Both "client-transport-security" and "peer-transport-security"
is mapped to Go struct embed.securityConfig.
Field for client certificate authentication is tagged as
"client-cert-auth", but it is misspelled in "peer-transport-security"
section.
Distribution would be:
0.1 second or more
...
25.6 seconds or more
51.2 seconds or more
etcd_network_snapshot_send_success
etcd_network_snapshot_send_failures
etcd_network_snapshot_send_total_duration_seconds
etcd_network_snapshot_receive_success
etcd_network_snapshot_receive_failures
etcd_network_snapshot_receive_total_duration_seconds
Signed-off-by: Gyuho Lee <leegyuho@amazon.com>
```
etcd_server_id{server_id="8e9e05c52164694d"} 1
```
Useful for automating membership change operations,
no need to run "endpoint status" or "member list"
command to get member IDs.
Also, useful for "etcd_network" metrics with "To/From" labels.
Signed-off-by: Gyuho Lee <leegyuho@amazon.com>
The benchmark as a sub command of put when provided will fetch hashkv and checks
the time taken to get it.
Fixed # 8910
Provider a way to etcd user to check how long it takes to get
hashkv in user enviornment. The command ouput will provide time taken
to get hashkv along with db file size.
etcd server
To improve debuggability of etcd v3. Added a grpc interceptor to log
info on incoming requests to etcd server. The log output includes
remote client info, request content (with value field redacted), request
handling latency, response size, etc. Uses zap logger if available,
otherwise uses capnslog.
Also did some clean up on the chaining of grpc interceptors on server
side.
We allow multiple in-flight append messages, but prior to this change
the only way we'd ever send them is if there is a steady stream of new
proposals. Catching up a follower that is far behind would be
unnecessarily slow (this is exacerbated by a quirk of CockroachDB's
use of raft which limits our ability to catch up via snapshot in some
cases).
See cockroachdb/cockroach#27983
The MaxSizePerMsg setting is now used to limit the size of
Ready.CommittedEntries. This prevents out-of-memory errors if the raft
log has become very large and commits all at once.
Since syscall.Errno implements net.Error and all cases
are matched sequentially, it's a mistake to put syscall.Errno
case after net.Error since it will never be executed.
This change swaps syscall.Errno case with net.Error
to give that clause chance to execute.
Joe has been helping fix lots of bugs and involved
in critical feature design/implementation,
plus leading patch release process for over a year.
Thanks for your excellent work.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
client should update next keepalive send time
even when lease keepalive response queue becomes full.
Otherwise, client sends keepalive request every 500ms
regardless of TTL when the send is only expected to happen
with the interval of TTL / 3 at minimum.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
gRPC has moved the transport package to an internal-only directory. This
eliminates direct use of the transport package in the stress test in
favor of the error code from gRPC that represents a connection problem.
https://godoc.org/google.golang.org/grpc/internal/transport is the new
location for the package, which says it's not intended to be imported
directly. Instead, the maintainers suggested to use the code Unavailable
to detect a connection problem.
This change slightly reorganizes the stresser test error handling.
The HTTP metrics have been deprecated and removed. In other places all
the HTTP related alerts have already been removed, this is just a clean
up of the remaining places these were left.
Using SRV discovery with TLS, the SRV record must be in the DNS SAN or clustering will fail.
This is a new requirement and may cause mysterious failures when upgrading from 3.1 to 3.2. I was only able to fix this in our configuration after reading through #8445; and now I understand the problem it's clear the docs have a hole here.
mirror_maker doc mentioned "Data under the destination cluster’s mirroring prefix should be treated as read only."
To avoid confusion, read only is being removed and a clear statement mentions that the updated values at destination wont be replicated from destination to origin.
wont -> won't
Fix
=== RUN TestEmbedEtcd
==================
WARNING: DATA RACE
Write at 0x000001df86d0 by goroutine 711:
github.com/coreos/etcd/embed.(*Config).setupLogging.func1()
/go/src/github.com/coreos/etcd/vendor/google.golang.org/grpc/grpclog/loggerv2.go:68 +0x16c
sync.(*Once).Do()
/usr/local/go/src/sync/once.go:44 +0xe1
github.com/coreos/etcd/embed.(*Config).setupLogging()
in gRPC proxy tests.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
"# github.com/coreos/etcd/raft
raft/logger.go:117: missing ... in args forwarded to print-like function"
New parameter check got added the golang to check the function parameter
c006036075 (diff-8fa5b0d6191706747ef5773f895781c9)
We only get 5 concurrent workers in Travis.
Save time by fetching test image only once
for all build tests.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Because the leases were sorted inside UnsafeLeases() the lessor mutex
ended up being locked while the whole map was sorted. This pulls the
soring outside of the lock, per feedback on
https://github.com/coreos/etcd/pull/9699
This also happens without gRPC proxy.
Fix panic when gRPC proxy leader watcher is restored:
```
go test -v -tags cluster_proxy -cpu 4 -race -run TestV3WatchRestoreSnapshotUnsync
=== RUN TestV3WatchRestoreSnapshotUnsync
panic: watcher minimum revision 9223372036854775805 should not exceed current revision 16
goroutine 156 [running]:
github.com/coreos/etcd/mvcc.(*watcherGroup).chooseAll(0xc4202b8720, 0x10, 0xffffffffffffffff, 0x1)
/home/gyuho/go/src/github.com/coreos/etcd/mvcc/watcher_group.go:242 +0x3b5
github.com/coreos/etcd/mvcc.(*watcherGroup).choose(0xc4202b8720, 0x200, 0x10, 0xffffffffffffffff, 0xc420253378, 0xc420253378)
/home/gyuho/go/src/github.com/coreos/etcd/mvcc/watcher_group.go:225 +0x289
github.com/coreos/etcd/mvcc.(*watchableStore).syncWatchers(0xc4202b86e0, 0x0)
/home/gyuho/go/src/github.com/coreos/etcd/mvcc/watchable_store.go:340 +0x237
github.com/coreos/etcd/mvcc.(*watchableStore).syncWatchersLoop(0xc4202b86e0)
/home/gyuho/go/src/github.com/coreos/etcd/mvcc/watchable_store.go:214 +0x280
created by github.com/coreos/etcd/mvcc.newWatchableStore
/home/gyuho/go/src/github.com/coreos/etcd/mvcc/watchable_store.go:90 +0x477
exit status 2
FAIL github.com/coreos/etcd/integration 2.551s
```
gRPC proxy spawns a watcher with a key "proxy-namespace__lostleader"
and watch revision "int64(math.MaxInt64 - 2)" to detect leader loss.
But, when the partitioned node restores, this watcher triggers
panic with "watcher minimum revision ... should not exceed current ...".
This check was added a long time ago, by my PR, when there was no gRPC proxy:
https://github.com/coreos/etcd/pull/4043#discussion_r48457145
> we can remove this checking actually. it is impossible for a unsynced watching to have a future rev. or we should just panic here.
However, now it's possible that a unsynced watcher has a future
revision, when it was moved from a synced watcher group through
restore operation.
This PR adds "restore" flag to indicate that a watcher was moved
from the synced watcher group with restore operation. Otherwise,
the watcher with future revision in an unsynced watcher group
would still panic.
Example logs with future revision watcher from restore operation:
```
{"level":"info","ts":1527196358.9057755,"caller":"mvcc/watcher_group.go:261","msg":"choosing future revision watcher from restore operation","watch-key":"proxy-namespace__lostleader","watch-revision":9223372036854775805,"current-revision":16}
{"level":"info","ts":1527196358.910349,"caller":"mvcc/watcher_group.go:261","msg":"choosing future revision watcher from restore operation","watch-key":"proxy-namespace__lostleader","watch-revision":9223372036854775805,"current-revision":16}
```
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
{"level":"warn","ts":1527101858.6985068,"caller":"etcdserver/util.go:115","msg":"apply request took too long","took":0.114101529,"expected-duration":0.1,"prefix":"","request":"header:<ID:1029181977902852337> put:<key:\"\\000\\000...
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
{"level":"warn","ts":1527101858.4149103,"caller":"etcdserver/raft.go:370","msg":"failed to send out heartbeat; took too long, server is overloaded likely from slow disk","heartbeat-interval":0.1,"expected-duration":0.2,"exceeded-duration":0.025771662}
{"level":"warn","ts":1527101858.4149644,"caller":"etcdserver/raft.go:370","msg":"failed to send out heartbeat; took too long, server is overloaded likely from slow disk","heartbeat-interval":0.1,"expected-duration":0.2,"exceeded-duration":0.034015766}
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
This commit adds two flags to etcdctl:
1. `--password` flag to etcdctl as a global option. It can be used for
specifying password for authentication required for the command
execution.
2. `--new-user-password` flag to `etcdctl user add`. It can be used
for specifying password of newly created user by the command.
The main motivation of the flags is allowing user to have : in its
name.
Fix https://github.com/coreos/etcd/issues/9691
systemd/journal should be configured manually.
Interpreting PPID 1 as systemd unit is wrong,
especially when embedded etcd runs under Docker
container. For such case, every single log write
will error and fall back to stderr, which is
inefficient.
Instead, add "systemd/journal" log output option
and returns error if systemd/journal socket cannot
be found.
"--logger=zap --log-outputs=default" will only error
for v3.4, since zap logger is still experimental.
v3.5 deprecates capnslog and changes "--log-outputs"
default value to "--log-outputs=stderr".
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
We replace/insert into in-memory B-tree, which means
we only keep a single node per key thus do not support
delete by revision on B-tree. So, (*keyIndex).tombstone
has always been marked with latest revision.
tombstone with key's modified revision panics:
panic: store.keyindex: put with unexpected smaller revision [{2 0} / {2 0}]
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
The `--listen-metrics-urls` also responds to the `/health` endpoint, but
that was not made clear in the documentation. These updates should help
explain how else that flag can be used.
When a new file is created under an non existent directly,
the v2 API automatically create the parent directly.
This commit aligns the behaviour of v2v3 emulation to comply with the v2
API.
"--log-output default,a.log" should error out when etcd
runs as a systemd unit. We expect 'default' to be overridden
explicitly.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
For compaction, clone the original Btree for traversal purposes, so as to
not hold the lock for the duration of compaction. This allows read/write
throughput by not blocking when the index tree is large (> 1M entries).
mvcc: add comment for index compaction lock
mvcc: explicitly unlock store to do index compaction synchronously
mvcc: formatting index bench
mvcc: add release note for index compaction changes
mvcc: add license header
- Add/Document "logger" to support structured logging.
- This makes functional tests run easier, since zap logger
provides built-in log redirect to files.
- "etcd --logger-option=zap" to enable structured logging.
- Current "capnslog" will still be used as "default".
- We may switch the default or deprecate "capnslog" in v3.5.
- Either way, will clearly be documented.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Balancer tests have been flaky since last balancer rewrite.
Should be from slow balancer failover. Once we re-rewrite balancer
with roundrobin, there should be no more flaky tests as below:
TestBalancerUnderServerStopInflightLinearizableGetOnRestart (27.62s)
server_shutdown_test.go:343: context deadline exceeded
TestBalancerUnderServerShutdownDelete (3.22s)
util.go:32: context deadline exceeded
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Since heap is already sorted, we can just check first element
to see if anything is expiry, rather than popping and pushing
it back. If nothing is expiry, pop operation is unnecessary.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
This adds a heap acting as a priority queue to keep track of lease
exiprations. Previously the whole lease map had to be iterated through
each time.
The queue allows us to check only those leases which might be expired.
When the expiration changes, we add an additional entry. If we check an
entry that isn't expired, it means that the lease got extended.
If we find a entry in the heap that doesn't have a corresponding entry in
the map, we know that the lease has already been expired or revoked.
Tests in `rafttest` would fail because they referred to field `Id` instead of
`ID`. This PR fixes that.
Closes#9504.
Signed-off-by: Kostas Christidis <kostas@christidis.io>
This confused me just looking at the protos; reading up on coreos/etcd#6518 helped clarify things. I highlighted the terms that show up as fields in protos and tried to clean up some of the language.
Current etcd code uses the string literals ("token", "authorization")
as field names of grpc and swappger metadata for passing token. It is
difficult to maintain so this commit introduces new constants for the
purpose.
Now Restart takes longer, and leasing Get blocks
on server restarts anyway, so fix CI failures in
slow machines.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Prometheus that scrapes etcd might also scrape other things, so we should
specifically ask for etcd metrics, especially when dealing with generic
metrics like `process_resident_memory_bytes`.
This commit adds a new option of --auth-token, ttl, for configuring
TTL of jwt tokens. It can be specified like this:
```
--auth-token jwt,pub-key=<pub key path>,priv-key=<priv key path>,sign-method=<sign method>,ttl=5m
```
In the above case, TTL will be 5 minutes.
This commit adds a new auth token provider named nop. The nop provider
refuses every Authenticate() request so CN based authentication can
only be allowed. If the tokenOpts parameter of auth.NewTokenProvider()
is empty, the provider will be used.
"ActiveSince" is read-only, and should be able to call
concurrently, as long as there is no routine holding
write lock (send, send snapshot).
e.g. "ActiveSince" is used around etcd server
"processInternalRaftRequestOnce", should be non-blocking
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Running check datascale can create large history of keyspaces and
internal fragmentation. Allow user to run auto compact and defrag
when run check datascale.
I noticed that the docs mention the grpc proxy will listen on port 8080,
but the above example is explicitly telling it to listen on 2379.
Removing confusion.
We enable "raft.Config.CheckQuorum" by default in other
Raft initial starts. So should start with "ForceNewCluster".
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
"stepCandidate" should reuse candidate's own term, not term in Message,
because pre-vote is requested with future term.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
`raft.Step` already ensures that when `m.Term > r.Term`,
candidate reverts back to follower with its term being
reset with `m.Term`, thus it's always true that
`m.Term == r.Term` in `stepCandidate`.
This just makes `r.becomeFollower` calls consistent.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Create a new command similar to check perf that can check the memory
consumption for putting different workloads on a given endpoint. If no endpoint
is provided, localhost will be used. Return user with a message that whether
there are enough memory for a given workload with pass or fail.
Fixed#9121
Due to boltdb behavior, dbSizeInUse can be unequal to dbSize even when
dbSize is growing in some conditions. The possible gap between the sizes
is up to a couple of pages.
Closes#9302
Fix govet warnings with Go tip
```
tools/etcd-test-proxy/main.go:47: Fprintln arg list ends with redundant
newline
```
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
In case syncWatchersLoop() starts before Restore() is called,
watchers already added by that moment are moved to s.synced by the loop.
However, there is a broken logic that moves watchers from s.synced
to s.uncyned without setting keyWatchers of the watcherGroup.
Eventually syncWatchers() fails to pickup those watchers from s.unsynced
and no events are sent to the watchers, because newWatcherBatch() called
in the function uses wg.watcherSetByKey() internally that requires
a proper keyWatchers value.
Membership reconfiguration may not be applied
when the new member joins. Also pass all endpoints
to check restore data in case of leader election or
network faults.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Existing dbSize shows physically allocated DB size and the backend
(boltdb) won't shrink it after a compaction until a user runs the defrag command.
The new dbSizeInUse shows the DB size that excludes free pages created
by compactions so that users can see the actual DB usage. dbSize >=
dbSizeInUse is always true.
Note that dbSizeInUse shows a page-based size and not byte level usage.
Reproducible builds are already done via Gopkg.lock.
Remove "constraint"s for external projects with conflicting
dependencies.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
This includes one theoretical logic change: A node that knows the
leader of the current term will no longer grant votes, even if it has
not yet voted in this term. It also adds a `m.Type == MsgPreVote`
guard on the `m.Term > r.Term` check, which was previously thought to
be incorrect (see #8517) but was actually just unclear.
Closes#8517Closes#8571
Large key writes (stressEntries[1].weight) should not take this
much weight. It was triggering "database size exceeded" errors.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Current etcdctl endpoint health --cluster asks password twice if auth
is enabled. This is because the command creates two client instances:
one for the purpose of checking endpoint health and another for
getting cluster members with MemberList(). The latter client doesn't
need to be authenticated because MemberList() is a public RPC. This
commit makes the client with no authed one.
Fix https://github.com/coreos/etcd/issues/9094
status.FromError can return nil, false. We are handling the return values
most places in code but some places we aren't. Fixing it herewith.
Fixes#9117
The context error with cancel code is typically for user cancellation which
should be at debug level. For other error codes we should display a warning.
Fixes#9085
As README declares in the beginning, flags can have dash and it should be
replaced with underscore while using. Also flags or commands are emphasized
in general. The range-end is not a flag or valid doc word and should not be
used as such.
bufio.NewReader.ReadString blocks even
when the process received syscall.SIGKILL.
Remove ptyMu mutex and make ReadString return
when *os.File is closed.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
This allows for watchers to be created concurrently
without needing potentially complex and latency-adding
queuing on the client.
Signed-off-by: Gyuho Lee <gyuhox@gmail.com>
Scanning the uncommitted portion of the raft log to determine whether
there are any pending config changes can be expensive. In
cockroachdb/cockroach#18601, we've seen that a new leader can spend so
much time scanning its log post-election that it fails to send
its first heartbeats in time to prevent a second election from
starting immediately.
Instead of tracking whether a pending config change exists with a
boolean, this commit tracks the latest log index at which a pending
config change *could* exist. This is a less expensive solution to
the problem, and the impact of false positives should be minimal since
a newly-elected leader should be able to quickly commit the tail of
its log.
Adding information that when config file is used other command line flags
and env variables will be ignored. This changes are a follow up of a
disucssion under PR,
https://github.com/coreos/etcd/pull/9066
A good bug report has some very specific qualities, so please read over our short document on [reporting bugs][report_bugs] before submitting a bug report.
See [code changes](https://github.com/coreos/etcd/compare/v3.0.15...v3.0.16) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Go
- Compile with [*Go 1.6.4*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.14...v3.0.15) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Fixed
- Fix cancel watch request with wrong range end.
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.13...v3.0.14) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Added
- v3 `etcdctl migrate` command now supports `--no-ttl` flag to discard keys on transform.
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.12...v3.0.13) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.11...v3.0.12) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.10...v3.0.11) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Added
- Server returns previous key-value (optional)
-`clientv3.WithPrevKV` option
- v3 etcdctl `put,watch,del --prev-kv` flag
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.9...v3.0.10) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.8...v3.0.9) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Added
- Warn on domain names on listen URLs (v3.2 will reject domain names).
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.7...v3.0.8) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Other
- Allow only IP addresses in listen URLs (domain names are rejected).
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.6...v3.0.7) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Other
- SRV records only allow A records (RFC 2052).
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.5...v3.0.6) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.4...v3.0.5) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Other
- SRV records (e.g., infra1.example.com) must match the discovery domain (i.e., example.com) if no custom certificate authority is given.
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.3...v3.0.4) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Added
- v2 `etcdctl ls` command now supports `--output=json`.
- Add /var/lib/etcd directory to etcd official Docker image.
### Other
- v2 auth can now use common name from TLS certificate when `--client-cert-auth` is enabled.
### Go
- Compile with [*Go 1.6.3*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.2...v3.0.3) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Other
- Revert Dockerfile to use `CMD`, instead of `ENTRYPOINT`, to support `etcdctl` run.
- Docker commands for v3.0.2 won't work without specifying executable binary paths.
- v3 etcdctl default endpoints are now `127.0.0.1:2379`.
### Go
- Compile with [*Go 1.6.2*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.1...v3.0.2) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Other
- Dockerfile uses `ENTRYPOINT`, instead of `CMD`, to run etcd without binary path specified.
### Go
- Compile with [*Go 1.6.2*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.0...v3.0.1) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Go
- Compile with [*Go 1.6.2*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v2.3.0...v3.0.0) and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_0.md).**
### Go
- Compile with [*Go 1.6.2*](https://golang.org/doc/devel/release.html#go1.6).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.18...v3.1.19) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Improved
- Improve [Raft Read Index timeout warning messages](https://github.com/coreos/etcd/pull/9897).
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-1) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Use it with `etcd_mvcc_db_total_size_in_bytes` and `etcd_mvcc_db_total_size_in_use_in_bytes`.
-`etcd_server_quota_backend_bytes 2.147483648e+09` means current quota size is 2 GB.
-`etcd_mvcc_db_total_size_in_bytes 20480` means current physically allocated DB size is 20 KB.
-`etcd_mvcc_db_total_size_in_use_in_bytes 16384` means future DB size if defragment operation is complete.
-`etcd_mvcc_db_total_size_in_bytes - etcd_mvcc_db_total_size_in_use_in_bytes` is the number of bytes that can be saved on disk with defragment operation.
- Use it with `etcd_mvcc_db_total_size_in_bytes` and `etcd_mvcc_db_total_size_in_use_in_bytes`.
-`etcd_server_quota_backend_bytes 2.147483648e+09` means current quota size is 2 GB.
-`etcd_mvcc_db_total_size_in_bytes 20480` means current physically allocated DB size is 20 KB.
-`etcd_mvcc_db_total_size_in_use_in_bytes 16384` means future DB size if defragment operation is complete.
-`etcd_mvcc_db_total_size_in_bytes - etcd_mvcc_db_total_size_in_use_in_bytes` is the number of bytes that can be saved on disk with defragment operation.
### client v3
- Fix [lease keepalive interval updates when response queue is full](https://github.com/coreos/etcd/pull/9952).
- If `<-chan *clientv3LeaseKeepAliveResponse` from `clientv3.Lease.KeepAlive` was never consumed or channel is full, client was [sending keepalive request every 500ms](https://github.com/coreos/etcd/issues/9911) instead of expected rate of every "TTL / 3" duration.
### Go
- Compile with [*Go 1.8.7*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.17...v3.1.18) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-1) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
See [code changes](https://github.com/coreos/etcd/compare/v3.1.16...v3.1.17) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
See [code changes](https://github.com/coreos/etcd/compare/v3.1.15...v3.1.16) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### etcd server
- Fix [`mvcc` server panic from restore operation](https://github.com/coreos/etcd/pull/9775).
- Let's assume that a watcher had been requested with a future revision X and sent to node A that became network-partitioned thereafter. Meanwhile, cluster makes progress. Then when the partition gets removed, the leader sends a snapshot to node A. Previously if the snapshot's latest revision is still lower than the watch revision X, **etcd server panicked** during snapshot restore operation.
- Now, this server-side panic has been fixed.
### Go
- Compile with [*Go 1.8.7*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.14...v3.1.15) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### etcd server
- Purge old [`*.snap.db` snapshot files](https://github.com/coreos/etcd/pull/7967).
- Previously, etcd did not respect `--max-snapshots` flag to purge old `*.snap.db` files.
- Now, etcd purges old `*.snap.db` files to keep maximum `--max-snapshots` number of files on disk.
### Go
- Compile with [*Go 1.8.7*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.13...v3.1.14) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-1) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Add [`--initial-election-tick-advance`](https://github.com/coreos/etcd/pull/9591) flag to configure initial election tick fast-forward.
- By default, `--initial-election-tick-advance=true`, then local member fast-forwards election ticks to speed up "initial" leader election trigger.
- This benefits the case of larger election ticks. For instance, cross datacenter deployment may require longer election timeout of 10-second. If true, local node does not need wait up to 10-second. Instead, forwards its election ticks to 8-second, and have only 2-second left before leader election.
- Major assumptions are that: cluster has no active leader thus advancing ticks enables faster leader election. Or cluster already has an established leader, and rejoining follower is likely to receive heartbeats from the leader after tick advance and before election timeout.
- However, when network from leader to rejoining follower is congested, and the follower does not receive leader heartbeat within left election ticks, disruptive election has to happen thus affecting cluster availabilities.
- Now, this can be disabled by setting `--initial-election-tick-advance=false`.
- Disabling this would slow down initial bootstrap process for cross datacenter deployments. Make tradeoffs by configuring `--initial-election-tick-advance` at the cost of slow initial bootstrap.
See [code changes](https://github.com/coreos/etcd/compare/v3.1.12...v3.1.13) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Improved
- Adjust [election timeout on server restart](https://github.com/coreos/etcd/pull/9415) to reduce [disruptive rejoining servers](https://github.com/coreos/etcd/issues/9333).
- Previously, etcd fast-forwards election ticks on server start, with only one tick left for leader election. This is to speed up start phase, without having to wait until all election ticks elapse. Advancing election ticks is useful for cross datacenter deployments with larger election timeouts. However, it was affecting cluster availability if the last tick elapses before leader contacts the restarted node.
- Now, when etcd restarts, it adjusts election ticks with more than one tick left, thus more time for leader to prevent disruptive restart.
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-1) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
See [code changes](https://github.com/coreos/etcd/compare/v3.1.11...v3.1.12) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
- "unsynced" watcher is watcher that needs to be in sync with events that have happened.
- That is, "unsynced" watcher is the slow watcher that was requested on old revision.
- "unsynced" watcher restore operation was not correctly populating its underlying watcher group.
- Which possibly causes [missing events from "unsynced" watchers](https://github.com/coreos/etcd/issues/9086).
- A node gets network partitioned with a watcher on a future revision, and falls behind receiving a leader snapshot after partition gets removed. When applying this snapshot, etcd watch storage moves current synced watchers to unsynced since sync watchers might have become stale during network partition. And reset synced watcher group to restart watcher routines. Previously, there was a bug when moving from synced watcher group to unsynced, thus client would miss events when the watcher was requested to the network-partitioned node.
### Go
- Compile with [*Go 1.8.7*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.10...v3.1.11) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### etcd server
- [#8411](https://github.com/coreos/etcd/issues/8411),[#8806](https://github.com/coreos/etcd/pull/8806) backport "mvcc: sending events after restore"
See [code changes](https://github.com/coreos/etcd/compare/v3.1.9...v3.1.10) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Added
- Tag docker images with minor versions.
- e.g. `docker pull quay.io/coreos/etcd:v3.1` to fetch latest v3.1 versions.
### Go
- Compile with [*Go 1.8.3*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.8...v3.1.9) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### etcd server
- Allow v2 snapshot over 512MB.
### Go
- Compile with [*Go 1.7.6*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.7...v3.1.8) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Go
- Compile with [*Go 1.7.5*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.6...v3.1.7) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Go
- Compile with [*Go 1.7.5*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.5...v3.1.6) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### etcd server
- Fill in Auth API response header.
- Remove auth check in Status API.
### Go
- Compile with [*Go 1.7.5*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.4...v3.1.5) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### etcd server
- Fix raft memory leak issue.
- Fix Windows file path issues.
### Other
- Add `/etc/nsswitch.conf` file to alpine-based Docker image.
### Go
- Compile with [*Go 1.7.5*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.3...v3.1.4) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Go
- Compile with [*Go 1.7.5*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.2...v3.1.3) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### etcd gateway
- Fix `etcd gateway` schema handling in DNS discovery.
- Fix sd_notify behaviors in `gateway`, `grpc-proxy`.
### gRPC Proxy
- Fix sd_notify behaviors in `gateway`, `grpc-proxy`.
### Other
- Use machine default host when advertise URLs are default values(`localhost:2379,2380`) AND if listen URL is `0.0.0.0`.
### Go
- Compile with [*Go 1.7.5*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.1...v3.1.2) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### etcd gateway
- Fix `etcd gateway` with multiple endpoints.
### Other
- Use IPv4 default host, by default (when IPv4 and IPv6 are available).
### Go
- Compile with [*Go 1.7.5*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.0...v3.1.1) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
### Go
- Compile with [*Go 1.7.5*](https://golang.org/doc/devel/release.html#go1.7).
See [code changes](https://github.com/coreos/etcd/compare/v3.0.0...v3.1.0) and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.1 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_1.md).**
- Deprecated following gRPC metrics in favor of [go-grpc-prometheus](https://github.com/grpc-ecosystem/go-grpc-prometheus).
-`etcd_grpc_requests_total`
-`etcd_grpc_requests_failed_total`
-`etcd_grpc_active_streams`
-`etcd_grpc_unary_requests_duration_seconds`
### Dependency
- Upgrade [`github.com/ugorji/go/codec`](https://github.com/ugorji/go) to [**`ugorji/go@9c7f9b7`**](https://github.com/ugorji/go/commit/9c7f9b7a2bc3a520f7c7b30b34b7f85f47fe27b6), and [regenerate v2 `client`](https://github.com/coreos/etcd/pull/6945).
### Security, Authentication
See [security doc](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
- SRV records (e.g., infra1.example.com) must match the discovery domain (i.e., example.com) if no custom certificate authority is given.
-`TLSConfig.ServerName` is ignored with user-provided certificates for backwards compatibility; to be deprecated.
- For example, `etcd --discovery-srv=example.com` will only authenticate peers/clients when the provided certs have root domain `example.com` as an entry in Subject Alternative Name (SAN) field.
### etcd server
- Automatic leadership transfer when leader steps down.
- etcd flags
-`--strict-reconfig-check` flag is set by default.
- Add `--log-output` flag.
- Add `--metrics` flag.
- etcd uses default route IP if advertise URL is not given.
- Cluster rejects removing members if quorum will be lost.
- Discovery now has upper limit for waiting on retries.
- Warn on binding listeners through domain names; to be deprecated.
- v3.0 and v3.1 with `--auto-compaction-retention=10` run periodic compaction on v3 key-value store for every 10-hour.
- Compactor only supports periodic compaction.
- Compactor records latest revisions every 5-minute, until it reaches the first compaction period (e.g. 10-hour).
- In order to retain key-value history of last compaction period, it uses the last revision that was fetched before compaction period, from the revision records that were collected every 5-minute.
- When `--auto-compaction-retention=10`, compactor uses revision 100 for compact revision where revision 100 is the latest revision fetched from 10 hours ago.
- If compaction succeeds or requested revision has already been compacted, it resets period timer and starts over with new historical revision records (e.g. restart revision collect and compact for the next 10-hour period).
- If compaction fails, it retries in 5 minutes.
### client v3
- Add `SetEndpoints` method; update endpoints at runtime.
- Add `Sync` method; auto-update endpoints at runtime.
See [code changes](https://github.com/coreos/etcd/compare/v3.2.23...v3.2.24) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Improved
- Improve [Raft Read Index timeout warning messages](https://github.com/coreos/etcd/pull/9897).
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-2) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Use it with `etcd_mvcc_db_total_size_in_bytes` and `etcd_mvcc_db_total_size_in_use_in_bytes`.
-`etcd_server_quota_backend_bytes 2.147483648e+09` means current quota size is 2 GB.
-`etcd_mvcc_db_total_size_in_bytes 20480` means current physically allocated DB size is 20 KB.
-`etcd_mvcc_db_total_size_in_use_in_bytes 16384` means future DB size if defragment operation is complete.
-`etcd_mvcc_db_total_size_in_bytes - etcd_mvcc_db_total_size_in_use_in_bytes` is the number of bytes that can be saved on disk with defragment operation.
- Use it with `etcd_mvcc_db_total_size_in_bytes` and `etcd_server_quota_backend_bytes`.
-`etcd_server_quota_backend_bytes 2.147483648e+09` means current quota size is 2 GB.
-`etcd_mvcc_db_total_size_in_bytes 20480` means current physically allocated DB size is 20 KB.
-`etcd_mvcc_db_total_size_in_use_in_bytes 16384` means future DB size if defragment operation is complete.
-`etcd_mvcc_db_total_size_in_bytes - etcd_mvcc_db_total_size_in_use_in_bytes` is the number of bytes that can be saved on disk with defragment operation.
### gRPC Proxy
- Add [flags for specifying TLS for connecting to proxy](https://github.com/coreos/etcd/pull/9894):
- Add [`grpc-proxy start --metrics-addr` flag for specifying a separate metrics listen address](https://github.com/coreos/etcd/pull/9894).
### client v3
- Fix [lease keepalive interval updates when response queue is full](https://github.com/coreos/etcd/pull/9952).
- If `<-chan *clientv3LeaseKeepAliveResponse` from `clientv3.Lease.KeepAlive` was never consumed or channel is full, client was [sending keepalive request every 500ms](https://github.com/coreos/etcd/issues/9911) instead of expected rate of every "TTL / 3" duration.
### Go
- Compile with [*Go 1.8.7*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.22...v3.2.23) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
See [code changes](https://github.com/coreos/etcd/compare/v3.2.21...v3.2.22) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Security, Authentication
- Support TLS cipher suite whitelisting.
- To block [weak cipher suites](https://github.com/coreos/etcd/issues/8320).
- TLS handshake fails when client hello is requested with invalid cipher suites.
See [code changes](https://github.com/coreos/etcd/compare/v3.2.20...v3.2.21) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcd server
- Fix [auth storage panic when simple token provider is disabled](https://github.com/coreos/etcd/pull/8695).
- Fix [`mvcc` server panic from restore operation](https://github.com/coreos/etcd/pull/9775).
- Let's assume that a watcher had been requested with a future revision X and sent to node A that became network-partitioned thereafter. Meanwhile, cluster makes progress. Then when the partition gets removed, the leader sends a snapshot to node A. Previously if the snapshot's latest revision is still lower than the watch revision X, **etcd server panicked** during snapshot restore operation.
- Now, this server-side panic has been fixed.
### Go
- Compile with [*Go 1.8.7*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.19...v3.2.20) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcd server
- Purge old [`*.snap.db` snapshot files](https://github.com/coreos/etcd/pull/7967).
- Previously, etcd did not respect `--max-snapshots` flag to purge old `*.snap.db` files.
- Now, etcd purges old `*.snap.db` files to keep maximum `--max-snapshots` number of files on disk.
### Go
- Compile with [*Go 1.8.7*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.18...v3.2.19) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-2) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Fix [TLS reload](https://github.com/coreos/etcd/pull/9570) when [certificate SAN field only includes IP addresses but no domain names](https://github.com/coreos/etcd/issues/9541).
- In Go, server calls `(*tls.Config).GetCertificate` for TLS reload if and only if server's `(*tls.Config).Certificates` field is not empty, or `(*tls.ClientHelloInfo).ServerName` is not empty with a valid SNI from the client. Previously, etcd always populates `(*tls.Config).Certificates` on the initial client TLS handshake, as non-empty. Thus, client was always expected to supply a matching SNI in order to pass the TLS verification and to trigger `(*tls.Config).GetCertificate` to reload TLS assets.
- However, a certificate whose SAN field does [not include any domain names but only IP addresses](https://github.com/coreos/etcd/issues/9541) would request `*tls.ClientHelloInfo` with an empty `ServerName` field, thus failing to trigger the TLS reload on initial TLS handshake; this becomes a problem when expired certificates need to be replaced online.
- Now, `(*tls.Config).Certificates` is created empty on initial TLS client handshake, first to trigger `(*tls.Config).GetCertificate`, and then to populate rest of the certificates on every new TLS connection, even when client SNI is empty (e.g. cert only includes IPs).
### etcd server
- Add [`etcd --initial-election-tick-advance`](https://github.com/coreos/etcd/pull/9591) flag to configure initial election tick fast-forward.
- By default, `etcd --initial-election-tick-advance=true`, then local member fast-forwards election ticks to speed up "initial" leader election trigger.
- This benefits the case of larger election ticks. For instance, cross datacenter deployment may require longer election timeout of 10-second. If true, local node does not need wait up to 10-second. Instead, forwards its election ticks to 8-second, and have only 2-second left before leader election.
- Major assumptions are that: cluster has no active leader thus advancing ticks enables faster leader election. Or cluster already has an established leader, and rejoining follower is likely to receive heartbeats from the leader after tick advance and before election timeout.
- However, when network from leader to rejoining follower is congested, and the follower does not receive leader heartbeat within left election ticks, disruptive election has to happen thus affecting cluster availabilities.
- Now, this can be disabled by setting `--initial-election-tick-advance=false`.
- Disabling this would slow down initial bootstrap process for cross datacenter deployments. Make tradeoffs by configuring `--initial-election-tick-advance` at the cost of slow initial bootstrap.
See [code changes](https://github.com/coreos/etcd/compare/v3.2.17...v3.2.18) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Improved
- Adjust [election timeout on server restart](https://github.com/coreos/etcd/pull/9415) to reduce [disruptive rejoining servers](https://github.com/coreos/etcd/issues/9333).
- Previously, etcd fast-forwards election ticks on server start, with only one tick left for leader election. This is to speed up start phase, without having to wait until all election ticks elapse. Advancing election ticks is useful for cross datacenter deployments with larger election timeouts. However, it was affecting cluster availability if the last tick elapses before leader contacts the restarted node.
- Now, when etcd restarts, it adjusts election ticks with more than one tick left, thus more time for leader to prevent disruptive restart.
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-2) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
See [code changes](https://github.com/coreos/etcd/compare/v3.2.16...v3.2.17) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcd server
- Fix [server panic on invalid Election Proclaim/Resign HTTP(S) requests](https://github.com/coreos/etcd/pull/9379).
- Previously, wrong-formatted HTTP requests to Election API could trigger panic in etcd server.
- e.g. `curl -L http://localhost:2379/v3/election/proclaim -X POST -d '{"value":""}'`, `curl -L http://localhost:2379/v3/election/resign -X POST -d '{"value":""}'`.
- Prevent [overflow by large `TTL` values for `Lease` `Grant`](https://github.com/coreos/etcd/pull/9399).
-`TTL` parameter to `Grant` request is unit of second.
- Leases with too large `TTL` values exceeding `math.MaxInt64` [expire in unexpected ways](https://github.com/coreos/etcd/issues/9374).
- Server now returns `rpctypes.ErrLeaseTTLTooLarge` to client, when the requested `TTL` is larger than *9,000,000,000 seconds* (which is >285 years).
- Again, etcd `Lease` is meant for short-periodic keepalives or sessions, in the range of seconds or minutes. Not for hours or days!
- Enable etcd server [`raft.Config.CheckQuorum` when starting with `ForceNewCluster`](https://github.com/coreos/etcd/pull/9347).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.15...v3.2.16) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
- "unsynced" watcher is watcher that needs to be in sync with events that have happened.
- That is, "unsynced" watcher is the slow watcher that was requested on old revision.
- "unsynced" watcher restore operation was not correctly populating its underlying watcher group.
- Which possibly causes [missing events from "unsynced" watchers](https://github.com/coreos/etcd/issues/9086).
- A node gets network partitioned with a watcher on a future revision, and falls behind receiving a leader snapshot after partition gets removed. When applying this snapshot, etcd watch storage moves current synced watchers to unsynced since sync watchers might have become stale during network partition. And reset synced watcher group to restart watcher routines. Previously, there was a bug when moving from synced watcher group to unsynced, thus client would miss events when the watcher was requested to the network-partitioned node.
### Go
- Compile with [*Go 1.8.5*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.14...v3.2.15) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcd server
- Prevent [server panic from member update/add](https://github.com/coreos/etcd/pull/9174) with [wrong scheme URLs](https://github.com/coreos/etcd/issues/9173).
- Log [user context cancel errors on stream APIs in debug level with TLS](https://github.com/coreos/etcd/pull/9178).
### Go
- Compile with [*Go 1.8.5*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.13...v3.2.14) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Improved
- Log [user context cancel errors on stream APIs in debug level](https://github.com/coreos/etcd/pull/9105).
### etcd server
- Fix [`mvcc/backend.defragdb` nil-pointer dereference on create bucket failure](https://github.com/coreos/etcd/pull/9119).
### Go
- Compile with [*Go 1.8.5*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.12...v3.2.13) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcd server
- Remove [verbose error messages on stream cancel and gRPC info-level logs](https://github.com/coreos/etcd/pull/9080) in server-side.
- Fix [gRPC server panic on `GracefulStop` TLS-enabled server](https://github.com/coreos/etcd/pull/8987).
### Go
- Compile with [*Go 1.8.5*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.11...v3.2.12) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Dependency
- Upgrade [`google.golang.org/grpc`](https://github.com/grpc/grpc-go/releases/tag) from [**`v1.7.4`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.4) to [**`v1.7.5`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.5).
- Upgrade [`github.com/grpc-ecosystem/grpc-gateway`](https://github.com/grpc-ecosystem/grpc-gateway/releases) from [**`v1.3`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.3) to [**`v1.3.0`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.3.0).
### etcd server
- Fix [error message of `Revision` compactor](https://github.com/coreos/etcd/pull/8999) in server-side.
### client v3
- Add [`MaxCallSendMsgSize` and `MaxCallRecvMsgSize`](https://github.com/coreos/etcd/pull/9047) fields to [`clientv3.Config`](https://godoc.org/github.com/coreos/etcd/clientv3#Config).
- Fix [exceeded response size limit error in client-side](https://github.com/coreos/etcd/issues/9043).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.10...v3.2.11) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Dependency
- Upgrade [`google.golang.org/grpc`](https://github.com/grpc/grpc-go/releases/tag) from [**`v1.7.3`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.3) to [**`v1.7.4`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.4).
### Security, Authentication
See [security doc](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
- Log [more details on TLS handshake failures](https://github.com/coreos/etcd/pull/8952/files).
### client v3
- Fix racey grpc-go's server handler transport `WriteStatus` call to prevent [TLS-enabled etcd server crash](https://github.com/coreos/etcd/issues/8904).
- Add [gRPC RPC failure warnings](https://github.com/coreos/etcd/pull/8939) to help debug such issues in the future.
### Documentation
- Remove `--listen-metrics-urls` flag in monitoring document (non-released in `v3.2.x`, planned for `v3.3.x`).
### Go
- Compile with [*Go 1.8.5*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.9...v3.2.10) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Dependency
- Upgrade [`google.golang.org/grpc`](https://github.com/grpc/grpc-go/releases/tag) from [**`v1.2.1`**](https://github.com/grpc/grpc-go/releases/tag/v1.2.1) to [**`v1.7.3`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.3).
- Upgrade [`github.com/grpc-ecosystem/grpc-gateway`](https://github.com/grpc-ecosystem/grpc-gateway/releases) from [**`v1.2.0`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.2.0) to [**`v1.3`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.3).
### Security, Authentication
See [security doc](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
- Revert [discovery SRV auth `ServerName` with `*.{ROOT_DOMAIN}`](https://github.com/coreos/etcd/pull/8651) to support non-wildcard subject alternative names in the certs (see [issue #8445](https://github.com/coreos/etcd/issues/8445) for more contexts).
- For instance, `etcd --discovery-srv=etcd.local` will only authenticate peers/clients when the provided certs have root domain `etcd.local` (**not `*.etcd.local`**) as an entry in Subject Alternative Name (SAN) field.
### etcd server
- Replace backend key-value database `boltdb/bolt` with [`coreos/bbolt`](https://github.com/coreos/bbolt/releases) to address [backend database size issue](https://github.com/coreos/etcd/issues/8009).
### client v3
- Rewrite balancer to handle [network partitions](https://github.com/coreos/etcd/issues/8711).
### Go
- Compile with [*Go 1.8.5*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.8...v3.2.9) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Security, Authentication
See [security doc](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
- Update `golang.org/x/crypto/bcrypt` (see [golang/crypto@6c586e1](https://github.com/golang/crypto/commit/6c586e17d90a7d08bbbc4069984180dce3b04117)).
- Fix discovery SRV bootstrapping to [authenticate `ServerName` with `*.{ROOT_DOMAIN}`](https://github.com/coreos/etcd/pull/8651), in order to support sub-domain wildcard matching (see [issue #8445](https://github.com/coreos/etcd/issues/8445) for more contexts).
- For instance, `etcd --discovery-srv=etcd.local` will only authenticate peers/clients when the provided certs have root domain `*.etcd.local` as an entry in Subject Alternative Name (SAN) field.
### Go
- Compile with [*Go 1.8.4*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.7...v3.2.8) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### client v2
- Fix v2 client failover to next endpoint on mutable operation.
See [code changes](https://github.com/coreos/etcd/compare/v3.2.6...v3.2.7) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Security, Authentication
- Fix [server-side auth so concurrent auth operations do not return old revision error](https://github.com/coreos/etcd/pull/8306).
### client v3
- Fix [`concurrency/stm` Put with serializable snapshot](https://github.com/coreos/etcd/pull/8439).
- Use store revision from first fetch to resolve write conflicts instead of modified revision.
### Go
- Compile with [*Go 1.8.3*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.5...v3.2.6) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcd server
- Fix watch restore from snapshot.
- Fix multiple URLs for `--listen-peer-urls` flag.
- Add `--enable-pprof` flag to etcd configuration file format.
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-2) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
See [code changes](https://github.com/coreos/etcd/compare/v3.2.4...v3.2.5) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcdctl v3
- Return non-zero exit code on unhealthy `endpoint health`.
### Security, Authentication
See [security doc](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
- [Server supports reverse-lookup on wildcard DNS `SAN`](https://github.com/coreos/etcd/pull/8281). For instance, if peer cert contains only DNS names (no IP addresses) in Subject Alternative Name (SAN) field, server first reverse-lookups the remote IP address to get a list of names mapping to that address (e.g. `nslookup IPADDR`). Then accepts the connection if those names have a matching name with peer cert's DNS names (either by exact or wildcard match). If none is matched, server forward-lookups each DNS entry in peer cert (e.g. look up `example.default.svc` when the entry is `*.example.default.svc`), and accepts connection only when the host's resolved addresses have the matching IP address with the peer's remote IP address. For example, peer B's CSR (with `cfssl`) SAN field is `["*.example.default.svc", "*.example.default.svc.cluster.local"]` when peer B's remote IP address is `10.138.0.2`. When peer B tries to join the cluster, peer A reverse-lookup the IP `10.138.0.2` to get the list of host names. And either exact or wildcard match the host names with peer B's cert DNS names in Subject Alternative Name (SAN) field. If none of reverse/forward lookups worked, it returns an error `"tls: "10.138.0.2" does not match any of DNSNames ["*.example.default.svc","*.example.default.svc.cluster.local"]`. See [issue#8268](https://github.com/coreos/etcd/issues/8268) for more detail.
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-2) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Fix unreachable `/metrics` endpoint when `--enable-v2=false`.
See [code changes](https://github.com/coreos/etcd/compare/v3.2.3...v3.2.4) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcd server
- Do not block on active client stream when stopping server
### gRPC proxy
- Fix gRPC proxy Snapshot RPC error handling
### Go
- Compile with [*Go 1.8.3*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.2...v3.2.3) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### client v3
- Let clients establish unlimited streams
### Other
- Tag docker images with minor versions
- e.g. `docker pull quay.io/coreos/etcd:v3.2` to fetch latest v3.2 versions
### Go
- Compile with [*Go 1.8.3*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.1...v3.2.2) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Improved
- Rate-limit lease revoke on expiration.
- Extend leases on promote to avoid queueing effect on lease expiration.
### Security, Authentication
See [security doc](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
- [Server accepts connections if IP matches, without checking DNS entries](https://github.com/coreos/etcd/pull/8223). For instance, if peer cert contains IP addresses and DNS names in Subject Alternative Name (SAN) field, and the remote IP address matches one of those IP addresses, server just accepts connection without further checking the DNS names. For example, peer B's CSR (with `cfssl`) SAN field is `["invalid.domain", "10.138.0.2"]` when peer B's remote IP address is `10.138.0.2` and `invalid.domain` is a invalid host. When peer B tries to join the cluster, peer A successfully authenticates B, since Subject Alternative Name (SAN) field has a valid matching IP address. See [issue#8206](https://github.com/coreos/etcd/issues/8206) for more detail.
### etcd server
- Accept connection with matched IP SAN but no DNS match.
- Don't check DNS entries in certs if there's a matching IP.
### gRPC gateway
- Use user-provided listen address to connect to gRPC gateway.
See [code changes](https://github.com/coreos/etcd/compare/v3.2.0...v3.2.1) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### etcd server
- Fix backend database in-memory index corruption issue on restore (only 3.2.0 is affected).
### gRPC gateway
- Fix Txn marshaling.
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-2) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Fix backend database size debugging metrics.
### Go
- Compile with [*Go 1.8.3*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.1.0...v3.2.0) and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.2 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_2.md).**
### Improved
- Improve backend read concurrency.
### Breaking Changes
- Increased [`--snapshot-count` default value from 10,000 to 100,000](https://github.com/coreos/etcd/pull/7160).
- Higher snapshot count means it holds Raft entries in memory for longer before discarding old entries.
- It is a trade-off between less frequent snapshotting and [higher memory usage](https://github.com/kubernetes/kubernetes/issues/60589#issuecomment-371977156).
- User lower `--snapshot-count` value for lower memory usage.
- User higher `--snapshot-count` value for better availabilities of slow followers (less frequent snapshots from leader).
-`clientv3.Lease.TimeToLive` returns `LeaseTimeToLiveResponse.TTL == -1` on lease not found.
-`clientv3.NewFromConfigFile` is moved to `clientv3/yaml.NewConfig`.
-`embed.Etcd.Peers` field is now `[]*peerListener`.
- Rejects domains names for `--listen-peer-urls` and `--listen-client-urls` (3.1 only prints out warnings), since [domain name is invalid for network interface binding](https://github.com/coreos/etcd/issues/6336).
### Dependency
- Upgrade [`google.golang.org/grpc`](https://github.com/grpc/grpc-go/releases) from [**`v1.0.4`**](https://github.com/grpc/grpc-go/releases/tag/v1.0.4) to [**`v1.2.1`**](https://github.com/grpc/grpc-go/releases/tag/v1.2.1).
- Upgrade [`github.com/grpc-ecosystem/grpc-gateway`](https://github.com/grpc-ecosystem/grpc-gateway/releases) to [**`v1.2.0`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.2.0).
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-2) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
See [security doc](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
- [TLS certificates get reloaded on every client connection](https://github.com/coreos/etcd/pull/7829). This is useful when replacing expiry certs without stopping etcd servers; it can be done by overwriting old certs with new ones. Refreshing certs for every connection should not have too much overhead, but can be improved in the future, with caching layer. Example tests can be found [here](https://github.com/coreos/etcd/blob/b041ce5d514a4b4aaeefbffb008f0c7570a18986/integration/v3_grpc_test.go#L1601-L1757).
- [Server denies incoming peer certs with wrong IP `SAN`](https://github.com/coreos/etcd/pull/7687). For instance, if peer cert contains any IP addresses in Subject Alternative Name (SAN) field, server authenticates a peer only when the remote IP address matches one of those IP addresses. This is to prevent unauthorized endpoints from joining the cluster. For example, peer B's CSR (with `cfssl`) SAN field is `["*.example.default.svc", "*.example.default.svc.cluster.local", "10.138.0.27"]` when peer B's actual IP address is `10.138.0.2`, not `10.138.0.27`. When peer B tries to join the cluster, peer A will reject B with the error `x509: certificate is valid for 10.138.0.27, not 10.138.0.2`, because B's remote IP address does not match the one in Subject Alternative Name (SAN) field.
- [Server resolves TLS `DNSNames` when checking `SAN`](https://github.com/coreos/etcd/pull/7767). For instance, if peer cert contains only DNS names (no IP addresses) in Subject Alternative Name (SAN) field, server authenticates a peer only when forward-lookups (`dig b.com`) on those DNS names have matching IP with the remote IP address. For example, peer B's CSR (with `cfssl`) SAN field is `["b.com"]` when peer B's remote IP address is `10.138.0.2`. When peer B tries to join the cluster, peer A looks up the incoming host `b.com` to get the list of IP addresses (e.g. `dig b.com`). And rejects B if the list does not contain the IP `10.138.0.2`, with the error `tls: 10.138.0.2 does not match any of DNSNames ["b.com"]`.
- Auth support JWT token.
### etcd server
- RPCs
- Add Election, Lock service.
- Native client `etcdserver/api/v3client`
- client "embedded" in the server.
- Logging, monitoring
- Server warns large snapshot operations.
- Add `etcd --enable-v2` flag to enable v2 API server.
- Compactor continues to record latest revisions every 5-minute.
- For every hour, it uses the last revision that was fetched before compaction period, from the revision records that were collected every 5-minute.
- That is, for every hour, compactor discards historical data created before compaction period.
- The retention window of compaction period moves to next hour.
- For instance, when hourly writes are 100 and `--auto-compaction-retention=10`, v3.1 compacts revision 1000, 2000, and 3000 for every 10-hour, while v3.2 compacts revision 1000, 1100, and 1200 for every 1-hour.
- If compaction succeeds or requested revision has already been compacted, it resets period timer and removes used compacted revision from historical revision records (e.g. start next revision collect and compaction from previously collected revisions).
- If compaction fails, it retries in 5 minutes.
- Allow snapshot over 512MB.
### client v3
- STM prefetching.
- Add namespace feature.
- Add `ErrOldCluster` with server version checking.
- Translate `WithPrefix()` into `WithFromKey()` for empty key.
### etcdctl v3
- Add `check perf` command.
- Add `etcdctl --from-key` flag to role grant-permission command.
-`lock` command takes an optional command to execute.
### gRPC Proxy
- Proxy endpoint discovery.
- Namespaces.
- Coalesce lease requests.
### etcd gateway
- Support [DNS SRV priority](https://github.com/coreos/etcd/pull/7882) for [smart proxy routing](https://github.com/coreos/etcd/issues/4378).
### Other
- v3 client
- concurrency package's elections updated to match RPC interfaces.
- let client dial endpoints not in the balancer.
- Release
- Annotate acbuild with supports-systemd-notify.
- Add `nsswitch.conf` to Docker container image.
- Add ppc64le, arm64(experimental) builds.
### Go
- Compile with [*Go 1.8.3*](https://golang.org/doc/devel/release.html#go1.8).
See [code changes](https://github.com/coreos/etcd/compare/v3.3.8...v3.3.9) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
### Improved
- Improve [Raft Read Index timeout warning messages](https://github.com/coreos/etcd/pull/9897).
### Security, Authentication
- Compile with [*Go 1.10.3*](https://golang.org/doc/devel/release.html#go1.10) to support [crypto/x509 "Name Constraints"](https://github.com/coreos/etcd/issues/9912).
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-3) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Use it with `etcd_mvcc_db_total_size_in_bytes` and `etcd_mvcc_db_total_size_in_use_in_bytes`.
-`etcd_server_quota_backend_bytes 2.147483648e+09` means current quota size is 2 GB.
-`etcd_mvcc_db_total_size_in_bytes 20480` means current physically allocated DB size is 20 KB.
-`etcd_mvcc_db_total_size_in_use_in_bytes 16384` means future DB size if defragment operation is complete.
-`etcd_mvcc_db_total_size_in_bytes - etcd_mvcc_db_total_size_in_use_in_bytes` is the number of bytes that can be saved on disk with defragment operation.
- Use it with `etcd_mvcc_db_total_size_in_bytes` and `etcd_mvcc_db_total_size_in_use_in_bytes`.
-`etcd_server_quota_backend_bytes 2.147483648e+09` means current quota size is 2 GB.
-`etcd_mvcc_db_total_size_in_bytes 20480` means current physically allocated DB size is 20 KB.
-`etcd_mvcc_db_total_size_in_use_in_bytes 16384` means future DB size if defragment operation is complete.
-`etcd_mvcc_db_total_size_in_bytes - etcd_mvcc_db_total_size_in_use_in_bytes` is the number of bytes that can be saved on disk with defragment operation.
### client v3
- Fix [lease keepalive interval updates when response queue is full](https://github.com/coreos/etcd/pull/9952).
- If `<-chan *clientv3LeaseKeepAliveResponse` from `clientv3.Lease.KeepAlive` was never consumed or channel is full, client was [sending keepalive request every 500ms](https://github.com/coreos/etcd/issues/9911) instead of expected rate of every "TTL / 3" duration.
### Go
- Compile with [*Go 1.10.3*](https://golang.org/doc/devel/release.html#go1.10).
See [code changes](https://github.com/coreos/etcd/compare/v3.3.7...v3.3.8) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
See [code changes](https://github.com/coreos/etcd/compare/v3.3.6...v3.3.7) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
### Security, Authentication
- Support TLS cipher suite whitelisting.
- To block [weak cipher suites](https://github.com/coreos/etcd/issues/8320).
- TLS handshake fails when client hello is requested with invalid cipher suites.
See [code changes](https://github.com/coreos/etcd/compare/v3.3.5...v3.3.6) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
- Previously, when auth token is an empty string, it returns [`failed to initialize the etcd server: auth: invalid auth options` error](https://github.com/coreos/etcd/issues/9349).
- Fix [auth storage panic on server lease revoke routine with JWT token](https://github.com/coreos/etcd/issues/9695).
- Fix [`mvcc` server panic from restore operation](https://github.com/coreos/etcd/pull/9775).
- Let's assume that a watcher had been requested with a future revision X and sent to node A that became network-partitioned thereafter. Meanwhile, cluster makes progress. Then when the partition gets removed, the leader sends a snapshot to node A. Previously if the snapshot's latest revision is still lower than the watch revision X, **etcd server panicked** during snapshot restore operation.
- Now, this server-side panic has been fixed.
### Go
- Compile with [*Go 1.9.6*](https://golang.org/doc/devel/release.html#go1.9).
See [code changes](https://github.com/coreos/etcd/compare/v3.3.4...v3.3.5) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
See [code changes](https://github.com/coreos/etcd/compare/v3.3.3...v3.3.4) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-3) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Fix [race conditions in v2 server stat collecting](https://github.com/coreos/etcd/pull/9562).
### Security, Authentication
- Fix [TLS reload](https://github.com/coreos/etcd/pull/9570) when [certificate SAN field only includes IP addresses but no domain names](https://github.com/coreos/etcd/issues/9541).
- In Go, server calls `(*tls.Config).GetCertificate` for TLS reload if and only if server's `(*tls.Config).Certificates` field is not empty, or `(*tls.ClientHelloInfo).ServerName` is not empty with a valid SNI from the client. Previously, etcd always populates `(*tls.Config).Certificates` on the initial client TLS handshake, as non-empty. Thus, client was always expected to supply a matching SNI in order to pass the TLS verification and to trigger `(*tls.Config).GetCertificate` to reload TLS assets.
- However, a certificate whose SAN field does [not include any domain names but only IP addresses](https://github.com/coreos/etcd/issues/9541) would request `*tls.ClientHelloInfo` with an empty `ServerName` field, thus failing to trigger the TLS reload on initial TLS handshake; this becomes a problem when expired certificates need to be replaced online.
- Now, `(*tls.Config).Certificates` is created empty on initial TLS client handshake, first to trigger `(*tls.Config).GetCertificate`, and then to populate rest of the certificates on every new TLS connection, even when client SNI is empty (e.g. cert only includes IPs).
### etcd server
- Add [`etcd --initial-election-tick-advance`](https://github.com/coreos/etcd/pull/9591) flag to configure initial election tick fast-forward.
- By default, `etcd --initial-election-tick-advance=true`, then local member fast-forwards election ticks to speed up "initial" leader election trigger.
- This benefits the case of larger election ticks. For instance, cross datacenter deployment may require longer election timeout of 10-second. If true, local node does not need wait up to 10-second. Instead, forwards its election ticks to 8-second, and have only 2-second left before leader election.
- Major assumptions are that: cluster has no active leader thus advancing ticks enables faster leader election. Or cluster already has an established leader, and rejoining follower is likely to receive heartbeats from the leader after tick advance and before election timeout.
- However, when network from leader to rejoining follower is congested, and the follower does not receive leader heartbeat within left election ticks, disruptive election has to happen thus affecting cluster availabilities.
- Now, this can be disabled by setting `--initial-election-tick-advance=false`.
- Disabling this would slow down initial bootstrap process for cross datacenter deployments. Make tradeoffs by configuring `etcd --initial-election-tick-advance` at the cost of slow initial bootstrap.
See [code changes](https://github.com/coreos/etcd/compare/v3.3.2...v3.3.3) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
### Improved
- Adjust [election timeout on server restart](https://github.com/coreos/etcd/pull/9415) to reduce [disruptive rejoining servers](https://github.com/coreos/etcd/issues/9333).
- Previously, etcd fast-forwards election ticks on server start, with only one tick left for leader election. This is to speed up start phase, without having to wait until all election ticks elapse. Advancing election ticks is useful for cross datacenter deployments with larger election timeouts. However, it was affecting cluster availability if the last tick elapses before leader contacts the restarted node.
- Now, when etcd restarts, it adjusts election ticks with more than one tick left, thus more time for leader to prevent disruptive restart.
- e.g. `etcd --auto-compaction-mode=revision --auto-compaction-retention=1000` automatically `Compact` on `"latest revision" - 1000` every 5-minute (when latest revision is 30000, compact on revision 29000).
- e.g. Previously, `etcd --auto-compaction-mode=periodic --auto-compaction-retention=72h` automatically `Compact` with 72-hour retention windown for every 7.2-hour. **Now, `Compact` happens, for every 1-hour but still with 72-hour retention window.**
- e.g. Previously, `etcd --auto-compaction-mode=periodic --auto-compaction-retention=30m` automatically `Compact` with 30-minute retention windown for every 3-minute. **Now, `Compact` happens, for every 30-minute but still with 30-minute retention window.**
- Periodic compactor keeps recording latest revisions for every compaction period when given period is less than 1-hour, or for every 1-hour when given compaction period is greater than 1-hour (e.g. 1-hour when `etcd --auto-compaction-mode=periodic --auto-compaction-retention=24h`).
- For every compaction period or 1-hour, compactor uses the last revision that was fetched before compaction period, to discard historical data.
- The retention window of compaction period moves for every given compaction period or hour.
- For instance, when hourly writes are 100 and `etcd --auto-compaction-mode=periodic --auto-compaction-retention=24h`, `v3.2.x`, `v3.3.0`, `v3.3.1`, and `v3.3.2` compact revision 2400, 2640, and 2880 for every 2.4-hour, while `v3.3.3`*or later* compacts revision 2400, 2500, 2600 for every 1-hour.
- Futhermore, when `etcd --auto-compaction-mode=periodic --auto-compaction-retention=30m` and writes per minute are about 1000, `v3.3.0`, `v3.3.1`, and `v3.3.2` compact revision 30000, 33000, and 36000, for every 3-minute, while `v3.3.3`*or later* compacts revision 30000, 60000, and 90000, for every 30-minute.
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-3) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
See [code changes](https://github.com/coreos/etcd/compare/v3.3.1...v3.3.2) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
### etcd server
- Fix [server panic on invalid Election Proclaim/Resign HTTP(S) requests](https://github.com/coreos/etcd/pull/9379).
- Previously, wrong-formatted HTTP requests to Election API could trigger panic in etcd server.
- e.g. `curl -L http://localhost:2379/v3/election/proclaim -X POST -d '{"value":""}'`, `curl -L http://localhost:2379/v3/election/resign -X POST -d '{"value":""}'`.
See [code changes](https://github.com/coreos/etcd/compare/v3.3.0...v3.3.1) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
### Improved
- Add [warnings on requests taking too long](https://github.com/coreos/etcd/pull/9288).
- e.g. `etcdserver: read-only range request "key:\"\\000\" range_end:\"\\000\" " took too long [3.389041388s] to execute`
- "unsynced" watcher is watcher that needs to be in sync with events that have happened.
- That is, "unsynced" watcher is the slow watcher that was requested on old revision.
- "unsynced" watcher restore operation was not correctly populating its underlying watcher group.
- Which possibly causes [missing events from "unsynced" watchers](https://github.com/coreos/etcd/issues/9086).
- A node gets network partitioned with a watcher on a future revision, and falls behind receiving a leader snapshot after partition gets removed. When applying this snapshot, etcd watch storage moves current synced watchers to unsynced since sync watchers might have become stale during network partition. And reset synced watcher group to restart watcher routines. Previously, there was a bug when moving from synced watcher group to unsynced, thus client would miss events when the watcher was requested to the network-partitioned node.
### Go
- Compile with [*Go 1.9.4*](https://golang.org/doc/devel/release.html#go1.9).
See [code changes](https://github.com/coreos/etcd/compare/v3.2.0...v3.3.0) and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.3 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_3.md).**
- [v3.3.0-rc.4](https://github.com/coreos/etcd/releases/tag/v3.3.0-rc.4) (2018-01-22), see [code changes](https://github.com/coreos/etcd/compare/v3.3.0-rc.3...v3.3.0-rc.4).
- [v3.3.0-rc.3](https://github.com/coreos/etcd/releases/tag/v3.3.0-rc.3) (2018-01-17), see [code changes](https://github.com/coreos/etcd/compare/v3.3.0-rc.2...v3.3.0-rc.3).
- [v3.3.0-rc.2](https://github.com/coreos/etcd/releases/tag/v3.3.0-rc.2) (2018-01-11), see [code changes](https://github.com/coreos/etcd/compare/v3.3.0-rc.1...v3.3.0-rc.2).
- [v3.3.0-rc.1](https://github.com/coreos/etcd/releases/tag/v3.3.0-rc.1) (2018-01-02), see [code changes](https://github.com/coreos/etcd/compare/v3.3.0-rc.0...v3.3.0-rc.1).
- [v3.3.0-rc.0](https://github.com/coreos/etcd/releases/tag/v3.3.0-rc.0) (2017-12-20), see [code changes](https://github.com/coreos/etcd/compare/v3.2.0...v3.3.0-rc.0).
### Improved
- Use [`coreos/bbolt`](https://github.com/coreos/bbolt/releases) to replace [`boltdb/bolt`](https://github.com/boltdb/bolt#project-status).
- Fix [etcd database size grows until `mvcc: database space exceeded`](https://github.com/coreos/etcd/issues/8009).
- [Support database size larger than 8GiB](https://github.com/coreos/etcd/pull/7525) (8GiB is now a suggested maximum size for normal environments)
- [Reduce memory allocation](https://github.com/coreos/etcd/pull/8428) on [Range operations](https://github.com/coreos/etcd/pull/8475).
- [Rate limit](https://github.com/coreos/etcd/pull/8099) and [randomize](https://github.com/coreos/etcd/pull/8101) lease revoke on restart or leader elections.
- Prevent [spikes in Raft proposal rate](https://github.com/coreos/etcd/issues/8096).
- Support `clientv3` balancer failover under [network faults/partitions](https://github.com/coreos/etcd/issues/8711).
- Better warning on [mismatched `etcd --initial-cluster`](https://github.com/coreos/etcd/pull/8083) flag.
- etcd compares `etcd --initial-advertise-peer-urls` against corresponding `etcd --initial-cluster` URLs with forward-lookup.
- If resolved IP addresses of `etcd --initial-advertise-peer-urls` and `etcd --initial-cluster` do not match (e.g. [due to DNS error](https://github.com/coreos/etcd/pull/9210)), etcd will exit with errors.
- v3.2 error: `etcd --initial-cluster must include s1=https://s1.test:2380 given --initial-advertise-peer-urls=https://s1.test:2380`.
- v3.3 error: `failed to resolve https://s1.test:2380 to match --initial-cluster=s1=https://s1.test:2380 (failed to resolve "https://s1.test:2380" (error ...))`.
### Breaking Changes
- Require [`google.golang.org/grpc`](https://github.com/grpc/grpc-go/releases) [**`v1.7.4`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.4) or [**`v1.7.5`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.5).
- Deprecate `grpclog.Logger`, upgrade to [`grpclog.LoggerV2`](https://github.com/coreos/etcd/pull/8533).
- Deprecate [`grpc.ErrClientConnTimeout`](https://github.com/coreos/etcd/pull/8505) errors in `clientv3`.
- Use [`MaxRecvMsgSize` and `MaxSendMsgSize`](https://github.com/coreos/etcd/pull/8437) to limit message size, in etcd server.
- Translate [gRPC status error in v3 client `Snapshot` API](https://github.com/coreos/etcd/pull/9038).
- v3 `etcdctl` [`lease timetolive LEASE_ID`](https://github.com/coreos/etcd/issues/9028) on expired lease now prints [`"lease LEASE_ID already expired"`](https://github.com/coreos/etcd/pull/9047).
-<=3.2 prints `"lease LEASE_ID granted with TTL(0s), remaining(-1s)"`.
- Replace [gRPC gateway](https://github.com/grpc-ecosystem/grpc-gateway) endpoint `/v3alpha` with [`/v3beta`](https://github.com/coreos/etcd/pull/8880).
- To deprecate [`/v3alpha`](https://github.com/coreos/etcd/issues/8125) in v3.4.
- In v3.3, `curl -L http://localhost:2379/v3alpha/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` still works as a fallback to `curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'`, but `curl -L http://localhost:2379/v3alpha/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` won't work in v3.4. Use `curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` instead.
- Change `etcd --auto-compaction-retention` flag to [accept string values](https://github.com/coreos/etcd/pull/8563) with [finer granularity](https://github.com/coreos/etcd/issues/8503).
- Now that `etcd --auto-compaction-retention` accepts string values, etcd configuration YAML file `auto-compaction-retention` field must be changed to `string` type.
- Previously, `--config-file etcd.config.yaml` can have `auto-compaction-retention: 24` field, now must be `auto-compaction-retention: "24"` or `auto-compaction-retention: "24h"`.
- If configured as `etcd --auto-compaction-mode periodic --auto-compaction-retention "24h"`, the time duration value for `etcd --auto-compaction-retention` flag must be valid for [`time.ParseDuration`](https://golang.org/pkg/time/#ParseDuration) function in Go.
### Dependency
- Upgrade [`boltdb/bolt`](https://github.com/boltdb/bolt#project-status) from [**`v1.3.0`**](https://github.com/boltdb/bolt/releases/tag/v1.3.0) to [`coreos/bbolt`](https://github.com/coreos/bbolt/releases) [**`v1.3.1-coreos.6`**](https://github.com/coreos/bbolt/releases/tag/v1.3.1-coreos.6).
- Upgrade [`google.golang.org/grpc`](https://github.com/grpc/grpc-go/releases) from [**`v1.2.1`**](https://github.com/grpc/grpc-go/releases/tag/v1.2.1) to [**`v1.7.5`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.5).
- Upgrade [`github.com/ugorji/go/codec`](https://github.com/ugorji/go) to [**`v1.1`**](https://github.com/ugorji/go/releases/tag/v1.1), and [regenerate v2 `client`](https://github.com/coreos/etcd/pull/8721).
- Upgrade [`github.com/ugorji/go/codec`](https://github.com/ugorji/go) to [**`ugorji/go@54210f4e0`**](https://github.com/ugorji/go/commit/54210f4e076c57f351166f0ed60e67d3fca57a36), and [regenerate v2 `client`](https://github.com/coreos/etcd/pull/8574).
- Upgrade [`github.com/grpc-ecosystem/grpc-gateway`](https://github.com/grpc-ecosystem/grpc-gateway/releases) from [**`v1.2.2`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.2.2) to [**`v1.3.0`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.3.0).
- Upgrade [`golang.org/x/crypto/bcrypt`](https://github.com/golang/crypto) to [**`golang/crypto@6c586e17d`**](https://github.com/golang/crypto/commit/6c586e17d90a7d08bbbc4069984180dce3b04117).
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#v3-3) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Add [`etcd --listen-metrics-urls`](https://github.com/coreos/etcd/pull/8242) flag for additional `/metrics` and `/health` endpoints.
- Useful for [bypassing critical APIs when monitoring etcd](https://github.com/coreos/etcd/issues/8060).
- Initialize gRPC server [metrics with zero values](https://github.com/coreos/etcd/pull/8878).
- Fix [range/put/delete operation metrics](https://github.com/coreos/etcd/pull/8054) with transaction.
-`etcd_debugging_mvcc_range_total`
-`etcd_debugging_mvcc_put_total`
-`etcd_debugging_mvcc_delete_total`
-`etcd_debugging_mvcc_txn_total`
- Fix [`etcd_debugging_mvcc_keys_total`](https://github.com/coreos/etcd/pull/8390) on restore.
- Fix [`etcd_debugging_mvcc_db_total_size_in_bytes`](https://github.com/coreos/etcd/pull/8120) on restore.
- Also change to [`prometheus.NewGaugeFunc`](https://github.com/coreos/etcd/pull/8150).
### Security, Authentication
See [security doc](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
- Add [CRL based connection rejection](https://github.com/coreos/etcd/pull/8124) to manage [revoked certs](https://github.com/coreos/etcd/issues/4034).
- [Server accepts connections if IP matches, without checking DNS entries](https://github.com/coreos/etcd/pull/8223). For instance, if peer cert contains IP addresses and DNS names in Subject Alternative Name (SAN) field, and the remote IP address matches one of those IP addresses, server just accepts connection without further checking the DNS names.
- [Server supports reverse-lookup on wildcard DNS `SAN`](https://github.com/coreos/etcd/pull/8281). For instance, if peer cert contains only DNS names (no IP addresses) in Subject Alternative Name (SAN) field, server first reverse-lookups the remote IP address to get a list of names mapping to that address (e.g. `nslookup IPADDR`). Then accepts the connection if those names have a matching name with peer cert's DNS names (either by exact or wildcard match). If none is matched, server forward-lookups each DNS entry in peer cert (e.g. look up `example.default.svc` when the entry is `*.example.default.svc`), and accepts connection only when the host's resolved addresses have the matching IP address with the peer's remote IP address.
- To support [CommonName(CN) based auth](https://github.com/coreos/etcd/issues/8262) for inter peer connection.
- [Swap priority](https://github.com/coreos/etcd/pull/8594) of cert CommonName(CN) and username + password.
- To address ["username and password specified in the request should take priority over CN in the cert"](https://github.com/coreos/etcd/issues/8584).
- Protect [lease revoke with auth](https://github.com/coreos/etcd/pull/8031).
- Provide user's role on [auth permission error](https://github.com/coreos/etcd/pull/8164).
- Fix [auth store panic with disabled token](https://github.com/coreos/etcd/pull/8695).
### etcd server
- Add [`etcd --experimental-initial-corrupt-check`](https://github.com/coreos/etcd/pull/8554) flag to [check cluster database hashes before serving client/peer traffic](https://github.com/coreos/etcd/issues/8313).
-`etcd --experimental-initial-corrupt-check=false` by default.
- v3.4 will enable `--initial-corrupt-check=true` by default.
- Add [`etcd --experimental-corrupt-check-time`](https://github.com/coreos/etcd/pull/8420) flag to [raise corrupt alarm monitoring](https://github.com/coreos/etcd/issues/7125).
-`etcd --experimental-corrupt-check-time=0s` disabled by default.
- Add [`etcd --experimental-enable-v2v3`](https://github.com/coreos/etcd/pull/8407) flag to [emulate v2 API with v3](https://github.com/coreos/etcd/issues/6925).
-`etcd --experimental-enable-v2v3=false` by default.
- Add [`etcd --max-txn-ops`](https://github.com/coreos/etcd/pull/7976) flag to [configure maximum number operations in transaction](https://github.com/coreos/etcd/issues/7826).
- Add [`etcd --max-request-bytes`](https://github.com/coreos/etcd/pull/7968) flag to [configure maximum client request size](https://github.com/coreos/etcd/issues/7923).
- If not configured, it defaults to 1.5 MiB.
- Add [`etcd --client-crl-file`, `--peer-crl-file`](https://github.com/coreos/etcd/pull/8124) flags for [Certificate revocation list](https://github.com/coreos/etcd/issues/4034).
- Add [`etcd --peer-cert-allowed-cn`](https://github.com/coreos/etcd/pull/8616) flag to support [CN-based auth for inter-peer connection](https://github.com/coreos/etcd/issues/8262).
- Add [`etcd --listen-metrics-urls`](https://github.com/coreos/etcd/pull/8242) flag for additional `/metrics` and `/health` endpoints.
- Support [additional (non) TLS `/metrics` endpoints for a TLS-enabled cluster](https://github.com/coreos/etcd/pull/8282).
- e.g. `etcd --listen-metrics-urls=https://localhost:2378,http://localhost:9379` to serve `/metrics` and `/health` on secure port 2378 and insecure port 9379.
- Useful for [bypassing critical APIs when monitoring etcd](https://github.com/coreos/etcd/issues/8060).
- Add [`etcd --auto-compaction-mode`](https://github.com/coreos/etcd/pull/8123) flag to [support revision-based compaction](https://github.com/coreos/etcd/issues/8098).
- Change `etcd --auto-compaction-retention` flag to [accept string values](https://github.com/coreos/etcd/pull/8563) with [finer granularity](https://github.com/coreos/etcd/issues/8503).
- Now that `etcd --auto-compaction-retention` accepts string values, etcd configuration YAML file `auto-compaction-retention` field must be changed to `string` type.
- Previously, `etcd --config-file etcd.config.yaml` can have `auto-compaction-retention: 24` field, now must be `auto-compaction-retention: "24"` or `auto-compaction-retention: "24h"`.
- If configured as `--auto-compaction-mode periodic --auto-compaction-retention "24h"`, the time duration value for `etcd --auto-compaction-retention` flag must be valid for [`time.ParseDuration`](https://golang.org/pkg/time/#ParseDuration) function in Go.
- e.g. `etcd --auto-compaction-mode=revision --auto-compaction-retention=1000` automatically `Compact` on `"latest revision" - 1000` every 5-minute (when latest revision is 30000, compact on revision 29000).
- e.g. `etcd --auto-compaction-mode=periodic --auto-compaction-retention=72h` automatically `Compact` with 72-hour retention windown, for every 7.2-hour.
- e.g. `etcd --auto-compaction-mode=periodic --auto-compaction-retention=30m` automatically `Compact` with 30-minute retention windown, for every 3-minute.
- Periodic compactor continues to record latest revisions for every 1/10 of given compaction period (e.g. 1-hour when `etcd --auto-compaction-mode=periodic --auto-compaction-retention=10h`).
- For every 1/10 of given compaction period, compactor uses the last revision that was fetched before compaction period, to discard historical data.
- The retention window of compaction period moves for every 1/10 of given compaction period.
- For instance, when hourly writes are 100 and `--auto-compaction-retention=10`, v3.1 compacts revision 1000, 2000, and 3000 for every 10-hour, while v3.2.x, v3.3.0, v3.3.1, and v3.3.2 compact revision 1000, 1100, and 1200 for every 1-hour. Futhermore, when writes per minute are 1000, v3.3.0, v3.3.1, and v3.3.2 with `--auto-compaction-mode=periodic --auto-compaction-retention=30m` compact revision 30000, 33000, and 36000, for every 3-minute with more finer granularity.
- Whether compaction succeeds or not, this process repeats for every 1/10 of given compaction period. If compaction succeeds, it just removes compacted revision from historical revision records.
- Serve [`/health` endpoint as unhealthy](https://github.com/coreos/etcd/pull/8272) when [alarm (e.g. `NOSPACE`) is raised or there's no leader](https://github.com/coreos/etcd/issues/8207).
- Define [`etcdhttp.Health`](https://godoc.org/github.com/coreos/etcd/etcdserver/api/etcdhttp#Health) struct with JSON encoder.
- Note that `"health"` field is [`string` type, not `bool`](https://github.com/coreos/etcd/pull/9143).
- e.g. `{"health":"false"}`, `{"health":"true"}`
- [Remove `"errors"` field](https://github.com/coreos/etcd/pull/9162) since `v3.3.0-rc.3` (did exist only in `v3.3.0-rc.0`, `v3.3.0-rc.1`, `v3.3.0-rc.2`).
- Move [logging setup to embed package](https://github.com/coreos/etcd/pull/8810)
- Disable gRPC server info-level logs by default (can be enabled with `etcd --debug` flag).
- Use [monotonic time in Go 1.9](https://github.com/coreos/etcd/pull/8507) for `lease` package.
- Warn on [empty hosts in advertise URLs](https://github.com/coreos/etcd/pull/8384).
- e.g. `etcd --advertise-client-urls=http://:2379`.
- Warn on [shadowed environment variables](https://github.com/coreos/etcd/pull/8385).
- Address [error on shadowed environment variables](https://github.com/coreos/etcd/issues/8380).
- etcd v3.4 will exit on this error.
### API
- Support [ranges in transaction comparisons](https://github.com/coreos/etcd/pull/8025) for [disconnected linearized reads](https://github.com/coreos/etcd/issues/7924).
- Add [nested transactions](https://github.com/coreos/etcd/pull/8102) to extend [proxy use cases](https://github.com/coreos/etcd/issues/7857).
- Add [lease comparison target in transaction](https://github.com/coreos/etcd/pull/8324).
- Add [hash by revision](https://github.com/coreos/etcd/pull/8263) for [better corruption checking against boltdb](https://github.com/coreos/etcd/issues/8016).
### client v3
- Add [health balancer](https://github.com/coreos/etcd/pull/8545) to fix [watch API hangs](https://github.com/coreos/etcd/issues/7247), improve [endpoint switch under network faults](https://github.com/coreos/etcd/issues/7941).
- [Refactor balancer](https://github.com/coreos/etcd/pull/8840) and add [client-side keepalive pings](https://github.com/coreos/etcd/pull/8199) to handle [network partitions](https://github.com/coreos/etcd/issues/8711).
- Add [`MaxCallSendMsgSize` and `MaxCallRecvMsgSize`](https://github.com/coreos/etcd/pull/9047) fields to [`clientv3.Config`](https://godoc.org/github.com/coreos/etcd/clientv3#Config).
- Fix [exceeded response size limit error in client-side](https://github.com/coreos/etcd/issues/9043).
- In previous versions(v3.2.10, v3.2.11), client response size was limited to only 4 MiB.
-`MaxCallSendMsgSize` default value is 2 MiB, if not configured.
-`MaxCallRecvMsgSize` default value is `math.MaxInt32`, if not configured.
- Accept [`Compare_LEASE`](https://github.com/coreos/etcd/pull/8324) in [`clientv3.Compare`](https://godoc.org/github.com/coreos/etcd/clientv3#Compare).
- Add [`LeaseValue` helper](https://github.com/coreos/etcd/pull/8488) to `Cmp``LeaseID` values in `Txn`.
- Add [`MoveLeader`](https://github.com/coreos/etcd/pull/8153) to `Maintenance`.
- Add [`HashKV`](https://github.com/coreos/etcd/pull/8351) to `Maintenance`.
- Add [`Leases`](https://github.com/coreos/etcd/pull/8358) to `Lease`.
- Add [`clientv3/ordering`](https://github.com/coreos/etcd/pull/8092) for enforce [ordering in serialized requests](https://github.com/coreos/etcd/issues/7623).
- Support [`etcdctl watch [key] [range_end] -- [exec-command…]`](https://github.com/coreos/etcd/pull/8919), equivalent to [v2 `etcdctl exec-watch`](https://github.com/coreos/etcd/issues/8814).
- Make `etcdctl watch -- [exec-command]` set environmental variables [`ETCD_WATCH_REVISION`, `ETCD_WATCH_EVENT_TYPE`, `ETCD_WATCH_KEY`, `ETCD_WATCH_VALUE`](https://github.com/coreos/etcd/pull/9142) for each event.
- Support [`etcdctl watch` with environmental variables `ETCDCTL_WATCH_KEY` and `ETCDCTL_WATCH_RANGE_END`](https://github.com/coreos/etcd/pull/9142).
- Enable [`clientv3.WithRequireLeader(context.Context)` for `watch`](https://github.com/coreos/etcd/pull/8672) command.
- Print [`"del"` instead of `"delete"`](https://github.com/coreos/etcd/pull/8297) in `txn` interactive mode.
- Print [`ETCD_INITIAL_ADVERTISE_PEER_URLS` in `member add`](https://github.com/coreos/etcd/pull/8332).
### etcdctl v3
- Handle [empty key permission](https://github.com/coreos/etcd/pull/8514) in `etcdctl`.
- Add [`grpc-proxy start --max-send-bytes`](https://github.com/coreos/etcd/pull/9250) flag to [configure maximum client request size](https://github.com/coreos/etcd/issues/7923).
- Add [`grpc-proxy start --max-recv-bytes`](https://github.com/coreos/etcd/pull/9250) flag to [configure maximum client request size](https://github.com/coreos/etcd/issues/7923).
- Fix [Snapshot API error handling](https://github.com/coreos/etcd/commit/dbd16d52fbf81e5fd806d21ff5e9148d5bf203ab).
- Fix [KV API `PrevKv` flag handling](https://github.com/coreos/etcd/pull/8366).
- Fix [KV API `KeysOnly` flag handling](https://github.com/coreos/etcd/pull/8552).
### gRPC gateway
- Replace [gRPC gateway](https://github.com/grpc-ecosystem/grpc-gateway) endpoint `/v3alpha` with [`/v3beta`](https://github.com/coreos/etcd/pull/8880).
- To deprecate [`/v3alpha`](https://github.com/coreos/etcd/issues/8125) in v3.4.
- In v3.3, `curl -L http://localhost:2379/v3alpha/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` still works as a fallback to `curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'`, but `curl -L http://localhost:2379/v3alpha/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` won't work in v3.4. Use `curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` instead.
- Support ["authorization" token](https://github.com/coreos/etcd/pull/7999).
- Support [websocket for bi-directional streams](https://github.com/coreos/etcd/pull/8257).
- Fix [`Watch` API with gRPC gateway](https://github.com/coreos/etcd/issues/8237).
- Upgrade gRPC gateway to [v1.3.0](https://github.com/coreos/etcd/issues/8838).
### etcd server
- Fix [backend database in-memory index corruption](https://github.com/coreos/etcd/pull/8127) issue on restore (only 3.2.0 is affected).
- Fix [watch restore from snapshot](https://github.com/coreos/etcd/pull/8427).
- Fix [`mvcc/backend.defragdb` nil-pointer dereference on create bucket failure](https://github.com/coreos/etcd/pull/9119).
- Fix [server crash](https://github.com/coreos/etcd/pull/8010) on [invalid transaction request from gRPC gateway](https://github.com/coreos/etcd/issues/7889).
- Prevent [server panic from member update/add](https://github.com/coreos/etcd/pull/9174) with [wrong scheme URLs](https://github.com/coreos/etcd/issues/9173).
- Make [peer dial timeout longer](https://github.com/coreos/etcd/pull/8599).
- See [coreos/etcd-operator#1300](https://github.com/coreos/etcd-operator/issues/1300) for more detail.
- Make server [wait up to request time-out](https://github.com/coreos/etcd/pull/8267) with [pending RPCs](https://github.com/coreos/etcd/issues/8224).
- Fix [`grpc.Server` panic on `GracefulStop`](https://github.com/coreos/etcd/pull/8987) with [TLS-enabled server](https://github.com/coreos/etcd/issues/8916).
Previous change logs can be found at [CHANGELOG-3.3](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.3.md).
## v3.4.0 (TBD 2018-09)
See [code changes](https://github.com/coreos/etcd/compare/v3.3.0...v3.4.0) and [v3.4 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_4.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.4 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_4.md).**
### Improved
- Rewrite [client balancer](https://github.com/coreos/etcd/pull/9860) with [new gRPC balancer interface](https://github.com/coreos/etcd/issues/9106).
- Add [backoff on watch retries on transient errors](https://github.com/coreos/etcd/pull/9840).
- Add [jitter to watch progress notify](https://github.com/coreos/etcd/pull/9278) to prevent [spikes in `etcd_network_client_grpc_sent_bytes_total`](https://github.com/coreos/etcd/issues/9246).
- e.g. `read-only range request "key:\"/a\" range_end:\"/b\" " with result "range_response_count:3 size:96" took too long (97.966µs) to execute`.
- Redact [request value field](https://github.com/coreos/etcd/pull/9822).
- Provide [response size](https://github.com/coreos/etcd/pull/9826).
- Improve [TLS setup error logging](https://github.com/coreos/etcd/pull/9518) to help debug [TLS-enabled cluster configuring issues](https://github.com/coreos/etcd/issues/9400).
- Improve [long-running concurrent read transactions under light write workloads](https://github.com/coreos/etcd/pull/9296).
- Previously, periodic commit on pending writes blocks incoming read transactions, even if there is no pending write.
- Now, periodic commit operation does not block concurrent read transactions, thus improves long-running read transaction performance.
- Improve [Raft Read Index timeout warning messages](https://github.com/coreos/etcd/pull/9897).
- Adjust [election timeout on server restart](https://github.com/coreos/etcd/pull/9415) to reduce [disruptive rejoining servers](https://github.com/coreos/etcd/issues/9333).
- Previously, etcd fast-forwards election ticks on server start, with only one tick left for leader election. This is to speed up start phase, without having to wait until all election ticks elapse. Advancing election ticks is useful for cross datacenter deployments with larger election timeouts. However, it was affecting cluster availability if the last tick elapses before leader contacts the restarted node.
- Now, when etcd restarts, it adjusts election ticks with more than one tick left, thus more time for leader to prevent disruptive restart.
- Add [Raft Pre-Vote feature](https://github.com/coreos/etcd/pull/9352) to reduce [disruptive rejoining servers](https://github.com/coreos/etcd/issues/9333).
- For instance, a flaky(or rejoining) member may drop in and out, and start campaign. This member will end up with a higher term, and ignore all incoming messages with lower term. In this case, a new leader eventually need to get elected, thus disruptive to cluster availability. Raft implements Pre-Vote phase to prevent this kind of disruptions. If enabled, Raft runs an additional phase of election to check if pre-candidate can get enough votes to win an election.
- e.g. `etcd --auto-compaction-mode=revision --auto-compaction-retention=1000` automatically `Compact` on `"latest revision" - 1000` every 5-minute (when latest revision is 30000, compact on revision 29000).
- e.g. Previously, `etcd --auto-compaction-mode=periodic --auto-compaction-retention=24h` automatically `Compact` with 24-hour retention windown for every 2.4-hour. Now, `Compact` happens for every 1-hour.
- e.g. Previously, `etcd --auto-compaction-mode=periodic --auto-compaction-retention=30m` automatically `Compact` with 30-minute retention windown for every 3-minute. Now, `Compact` happens for every 30-minute.
- Periodic compactor keeps recording latest revisions for every compaction period when given period is less than 1-hour, or for every 1-hour when given compaction period is greater than 1-hour (e.g. 1-hour when `etcd --auto-compaction-mode=periodic --auto-compaction-retention=24h`).
- For every compaction period or 1-hour, compactor uses the last revision that was fetched before compaction period, to discard historical data.
- The retention window of compaction period moves for every given compaction period or hour.
- For instance, when hourly writes are 100 and `etcd --auto-compaction-mode=periodic --auto-compaction-retention=24h`, `v3.2.x`, `v3.3.0`, `v3.3.1`, and `v3.3.2` compact revision 2400, 2640, and 2880 for every 2.4-hour, while `v3.3.3`*or later* compacts revision 2400, 2500, 2600 for every 1-hour.
- Futhermore, when `etcd --auto-compaction-mode=periodic --auto-compaction-retention=30m` and writes per minute are about 1000, `v3.3.0`, `v3.3.1`, and `v3.3.2` compact revision 30000, 33000, and 36000, for every 3-minute, while `v3.3.3`*or later* compacts revision 30000, 60000, and 90000, for every 30-minute.
- Make [Lease `Lookup` non-blocking with concurrent `Grant`/`Revoke`](https://github.com/coreos/etcd/pull/9229).
- Make etcd server return `raft.ErrProposalDropped` on internal Raft proposal drop in [v3 applier](https://github.com/coreos/etcd/pull/9549) and [v2 applier](https://github.com/coreos/etcd/pull/9558).
- e.g. a node is removed from cluster, or [`raftpb.MsgProp` arrives at current leader while there is an ongoing leadership transfer](https://github.com/coreos/etcd/issues/8975).
- Add [`snapshot`](https://github.com/coreos/etcd/pull/9118) package for easier snapshot workflow (see [`godoc.org/github.com/etcd/clientv3/snapshot`](https://godoc.org/github.com/coreos/etcd/clientv3/snapshot) for more).
- Improve [functional tester](https://github.com/coreos/etcd/tree/master/functional) coverage: [proxy layer to run network fault tests in CI](https://github.com/coreos/etcd/pull/9081), [TLS is enabled both for server and client](https://github.com/coreos/etcd/pull/9534), [liveness mode](https://github.com/coreos/etcd/issues/9230), [shuffle test sequence](https://github.com/coreos/etcd/issues/9381), [membership reconfiguration failure cases](https://github.com/coreos/etcd/pull/9564), [disastrous quorum loss and snapshot recover from a seed member](https://github.com/coreos/etcd/pull/9565), [embedded etcd](https://github.com/coreos/etcd/pull/9572).
- Improve [index compaction blocking](https://github.com/coreos/etcd/pull/9511) by using a copy on write clone to avoid holding the lock for the traversal of the entire index.
- Update [JWT methods](https://github.com/coreos/etcd/pull/9883) to allow for use of any supported signature method/algorithm.
- Add [Lease checkpointing](https://github.com/coreos/etcd/pull/9924) to persist remaining TTLs to the consensus log periodically so that long lived leases progress toward expiry in the presence of leader elections and server restarts.
### Breaking Changes
- Make [`ETCDCTL_API=3 etcdctl` default](https://github.com/coreos/etcd/issues/9600).
- Now, `etcdctl set foo bar` must be `ETCDCTL_API=2 etcdctl set foo bar`.
- Now, `ETCDCTL_API=3 etcdctl put foo bar` could be just `etcdctl put foo bar`.
- **Remove `etcd --ca-file` flag**, instead [use `etcd --trusted-ca-file`](https://github.com/coreos/etcd/pull/9470) (`etcd --ca-file` flag has been marked deprecated since v2.1).
- **Remove `etcd --peer-ca-file` flag**, instead [use `etcd --peer-trusted-ca-file`](https://github.com/coreos/etcd/pull/9470) (`etcd --peer-ca-file` flag has been marked deprecated since v2.1).
- **Remove `pkg/transport.TLSInfo.CAFile` field**, instead [use `pkg/transport.TLSInfo.TrustedCAFile`](https://github.com/coreos/etcd/pull/9470) (`CAFile` field has been marked deprecated since v2.1).
- e.g. exit with error on `--advertise-client-urls=http://:2379`.
- e.g. exit with error on `--initial-advertise-peer-urls=http://:2380`.
- Exit on [shadowed environment variables](https://github.com/coreos/etcd/pull/9382).
- Address [error on shadowed environment variables](https://github.com/coreos/etcd/issues/8380).
- e.g. exit with error on `ETCD_NAME=abc etcd --name=def`.
- e.g. exit with error on `ETCD_INITIAL_CLUSTER_TOKEN=abc etcd --initial-cluster-token=def`.
- e.g. exit with error on `ETCDCTL_ENDPOINTS=abc.com ETCDCTL_API=3 etcdctl endpoint health --endpoints=def.com`.
- Change [`etcdserverpb.AuthRoleRevokePermissionRequest/key,range_end` fields type from `string` to `bytes`](https://github.com/coreos/etcd/pull/9433).
- Rename [`etcd_debugging_mvcc_db_total_size_in_bytes` Prometheus metric to `etcd_mvcc_db_total_size_in_bytes`](https://github.com/coreos/etcd/pull/9819).
- Rename `etcdserver.ServerConfig.SnapCount` field to `etcdserver.ServerConfig.SnapshotCount`, to be consistent with the flag name `etcd --snapshot-count`.
- Rename `embed.Config.SnapCount` field to [`embed.Config.SnapshotCount`](https://github.com/coreos/etcd/pull/9745), to be consistent with the flag name `etcd --snapshot-count`.
- Change [`embed.Config.CorsInfo` in `*cors.CORSInfo` type to `embed.Config.CORS` in `map[string]struct{}` type](https://github.com/coreos/etcd/pull/9490).
- Now logger is set up automatically based on [`embed.Config.Logger`, `embed.Config.LogOutputs`, `embed.Config.Debug` fields](https://github.com/coreos/etcd/pull/9572).
- Rename [`etcd --log-output` to `etcd --log-outputs`](https://github.com/coreos/etcd/pull/9624) to support multiple log outputs.
- **`etcd --log-output`** will be deprecated in v3.5.
- Rename [**`embed.Config.LogOutput`** to **`embed.Config.LogOutputs`**](https://github.com/coreos/etcd/pull/9624) to support multiple log outputs.
- Change [**`embed.Config.LogOutputs`** type from `string` to `[]string`](https://github.com/coreos/etcd/pull/9579) to support multiple log outputs.
- Now that `etcd --log-outputs` accepts multiple writers, etcd configuration YAML file `log-outputs` field must be changed to `[]string` type.
- Previously, `etcd --config-file etcd.config.yaml` can have `log-outputs: default` field, now must be `log-outputs: [default]`.
- Change v3 `etcdctl snapshot` exit codes with [`snapshot` package](https://github.com/coreos/etcd/pull/9118/commits/df689f4280e1cce4b9d61300be13ca604d41670a).
- Exit on error with exit code 1 (no more exit code 5 or 6 on `snapshot save/restore` commands).
- Migrate dependency management tool from `glide` to [`golang/dep`](https://github.com/coreos/etcd/pull/9155).
-<= 3.3 puts `vendor` directory under `cmd/vendor` directory to [prevent conflicting transitive dependencies](https://github.com/coreos/etcd/issues/4913).
- 3.4 moves `cmd/vendor` directory to `vendor` at repository root.
- Remove recursive symlinks in `cmd` directory.
- Now `go get/install/build` on `etcd` packages (e.g. `clientv3`, `tools/benchmark`) enforce builds with etcd `vendor` directory.
- Replace [gRPC gateway](https://github.com/grpc-ecosystem/grpc-gateway) endpoint `/v3beta` with [`/v3`](https://github.com/coreos/etcd/pull/9298).
- To deprecate [`/v3beta`](https://github.com/coreos/etcd/issues/9189) in v3.5.
- In v3.4, `curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` still works as a fallback to `curl -L http://localhost:2379/v3/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'`, but `curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` won't work in v3.5. Use `curl -L http://localhost:2379/v3/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` instead.
- Change [`wal` package function signatures](https://github.com/coreos/etcd/pull/9572) to support [structured logger and logging to file](https://github.com/coreos/etcd/issues/9438) in server-side.
- Change [`etcd --experimental-enable-v2v3`](TODO) flag to `etcd --enable-v2v3`; v2 storage emulation is now stable.
- Move internal packages to `etcdserver`.
-`"github.com/coreos/etcd/alarm"` to `"github.com/coreos/etcd/etcdserver/api/v3alarm"`.
-`"github.com/coreos/etcd/compactor"` to `"github.com/coreos/etcd/etcdserver/api/v3compactor"`.
-`"github.com/coreos/etcd/discovery"` to `"github.com/coreos/etcd/etcdserver/api/v2discovery"`.
-`"github.com/coreos/etcd/etcdserver/auth"` to `"github.com/coreos/etcd/etcdserver/api/v2auth"`.
-`"github.com/coreos/etcd/etcdserver/membership"` to `"github.com/coreos/etcd/etcdserver/api/membership"`.
-`"github.com/coreos/etcd/etcdserver/stats"` to `"github.com/coreos/etcd/etcdserver/api/v2stats"`.
-`"github.com/coreos/etcd/error"` to `"github.com/coreos/etcd/etcdserver/api/v2error"`.
-`"github.com/coreos/etcd/rafthttp"` to `"github.com/coreos/etcd/etcdserver/api/rafthttp"`.
-`"github.com/coreos/etcd/snap"` to `"github.com/coreos/etcd/etcdserver/api/snap"`.
-`"github.com/coreos/etcd/store"` to `"github.com/coreos/etcd/etcdserver/api/v2store"`.
- Change [snapshot file permissions](https://github.com/coreos/etcd/pull/9977): On Linux, the snapshot file changes from readable by all (mode 0644) to readable by the user only (mode 0600).
### Dependency
- Upgrade [`google.golang.org/grpc`](https://github.com/grpc/grpc-go/releases) from [**`v1.7.5`**](https://github.com/grpc/grpc-go/releases/tag/v1.7.5) to [**`v1.13.0`**](https://github.com/grpc/grpc-go/releases/tag/v1.13.0).
- Upgrade [`github.com/golang/protobuf`](https://github.com/golang/protobuf/releases) from [**`golang/protobuf@1e59b77b5`**](https://github.com/golang/protobuf/commit/1e59b77b52bf8e4b449a57e6f79f21226d571845) to [**`v1.1.0`**](https://github.com/golang/protobuf/releases/tag/v1.1.0).
- Upgrade [`golang.org/x/crypto`](https://github.com/golang/crypto) from [**`crypto@9419663f5`**](https://github.com/golang/crypto/commit/9419663f5a44be8b34ca85f08abc5fe1be11f8a3) to [**`crypto@8ac0e0d97`**](https://github.com/golang/crypto/commit/8ac0e0d97ce45cd83d1d7243c060cb8461dda5e9).
- Upgrade [`golang.org/x/net`](https://github.com/golang/net) from [**`net@66aacef3d`**](https://github.com/golang/net/commit/66aacef3dd8a676686c7ae3716979581e8b03c47) to [**`net@db08ff08e`**](https://github.com/golang/net/commit/db08ff08e8622530d9ed3a0e8ac279f6d4c02196).
- Upgrade [`golang.org/x/sys`](https://github.com/golang/sys) from [**`sys@ebfc5b463`**](https://github.com/golang/sys/commit/ebfc5b4631820b793c9010c87fd8fef0f39eb082) to [**`sys@56ede360e`**](https://github.com/golang/sys/commit/56ede360ec1c541828fb88741b3f1049406d28f5).
- Upgrade [`golang.org/x/text`](https://github.com/golang/text) from [**`text@b19bf474d`**](https://github.com/golang/text/commit/b19bf474d317b857955b12035d2c5acb57ce8b01) to [**`text@f21a4dfb5`**](https://github.com/golang/text/commit/f21a4dfb5e38f5895301dc265a8def02365cc3d0).
- Upgrade [`golang.org/x/time`](https://github.com/golang/time) from [**`time@c06e80d93`**](https://github.com/golang/time/commit/c06e80d9300e4443158a03817b8a8cb37d230320) to [**`time@fbb02b229`**](https://github.com/golang/time/commit/fbb02b2291d28baffd63558aa44b4b56f178d650).
- Upgrade [`github.com/golang/protobuf`](https://github.com/golang/protobuf/releases) from [**`golang/protobuf@1e59b77b5`**](https://github.com/golang/protobuf/commit/1e59b77b52bf8e4b449a57e6f79f21226d571845) to [**`v1.1.0`**](https://github.com/golang/protobuf/releases/tag/v1.1.0).
- Upgrade [`gopkg.in/yaml.v2`](https://github.com/go-yaml/yaml/releases) from [**`yaml@cd8b52f82`**](https://github.com/go-yaml/yaml/commit/cd8b52f8269e0feb286dfeef29f8fe4d5b397e0b) to [**`yaml@5420a8b67`**](https://github.com/go-yaml/yaml/commit/5420a8b6744d3b0345ab293f6fcba19c978f1183).
- Upgrade [`github.com/dgrijalva/jwt-go`](https://github.com/dgrijalva/jwt-go/releases) from [**`v3.0.0`**](https://github.com/dgrijalva/jwt-go/releases/tag/v3.0.0) to [**`v3.2.0`**](https://github.com/dgrijalva/jwt-go/releases/tag/v3.2.0).
- Upgrade [`github.com/ugorji/go/codec`](https://github.com/ugorji/go/releases) to [**`v1.1.1`**](https://github.com/ugorji/go/releases/tag/v1.1.1), and [regenerate v2 `client`](https://github.com/coreos/etcd/pull/9494).
- Upgrade [`github.com/soheilhy/cmux`](https://github.com/soheilhy/cmux/releases) from [**`v0.1.3`**](https://github.com/soheilhy/cmux/releases/tag/v0.1.3) to [**`v0.1.4`**](https://github.com/soheilhy/cmux/releases/tag/v0.1.4).
- Upgrade [`github.com/google/btree`](https://github.com/google/btree/releases) from [**`google/btree@925471ac9`**](https://github.com/google/btree/commit/925471ac9e2131377a91e1595defec898166fe49) to [**`google/btree@e89373fe6`**](https://github.com/google/btree/commit/e89373fe6b4a7413d7acd6da1725b83ef713e6e4).
- Upgrade [`github.com/spf13/cobra`](https://github.com/spf13/cobra/releases) from [**`spf13/cobra@1c44ec8d3`**](https://github.com/spf13/cobra/commit/1c44ec8d3f1552cac48999f9306da23c4d8a288b) to [**`v0.0.3`**](https://github.com/spf13/cobra/releases/tag/v0.0.3).
- Upgrade [`github.com/spf13/pflag`](https://github.com/spf13/pflag/releases) from [**`v1.0.0`**](https://github.com/spf13/pflag/releases/tag/v1.0.0) to [**`spf13/pflag@1ce0cc6db`**](https://github.com/spf13/pflag/commit/1ce0cc6db4029d97571db82f85092fccedb572ce).
- Upgrade [`github.com/coreos/go-systemd`](https://github.com/coreos/go-systemd/releases) from [**`v15`**](https://github.com/coreos/go-systemd/releases/tag/v15) to [**`v17`**](https://github.com/coreos/go-systemd/releases/tag/v17).
- Upgrade [`github.com/prometheus/client_golang`](https://github.com/prometheus/client_golang/releases) from [**``prometheus/client_golang@5cec1d042``**](https://github.com/prometheus/client_golang/commit/5cec1d0429b02e4323e042eb04dafdb079ddf568) to [**`v0.8.0`**](https://github.com/prometheus/client_golang/releases/tag/v0.8.0).
- Upgrade [`github.com/grpc-ecosystem/go-grpc-prometheus`](https://github.com/grpc-ecosystem/go-grpc-prometheus/releases) from [**``grpc-ecosystem/go-grpc-prometheus@0dafe0d49``**](https://github.com/grpc-ecosystem/go-grpc-prometheus/commit/0dafe0d496ea71181bf2dd039e7e3f44b6bd11a7) to [**`v1.2.0`**](https://github.com/grpc-ecosystem/go-grpc-prometheus/releases/tag/v1.2.0).
- Upgrade [`github.com/grpc-ecosystem/grpc-gateway`](https://github.com/grpc-ecosystem/grpc-gateway/releases) from [**`v1.3.1`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.3.1) to [**`v1.4.1`**](https://github.com/grpc-ecosystem/grpc-gateway/releases/tag/v1.4.1).
### Metrics, Monitoring
See [List of metrics](https://etcd.readthedocs.io/en/latest/operate.html#latest) for all metrics per release.
Note that any `etcd_debugging_*` metrics are experimental and subject to change.
- Let's say `"7339c4e5e833c029"` server `/metrics` returns `etcd_network_active_peers{Local="7339c4e5e833c029",Remote="729934363faa4a24"} 1` and `etcd_network_active_peers{Local="7339c4e5e833c029",Remote="b548c2511513015"} 1`. This indicates that the local node `"7339c4e5e833c029"` currently has two active remote peers `"729934363faa4a24"` and `"b548c2511513015"` in a 3-node cluster. If the node `"b548c2511513015"` is down, the local node `"7339c4e5e833c029"` will show `etcd_network_active_peers{Local="7339c4e5e833c029",Remote="729934363faa4a24"} 1` and `etcd_network_active_peers{Local="7339c4e5e833c029",Remote="b548c2511513015"} 0`.
- If a remote peer `"b548c2511513015"` is down, the local node `"7339c4e5e833c029"` server `/metrics` would return `etcd_network_disconnected_peers_total{Local="7339c4e5e833c029",Remote="b548c2511513015"} 1`, while active peer metrics will show `etcd_network_active_peers{Local="7339c4e5e833c029",Remote="729934363faa4a24"} 1` and `etcd_network_active_peers{Local="7339c4e5e833c029",Remote="b548c2511513015"} 0`.
- Use it with `etcd_mvcc_db_total_size_in_bytes` and `etcd_mvcc_db_total_size_in_use_in_bytes`.
-`etcd_server_quota_backend_bytes 2.147483648e+09` means current quota size is 2 GB.
-`etcd_mvcc_db_total_size_in_bytes 20480` means current physically allocated DB size is 20 KB.
-`etcd_mvcc_db_total_size_in_use_in_bytes 16384` means future DB size if defragment operation is complete.
-`etcd_mvcc_db_total_size_in_bytes - etcd_mvcc_db_total_size_in_use_in_bytes` is the number of bytes that can be saved on disk with defragment operation.
- Use it with `etcd_mvcc_db_total_size_in_bytes` and `etcd_mvcc_db_total_size_in_use_in_bytes`.
-`etcd_server_quota_backend_bytes 2.147483648e+09` means current quota size is 2 GB.
-`etcd_mvcc_db_total_size_in_bytes 20480` means current physically allocated DB size is 20 KB.
-`etcd_mvcc_db_total_size_in_use_in_bytes 16384` means future DB size if defragment operation is complete.
-`etcd_mvcc_db_total_size_in_bytes - etcd_mvcc_db_total_size_in_use_in_bytes` is the number of bytes that can be saved on disk with defragment operation.
- Add [`etcd --host-whitelist`](https://github.com/coreos/etcd/pull/9372) flag, [`etcdserver.Config.HostWhitelist`](https://github.com/coreos/etcd/pull/9372), and [`embed.Config.HostWhitelist`](https://github.com/coreos/etcd/pull/9372), to prevent ["DNS Rebinding"](https://en.wikipedia.org/wiki/DNS_rebinding) attack.
- Any website can simply create an authorized DNS name, and direct DNS to `"localhost"` (or any other address). Then, all HTTP endpoints of etcd server listening on `"localhost"` becomes accessible, thus vulnerable to [DNS rebinding attacks (CVE-2018-5702)](https://bugs.chromium.org/p/project-zero/issues/detail?id=1447#c2).
- Client origin enforce policy works as follow:
- If client connection is secure via HTTPS, allow any hostnames..
- If client connection is not secure and `"HostWhitelist"` is not empty, only allow HTTP requests whose Host field is listed in whitelist.
- By default, `"HostWhitelist"` is `"*"`, which means insecure server allows all client HTTP requests.
- Note that the client origin policy is enforced whether authentication is enabled or not, for tighter controls.
- When specifying hostnames, loopback addresses are not added automatically. To allow loopback interfaces, add them to whitelist manually (e.g. `"localhost"`, `"127.0.0.1"`, etc.).
- e.g. `etcd --host-whitelist example.com`, then the server will reject all HTTP requests whose Host field is not `example.com` (also rejects requests to `"localhost"`).
- Support [`etcd --cors`](https://github.com/coreos/etcd/pull/9490) in v3 HTTP requests (gRPC gateway).
- Support [`ttl` field for `etcd` Authentication JWT token](https://github.com/coreos/etcd/pull/8302).
- e.g. `etcd --auth-token jwt,pub-key=<pub key path>,priv-key=<priv key path>,sign-method=<sign method>,ttl=5m`.
- Allow empty token provider in [`etcdserver.ServerConfig.AuthToken`](https://github.com/coreos/etcd/pull/9369).
- Fix [TLS reload](https://github.com/coreos/etcd/pull/9570) when [certificate SAN field only includes IP addresses but no domain names](https://github.com/coreos/etcd/issues/9541).
- In Go, server calls `(*tls.Config).GetCertificate` for TLS reload if and only if server's `(*tls.Config).Certificates` field is not empty, or `(*tls.ClientHelloInfo).ServerName` is not empty with a valid SNI from the client. Previously, etcd always populates `(*tls.Config).Certificates` on the initial client TLS handshake, as non-empty. Thus, client was always expected to supply a matching SNI in order to pass the TLS verification and to trigger `(*tls.Config).GetCertificate` to reload TLS assets.
- However, a certificate whose SAN field does [not include any domain names but only IP addresses](https://github.com/coreos/etcd/issues/9541) would request `*tls.ClientHelloInfo` with an empty `ServerName` field, thus failing to trigger the TLS reload on initial TLS handshake; this becomes a problem when expired certificates need to be replaced online.
- Now, `(*tls.Config).Certificates` is created empty on initial TLS client handshake, first to trigger `(*tls.Config).GetCertificate`, and then to populate rest of the certificates on every new TLS connection, even when client SNI is empty (e.g. cert only includes IPs).
### etcd server
- Add [`etcd --initial-election-tick-advance`](https://github.com/coreos/etcd/pull/9591) flag to configure initial election tick fast-forward.
- By default, `etcd --initial-election-tick-advance=true`, then local member fast-forwards election ticks to speed up "initial" leader election trigger.
- This benefits the case of larger election ticks. For instance, cross datacenter deployment may require longer election timeout of 10-second. If true, local node does not need wait up to 10-second. Instead, forwards its election ticks to 8-second, and have only 2-second left before leader election.
- Major assumptions are that: cluster has no active leader thus advancing ticks enables faster leader election. Or cluster already has an established leader, and rejoining follower is likely to receive heartbeats from the leader after tick advance and before election timeout.
- However, when network from leader to rejoining follower is congested, and the follower does not receive leader heartbeat within left election ticks, disruptive election has to happen thus affecting cluster availabilities.
- Now, this can be disabled by setting `etcd --initial-election-tick-advance=false`.
- Disabling this would slow down initial bootstrap process for cross datacenter deployments. Make tradeoffs by configuring `etcd --initial-election-tick-advance` at the cost of slow initial bootstrap.
- Add [`etcd --pre-vote`](https://github.com/coreos/etcd/pull/9352) flag to enable to run an additional Raft election phase.
- For instance, a flaky(or rejoining) member may drop in and out, and start campaign. This member will end up with a higher term, and ignore all incoming messages with lower term. In this case, a new leader eventually need to get elected, thus disruptive to cluster availability. Raft implements Pre-Vote phase to prevent this kind of disruptions. If enabled, Raft runs an additional phase of election to check if pre-candidate can get enough votes to win an election.
-`etcd --pre-vote=false` by default.
- v3.5 will enable `etcd --pre-vote=true` by default.
- [`etcd --initial-corrupt-check`](TODO) flag is now stable (`etcd --experimental-initial-corrupt-check`haisbeen deprecated).
-`etcd --initial-corrupt-check=true` by default, to check cluster database hashes before serving client/peer traffic.
- [`etcd --corrupt-check-time`](TODO) flag is now stable (`etcd --experimental-corrupt-check-time`haisbeen deprecated).
-`etcd --corrupt-check-time=12h` by default, to check cluster database hashes for every 12-hour.
- [`etcd --enable-v2v3`](TODO) flag is now stable.
-`etcd --experimental-enable-v2v3` has been deprecated.
-`etcd --enable-v2=true --enable-v2v3=''` by default, to enable v2 API server that is backed by **v2 store**.
-`etcd --enable-v2=true --enable-v2v3=/aaa` to enable v2 API server that is backed by **v3 storage**.
-`etcd --enable-v2=false --enable-v2v3=''` to disable v2 API server.
-`etcd --enable-v2=false --enable-v2v3=/aaa` to disable v2 API server. TODO: error?
- Automatically [create parent directory if it does not exist](https://github.com/coreos/etcd/pull/9626) (fix [issue#9609](https://github.com/coreos/etcd/issues/9609)).
- v4.0 will configure `etcd --enable-v2=true --enable-v2v3=/aaa` to enable v2 API server that is backed by **v3 storage**.
- Add [`etcd --discovery-srv-name`](https://github.com/coreos/etcd/pull/8690) flag to support custom DNS SRV name with discovery.
- If not given, etcd queries `_etcd-server-ssl._tcp.[YOUR_HOST]` and `_etcd-server._tcp.[YOUR_HOST]`.
- If `etcd --discovery-srv-name="foo"`, then query `_etcd-server-ssl-foo._tcp.[YOUR_HOST]` and `_etcd-server-foo._tcp.[YOUR_HOST]`.
- Useful for operating multiple etcd clusters under the same domain.
- Support TLS cipher suite whitelisting.
- To block [weak cipher suites](https://github.com/coreos/etcd/issues/8320).
- TLS handshake fails when client hello is requested with invalid cipher suites.
- Support [`etcd --cors`](https://github.com/coreos/etcd/pull/9490) in v3 HTTP requests (gRPC gateway).
- Rename [`etcd --log-output` to `etcd --log-outputs`](https://github.com/coreos/etcd/pull/9624) to support multiple log outputs.
- **`etcd --log-output` will be deprecated in v3.5**.
- Add [`etcd --logger`](https://github.com/coreos/etcd/pull/9572) flag to support [structured logger and multiple log outputs](https://github.com/coreos/etcd/issues/9438) in server-side.
- **`etcd --logger=capnslog` will be deprecated in v3.5**.
- Main motivation is to promote automated etcd monitoring, rather than looking back server logs when it starts breaking. Future development will make etcd log as few as possible, and make etcd easier to monitor with metrics and alerts.
-`etcd --logger=capnslog --log-outputs=default` is the default setting and same as previous etcd server logging format.
-`etcd --logger=zap --log-outputs=default` is not supported when `etcd --logger=zap`.
- Instead, use `etcd --logger=zap --log-outputs=stderr`.
- Or, use `etcd --logger=zap --log-outputs=systemd/journal` to send logs to the local systemd journal.
- Previously, if etcd parent process ID (PPID) is 1 (e.g. run with systemd), `etcd --logger=capnslog --log-outputs=default` redirects server logs to local systemd journal. And if write to journald fails, it writes to `os.Stderr` as a fallback.
- However, even with PPID 1, it can fail to dial systemd journal (e.g. run embedded etcd with Docker container). Then, [every single log write will fail](https://github.com/coreos/etcd/pull/9729) and fall back to `os.Stderr`, which is inefficient.
- To avoid this problem, systemd journal logging must be configured manually.
-`etcd --logger=zap --log-outputs=stderr` will log server operations in [JSON-encoded format](https://godoc.org/go.uber.org/zap#NewProductionEncoderConfig) and writes logs to `os.Stderr`. Use this to override journald log redirects.
-`etcd --logger=zap --log-outputs=stdout` will log server operations in [JSON-encoded format](https://godoc.org/go.uber.org/zap#NewProductionEncoderConfig) and writes logs to `os.Stdout` Use this to override journald log redirects.
-`etcd --logger=zap --log-outputs=a.log` will log server operations in [JSON-encoded format](https://godoc.org/go.uber.org/zap#NewProductionEncoderConfig) and writes logs to the specified file `a.log`.
-`etcd --logger=zap --log-outputs=a.log,b.log,c.log,stdout` [writes server logs to multiple files `a.log`, `b.log` and `c.log` at the same time](https://github.com/coreos/etcd/pull/9579) and outputs to `os.Stderr`, in [JSON-encoded format](https://godoc.org/go.uber.org/zap#NewProductionEncoderConfig).
-`etcd --logger=zap --log-outputs=/dev/null` will discard all server logs.
- "unsynced" watcher is watcher that needs to be in sync with events that have happened.
- That is, "unsynced" watcher is the slow watcher that was requested on old revision.
- "unsynced" watcher restore operation was not correctly populating its underlying watcher group.
- Which possibly causes [missing events from "unsynced" watchers](https://github.com/coreos/etcd/issues/9086).
- A node gets network partitioned with a watcher on a future revision, and falls behind receiving a leader snapshot after partition gets removed. When applying this snapshot, etcd watch storage moves current synced watchers to unsynced since sync watchers might have become stale during network partition. And reset synced watcher group to restart watcher routines. Previously, there was a bug when moving from synced watcher group to unsynced, thus client would miss events when the watcher was requested to the network-partitioned node.
- Fix [`mvcc` server panic from restore operation](https://github.com/coreos/etcd/pull/9775).
- Let's assume that a watcher had been requested with a future revision X and sent to node A that became network-partitioned thereafter. Meanwhile, cluster makes progress. Then when the partition gets removed, the leader sends a snapshot to node A. Previously if the snapshot's latest revision is still lower than the watch revision X, **etcd server panicked** during snapshot restore operation.
- Now, this server-side panic has been fixed.
- Fix [server panic on invalid Election Proclaim/Resign HTTP(S) requests](https://github.com/coreos/etcd/pull/9379).
- Previously, wrong-formatted HTTP requests to Election API could trigger panic in etcd server.
- e.g. `curl -L http://localhost:2379/v3/election/proclaim -X POST -d '{"value":""}'`, `curl -L http://localhost:2379/v3/election/resign -X POST -d '{"value":""}'`.
- Previously, `etcd --auto-compaction-mode revision --auto-compaction-retention 1` was [translated to revision retention 3600000000000](https://github.com/coreos/etcd/issues/9337).
- Now, `etcd --auto-compaction-mode revision --auto-compaction-retention 1` is correctly parsed as revision retention 1.
- Prevent [overflow by large `TTL` values for `Lease` `Grant`](https://github.com/coreos/etcd/pull/9399).
-`TTL` parameter to `Grant` request is unit of second.
- Leases with too large `TTL` values exceeding `math.MaxInt64` [expire in unexpected ways](https://github.com/coreos/etcd/issues/9374).
- Server now returns `rpctypes.ErrLeaseTTLTooLarge` to client, when the requested `TTL` is larger than *9,000,000,000 seconds* (which is >285 years).
- Again, etcd `Lease` is meant for short-periodic keepalives or sessions, in the range of seconds or minutes. Not for hours or days!
- Enable etcd server [`raft.Config.CheckQuorum` when starting with `ForceNewCluster`](https://github.com/coreos/etcd/pull/9347).
- Allow [non-WAL files in `etcd --wal-dir` directory](https://github.com/coreos/etcd/pull/9743).
- Previously, existing files such as [`lost+found`](https://github.com/coreos/etcd/issues/7287) in WAL directory prevent etcd server boot.
- Now, WAL directory that contains only `lost+found` or a file that's not suffixed with `.wal` is considered non-initialized.
### API
- Add [`snapshot`](https://github.com/coreos/etcd/pull/9118) package for snapshot restore/save operations (see [`godoc.org/github.com/etcd/clientv3/snapshot`](https://godoc.org/github.com/coreos/etcd/clientv3/snapshot) for more).
- Add [`watch_id` field to `etcdserverpb.WatchCreateRequest`](https://github.com/coreos/etcd/pull/9065) to allow user-provided watch ID to `mvcc`.
- Corresponding `watch_id` is returned via `etcdserverpb.WatchResponse`, if any.
- Add [`fragment` field to `etcdserverpb.WatchCreateRequest`](https://github.com/coreos/etcd/pull/9291) to request etcd server to [split watch events](https://github.com/coreos/etcd/issues/9294) when the total size of events exceeds `etcd --max-request-bytes` flag value plus gRPC-overhead 512 bytes.
- The default server-side request bytes limit is `embed.DefaultMaxRequestBytes` which is 1.5 MiB plus gRPC-overhead 512 bytes.
- If watch response events exceed this server-side request limit and watch request is created with `fragment` field `true`, the server will split watch events into a set of chunks, each of which is a subset of watch events below server-side request limit.
- Useful when client-side has limited bandwidths.
- For example, watch response contains 10 events, where each event is 1 MiB. And server `etcd --max-request-bytes` flag value is 1 MiB. Then, server will send 10 separate fragmented events to the client.
- For example, watch response contains 5 events, where each event is 2 MiB. And server `etcd --max-request-bytes` flag value is 1 MiB and `clientv3.Config.MaxCallRecvMsgSize` is 1 MiB. Then, server will try to send 5 separate fragmented events to the client, and the client will error with `"code = ResourceExhausted desc = grpc: received message larger than max (...)"`.
- Client must implement fragmented watch event merge (which `clientv3` does in etcd v3.4).
- Add [`raftAppliedIndex` field to `etcdserverpb.StatusResponse`](https://github.com/coreos/etcd/pull/9176) for current Raft applied index.
- Add [`errors` field to `etcdserverpb.StatusResponse`](https://github.com/coreos/etcd/pull/9206) for server-side error.
- e.g. `"etcdserver: no leader", "NOSPACE", "CORRUPT"`
- Add [`dbSizeInUse` field to `etcdserverpb.StatusResponse`](https://github.com/coreos/etcd/pull/9256) for actual DB size after compaction.
- To manually trigger broadcasting watch progress event (empty watch response with latest header) to all associated watch streams.
- Think of it as `WithProgressNotify` that can be triggered manually.
Note: **v3.5 will deprecate `etcd --log-package-levels` flag for `capnslog`**; `etcd --logger=zap --log-outputs=stderr` will the default. **v3.5 will deprecate `[CLIENT-URL]/config/local/log` endpoint.**
### Package `embed`
- Add [`embed.Config.CipherSuites`](https://github.com/coreos/etcd/pull/9801) to specify a list of supported cipher suites for TLS handshake between client/server and peers.
- If empty, Go auto-populates the list.
- Both `embed.Config.ClientTLSInfo.CipherSuites` and `embed.Config.CipherSuites` cannot be non-empty at the same time.
- If not empty, specify either `embed.Config.ClientTLSInfo.CipherSuites` or `embed.Config.CipherSuites`.
- Add [`embed.Config.InitialElectionTickAdvance`](https://github.com/coreos/etcd/pull/9591) to enable/disable initial election tick fast-forward.
-`embed.NewConfig()` would return `*embed.Config` with `InitialElectionTickAdvance` as true by default.
- Define [`embed.CompactorModePeriodic`](https://godoc.org/github.com/coreos/etcd/embed#pkg-variables) for `compactor.ModePeriodic`.
- Define [`embed.CompactorModeRevision`](https://godoc.org/github.com/coreos/etcd/embed#pkg-variables) for `compactor.ModeRevision`.
- Change [`embed.Config.CorsInfo` in `*cors.CORSInfo` type to `embed.Config.CORS` in `map[string]struct{}` type](https://github.com/coreos/etcd/pull/9490).
- Now logger is set up automatically based on [`embed.Config.Logger`, `embed.Config.LogOutputs`, `embed.Config.Debug` fields](https://github.com/coreos/etcd/pull/9572).
- Add [`embed.Config.Logger`](https://github.com/coreos/etcd/pull/9518) to support [structured logger `zap`](https://github.com/uber-go/zap) in server-side.
- Rename `embed.Config.SnapCount` field to [`embed.Config.SnapshotCount`](https://github.com/coreos/etcd/pull/9745), to be consistent with the flag name `etcd --snapshot-count`.
- Rename [**`embed.Config.LogOutput`** to **`embed.Config.LogOutputs`**](https://github.com/coreos/etcd/pull/9624) to support multiple log outputs.
- Change [**`embed.Config.LogOutputs`** type from `string` to `[]string`](https://github.com/coreos/etcd/pull/9579) to support multiple log outputs.
### Package `integration`
- Add [`CLUSTER_DEBUG` to enable test cluster logging](https://github.com/coreos/etcd/pull/9678).
- Deprecated `capnslog` in integration tests.
### client v3
- Add [`WithFragment` `OpOption`](https://github.com/coreos/etcd/pull/9291) to support [watch events fragmentation](https://github.com/coreos/etcd/issues/9294) when the total size of events exceeds `etcd --max-request-bytes` flag value plus gRPC-overhead 512 bytes.
- Watch fragmentation is disabled by default.
- The default server-side request bytes limit is `embed.DefaultMaxRequestBytes` which is 1.5 MiB plus gRPC-overhead 512 bytes.
- If watch response events exceed this server-side request limit and watch request is created with `fragment` field `true`, the server will split watch events into a set of chunks, each of which is a subset of watch events below server-side request limit.
- Useful when client-side has limited bandwidths.
- For example, watch response contains 10 events, where each event is 1 MiB. And server `etcd --max-request-bytes` flag value is 1 MiB. Then, server will send 10 separate fragmented events to the client.
- For example, watch response contains 5 events, where each event is 2 MiB. And server `etcd --max-request-bytes` flag value is 1 MiB and `clientv3.Config.MaxCallRecvMsgSize` is 1 MiB. Then, server will try to send 5 separate fragmented events to the client, and the client will error with `"code = ResourceExhausted desc = grpc: received message larger than max (...)"`.
- To manually trigger broadcasting watch progress event (empty watch response with latest header) to all associated watch streams.
- Think of it as `WithProgressNotify` that can be triggered manually.
- Fix [lease keepalive interval updates when response queue is full](https://github.com/coreos/etcd/pull/9952).
- If `<-chan *clientv3LeaseKeepAliveResponse` from `clientv3.Lease.KeepAlive` was never consumed or channel is full, client was [sending keepalive request every 500ms](https://github.com/coreos/etcd/issues/9911) instead of expected rate of every "TTL / 3" duration.
- Change [snapshot file permissions](https://github.com/coreos/etcd/pull/9977): On Linux, the snapshot file changes from readable by all (mode 0644) to readable by the user only (mode 0600).
### etcdctl v3
- Make [`ETCDCTL_API=3 etcdctl` default](https://github.com/coreos/etcd/issues/9600).
- Now, `etcdctl set foo bar` must be `ETCDCTL_API=2 etcdctl set foo bar`.
- Now, `ETCDCTL_API=3 etcdctl put foo bar` could be just `etcdctl put foo bar`.
- Fix [`etcdctl move-leader` command for TLS-enabled endpoints](https://github.com/coreos/etcd/pull/9807).
- Add [`progress` command to `etcdctl watch --interactive`](https://github.com/coreos/etcd/pull/9869).
- To manually trigger broadcasting watch progress event (empty watch response with latest header) to all associated watch streams.
- Think of it as `WithProgressNotify` that can be triggered manually.
### gRPC proxy
- Fix [etcd server panic from restore operation](https://github.com/coreos/etcd/pull/9775).
- Let's assume that a watcher had been requested with a future revision X and sent to node A that became network-partitioned thereafter. Meanwhile, cluster makes progress. Then when the partition gets removed, the leader sends a snapshot to node A. Previously if the snapshot's latest revision is still lower than the watch revision X, **etcd server panicked** during snapshot restore operation.
- Especially, gRPC proxy was affected, since it detects a leader loss with a key `"proxy-namespace__lostleader"` and a watch revision `"int64(math.MaxInt64 - 2)"`.
- Now, this server-side panic has been fixed.
### gRPC gateway
- Replace [gRPC gateway](https://github.com/grpc-ecosystem/grpc-gateway) endpoint `/v3beta` with [`/v3`](https://github.com/coreos/etcd/pull/9298).
- To deprecate [`/v3beta`](https://github.com/coreos/etcd/issues/9189) in v3.5.
- In v3.4, `curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` still works as a fallback to `curl -L http://localhost:2379/v3/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'`, but `curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` won't work in v3.5. Use `curl -L http://localhost:2379/v3/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` instead.
- Add API endpoints [`/{v3beta,v3}/lease/leases, /{v3beta,v3}/lease/revoke, /{v3beta,v3}/lease/timetolive`](https://github.com/coreos/etcd/pull/9450).
- To deprecate [`/{v3beta,v3}/kv/lease/leases, /{v3beta,v3}/kv/lease/revoke, /{v3beta,v3}/kv/lease/timetolive`](https://github.com/coreos/etcd/issues/9430) in v3.5.
- Support [`etcd --cors`](https://github.com/coreos/etcd/pull/9490) in v3 HTTP requests (gRPC gateway).
### Package `raft`
- Fix [deadlock during PreVote migration process](https://github.com/coreos/etcd/pull/8525).
- Now [`(r *raft) Step` returns `raft.ErrProposalDropped`](https://github.com/coreos/etcd/pull/9137) if a proposal has been ignored.
- e.g. a node is removed from cluster, or [`raftpb.MsgProp` arrives at current leader while there is an ongoing leadership transfer](https://github.com/coreos/etcd/issues/8975).
- Improve [Raft `becomeLeader` and `stepLeader`](https://github.com/coreos/etcd/pull/9073) by keeping track of latest `pb.EntryConfChange` index.
- Previously record `pendingConf` boolean field scanning the entire tail of the log, which can delay hearbeat send.
- Fix [missing learner nodes on `(n *node) ApplyConfChange`](https://github.com/coreos/etcd/pull/9116).
### Tooling
- Add [`etcd-dump-logs --entry-type`](https://github.com/coreos/etcd/pull/9628) flag to support WAL log filtering by entry type.
- Add [`etcd-dump-logs --stream-decoder`](https://github.com/coreos/etcd/pull/9790) flag to support custom decoder.
### Go
- Require *Go 1.10+*.
- Compile with [*Go 1.10.3*](https://golang.org/doc/devel/release.html#go1.10).
Previous change logs can be found at [CHANGELOG-3.4](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.4.md).
## v3.5.0 (TBD 2018-12)
See [code changes](https://github.com/coreos/etcd/compare/v3.4.0...v3.5.0) and [v3.5 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_5.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v3.5 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_5.md).**
### Breaking Changes
- [gRPC gateway](https://github.com/grpc-ecosystem/grpc-gateway) only supports [`/v3`](TODO) endpoint.
-`curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` does work in v3.5. Use `curl -L http://localhost:2379/v3/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` instead.
- **`etcd --log-output` flag has been deprecated.** Use **`etcd --log-outputs`** instead.
- **`etcd --logger=zap --log-outputs=stderr`** is now the default.
- **`etcd --logger=capnslog` flag has been deprecated.**
- **`etcd --logger=zap --log-outputs=default` flag value is not supported.**.
- Instead, use `etcd --logger=zap --log-outputs=stderr`.
- Or, use `etcd --logger=zap --log-outputs=systemd/journal` to send logs to the local systemd journal.
- Previously, if etcd parent process ID (PPID) is 1 (e.g. run with systemd), `etcd --logger=capnslog --log-outputs=default` redirects server logs to local systemd journal. And if write to journald fails, it writes to `os.Stderr` as a fallback.
- However, even with PPID 1, it can fail to dial systemd journal (e.g. run embedded etcd with Docker container). Then, [every single log write will fail](https://github.com/coreos/etcd/pull/9729) and fall back to `os.Stderr`, which is inefficient.
- To avoid this problem, systemd journal logging must be configured manually.
- **`etcd --log-outputs=stderr`** is now the default.
- **`etcd --log-package-levels` flag for `capnslog` has been deprecated.** Now, **`etcd --logger=zap --log-outputs=stderr`** is the default.
- **`[CLIENT-URL]/config/local/log` endpoint has been deprecated, as is `etcd --log-package-levels` flag.**
-`curl -L http://localhost:2379/v3beta/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` does work in v3.5. Use `curl -L http://localhost:2379/v3/kv/put -X POST -d '{"key": "Zm9v", "value": "YmFy"}'` instead.
Previous change logs can be found at [CHANGELOG-3.x](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.x.md).
Planned breaking changes.
## v4.0.0 (TBD)
See [code changes](https://github.com/coreos/etcd/compare/v3.3.0...v4.0.0) and [v4.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_4_0.md) for any breaking changes. **Again, before running upgrades from any previous release, please make sure to read change logs below and [v4.0 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_4_0.md).**
### Breaking Changes
- [Secure etcd by default](https://github.com/coreos/etcd/issues/9475)?
- Change `/health` endpoint output.
- Previously, `{"health":"true"}`.
- Now, `{"health":true}`.
- Breaks [Kubernetes `kubectl get componentstatuses` command](https://github.com/kubernetes/kubernetes/issues/58240).
- Deprecate [`etcd --proxy*`](TODO) flags; **no more v2 proxy**.
- Deprecate [v2 storage backend](https://github.com/coreos/etcd/issues/9232); **no more v2 store**.
- v2 API is still supported via [v2 emulation](TODO).
- Deprecate [`etcdctl backup`](TODO) command.
-`clientv3.Client.KeepAlive(ctx context.Context, id LeaseID) (<-chan *LeaseKeepAliveResponse, error)` is now [`clientv4.Client.KeepAlive(ctx context.Context, id LeaseID) <-chan *LeaseKeepAliveResponse`](TODO).
- Similar to `Watch`, [`KeepAlive` does not return errors](https://github.com/coreos/etcd/issues/7488).
- If there's an unknown server error, kill all open channels and create a new stream on the next `KeepAlive` call.
- Rename `github.com/coreos/client` to `github.com/coreos/clientv2`.
*NOTE*: The watch features are under active development, and their memory usage may change as that development progresses. We do not expect it to significantly increase beyond the figures stated below.
Two components of etcd storage consume physical memory. The etcd process allocates an *in-memory index* to speed key lookup. The process's *page cache*, managed by the operating system, stores recently-accessed data from disk for quick re-use.
etcd v3 uses [gRPC][grpc] for its messaging protocol. The etcd project includes a gRPC-based [Go client][go-client] and a command line utility, [etcdctl][etcdctl], for communicating with an etcd cluster through gRPC. For languages with no gRPC support, etcd provides a JSON [gRPC gateway][grpc-gateway]. This gateway serves a RESTful proxy that translates HTTP/JSON requests into gRPC messages.
@@ -19,8 +18,6 @@ gRPC gateway endpoint has changed since etcd v3.3:
- etcd v3.5 or later uses only `[CLIENT-URL]/v3/*`.
- **`[CLIENT-URL]/v3beta/*` is deprecated**.
gRPC-gateway does not support authentication using TLS Common Name.
### Put and get keys
Use the `/v3/kv/range` and `/v3/kv/put` services to read and write keys:
For the most part, the etcd project is stable, but we are still moving fast! We believe in the release fast philosophy. We want to get early feedback on features still in development and stabilizing. Thus, there are, and will be more, experimental features and APIs. We plan to improve these features based on the early feedback from the community, or abandon them if there is little interest, in the next few releases. Please do not rely on any experimental features or APIs in production environment.
## The current experimental API/features are:
- [KV ordering](https://godoc.org/github.com/etcd-io/etcd/clientv3/ordering) wrapper. When an etcd client switches endpoints, responses to serializable reads may go backward in time if the new endpoint is lagging behind the rest of the cluster. The ordering wrapper caches the current cluster revision from response headers. If a response revision is less than the cached revision, the client selects another endpoint and reissues the read. Enable in grpcproxy with `--experimental-serializable-ordering`.
- [KV ordering](https://godoc.org/github.com/coreos/etcd/clientv3/ordering) wrapper. When an etcd client switches endpoints, responses to serializable reads may go backward in time if the new endpoint is lagging behind the rest of the cluster. The ordering wrapper caches the current cluster revision from response headers. If a response revision is less than the cached revision, the client selects another endpoint and reissues the read. Enable in grpcproxy with `--experimental-serializable-ordering`.
etcd provides a gRPC resolver to support an alternative name system that fetches endpoints from etcd for discovering gRPC services. The underlying mechanism is based on watching updates to keys prefixed with the service name.
@@ -10,8 +8,8 @@ The etcd client provides a gRPC resolver for resolving gRPC endpoints with an et
Users mostly interact with etcd by putting or getting the value of a key. This section describes how to do that by using etcdctl, a command line tool for interacting with etcd server. The concepts described here should apply to the gRPC APIs or client library APIs.
The API version used by etcdctl to speak to etcd may be set to version `2` or `3` via the `ETCDCTL_API` environment variable. By default, etcdctl on master (3.4) uses the v3 API and earlier versions (3.3 and earlier) default to the v2 API.
Note that any key that was created using the v2 API will not be able to be queried via the v2 API. A v3 API ```etcdctl get``` of a v2 key will exit with 0 and no key data, this is the expected behaviour.
By default, etcdctl talks to the etcd server with the v2 API for backward compatibility. For etcdctl to speak to etcd using the v3 API, the API version must be set to version 3 via the `ETCDCTL_API` environment variable. However note that any key that was created using the v2 API will not be able to be queried via the v3 API. A v3 API ```etcdctl get``` of a v2 key will exit with 0 and no key data, this is the expected behaviour.
For testing and development deployments, the quickest and easiest way is to configure a local cluster. For a production deployment, refer to the [clustering][clustering] section.
@@ -23,7 +21,14 @@ The running etcd member listens on `localhost:2379` for client requests.
Use `etcdctl` to interact with the running cluster:
1.Store an example key-value pair in the cluster:
1.Configure the environment to have `ETCDCTL_API=3` so `etcdctl` uses the etcd API version 3 instead of defaulting to version 2.
```
# use API version 3
$ export ETCDCTL_API=3
```
2. Store an example key-value pair in the cluster:
```
$ ./etcdctl put foo bar
@@ -32,7 +37,7 @@ Use `etcdctl` to interact with the running cluster:
If OK is printed, storing key-value pair is successful.
2. Retrieve the value of `foo`:
3. Retrieve the value of `foo`:
```
$ ./etcdctl get foo
@@ -65,7 +70,14 @@ A `Procfile` at the base of the etcd git repository is provided to easily config
Use `etcdctl` to interact with the running cluster:
1. Print the list of members:
1. Configure the environment to have `ETCDCTL_API=3` so `etcdctl` uses the etcd API version 3 instead of defaulting to version 2.
```
# use API version 3
$ export ETCDCTL_API=3
```
2. Print the list of members:
```
$ etcdctl --write-out=table --endpoints=localhost:2379 member list
@@ -82,7 +94,7 @@ Use `etcdctl` to interact with the running cluster:
etcd uses the [capnslog][capnslog] library for logging application output categorized into *levels*. A log message's level is determined according to these conventions:
- Create new stable branch through `git push origin ${VERSION_MAJOR}.${VERSION_MINOR}` if this is a major stable release. This assumes `origin` corresponds to "https://github.com/etcd-io/etcd".
- Bump [hardcoded Version in the repository](https://github.com/etcd-io/etcd/blob/master/version/version.go#L30) to the version `${VERSION}+git`.
- Create new stable branch through `git push origin ${VERSION_MAJOR}.${VERSION_MINOR}` if this is a major stable release. This assumes `origin` corresponds to "https://github.com/coreos/etcd".
- Bump [hardcoded Version in the repository](https://github.com/coreos/etcd/blob/master/version/version.go#L30) to the version `${VERSION}+git`.
etcd is a distributed key-value store designed to reliably and quickly preserve and provide access to critical data. It enables reliable distributed coordination through distributed locking, leader elections, and write barriers. An etcd cluster is intended for high availability and permanent data storage and retrieval.
## Getting started
New etcd users and developers should get started by [downloading and building][download_build] etcd. After getting etcd, follow this [quick demo][demo] to see the basics of creating and working with an etcd cluster.
## Developing with etcd
The easiest way to get started using etcd as a distributed key-value store is to [set up a local cluster][local_cluster].
- [Setting up local clusters][local_cluster]
- [Interacting with etcd][interacting]
- gRPC [etcd core][api_ref] and [etcd concurrency][api_concurrency_ref] API references
- [HTTP JSON API through the gRPC gateway][api_grpc_gateway]
- [gRPC naming and discovery][grpc_naming]
- [Client][namespace_client] and [proxy][namespace_proxy] namespacing
- [Embedding etcd][embed_etcd]
- [Experimental features and APIs][experimental]
- [System limits][system-limit]
## Operating etcd clusters
Administrators who need a fault-tolerant etcd cluster for either development or production should begin with a [cluster on multiple machines][clustering].
> NOTE: This project is *alpha* stage. Flags, configuration, behaviour and design may change significantly in following releases.
A set of customisable Prometheus alerts for etcd.
Instructions for use are the same as the [kubernetes-mixin](https://github.com/kubernetes-monitoring/kubernetes-mixin).
## Background
* For more information about monitoring mixins, see this [design doc](https://docs.google.com/document/d/1A9xvzwqnFVSOZ5fD3blKODXfsat5fg6ZhnKu9LK3lB4/edit#).
@@ -108,7 +106,7 @@ To recover from the low space quota alarm:
This is gRPC-side warning when a server receives a TCP RST flag with client-side streams being prematurely closed. For example, a client closes its connection, while gRPC server has not yet processed all HTTP/2 frames in the TCP queue. Some data may have been lost in server side, but it is ok so long as client connection has already been closed.
Only [old versions of gRPC](https://github.com/grpc/grpc-go/issues/1362) log this. etcd [>=v3.2.13 by default log this with DEBUG level](https://github.com/etcd-io/etcd/pull/9080), thus only visible with `--debug` flag enabled.
Only [old versions of gRPC](https://github.com/grpc/grpc-go/issues/1362) log this. etcd [>=v3.2.13 by default log this with DEBUG level](https://github.com/coreos/etcd/pull/9080), thus only visible with `--debug` flag enabled.
## Performance
@@ -149,14 +147,14 @@ etcd sends a snapshot of its complete key-value store to refresh slow followers
This document is meant to give an overview of the etcd3 API's central design. It is by no means all encompassing, but intended to focus on the basic ideas needed to understand etcd without the distraction of less common API calls. All etcd3 API's are defined in [gRPC services][grpc-service], which categorize remote procedure calls (RPCs) understood by the etcd server. A full listing of all etcd RPCs are documented in markdown in the [gRPC API listing][grpc-api].
etcd is a consistent and durable key value store with [mini-transaction][txn] support. The key value store is exposed through the KV APIs. etcd tries to ensure the strongest consistency and durability guarantees for a distributed system. This specification enumerates the KV API guarantees made by etcd.
@@ -28,7 +26,7 @@ The metadata for auth should also be stored and managed in the storage controlle
The authentication mechanism in the etcd v2 protocol has a tricky part because the metadata consistency should work as in the above, but does not: each permission check is processed by the etcd member that receives the client request (etcdserver/api/v2http/client.go), including follower members. Therefore, it's possible the check may be based on stale metadata.
This staleness means that auth configuration cannot be reflected as soon as operators execute etcdctl. Therefore there is no way to know how long the stale metadata is active. Practically, the configuration change is reflected immediately after the command execution. However, in some cases of heavy load, the inconsistent state can be prolonged and it might result in counter-intuitive situations for users and developers. It requires a workaround like this: https://github.com/etcd-io/etcd/pull/4317#issuecomment-179037582
This staleness means that auth configuration cannot be reflected as soon as operators execute etcdctl. Therefore there is no way to know how long the stale metadata is active. Practically, the configuration change is reflected immediately after the command execution. However, in some cases of heavy load, the inconsistent state can be prolonged and it might result in counter-intuitive situations for users and developers. It requires a workaround like this: https://github.com/coreos/etcd/pull/4317#issuecomment-179037582
### Inconsistent permissions are unsafe for linearized requests
etcd server has proven its robustness with years of failure injection testing. Most complex application logic is already handled by etcd server and its data stores (e.g. cluster membership is transparent to clients, with Raft-layer forwarding proposals to leader). Although server components are correct, its composition with client requires a different set of intricate protocols to guarantee its correctness and high availability under faulty conditions. Ideally, etcd server provides one logical cluster view of many physical machines, and client implements automatic failover between replicas. This documents client architectural decisions and its implementation details.
## Glossary
**clientv3** --- etcd Official Go client for etcd v3 API.
**clientv3-grpc1.0** --- Official client implementation, with [grpc-go v1.0.x](https://github.com/grpc/grpc-go/releases/tag/v1.0.0), which is used in latest etcd v3.1.
**clientv3-grpc1.7** --- Official client implementation, with [grpc-go v1.7.x](https://github.com/grpc/grpc-go/releases/tag/v1.7.0), which is used in latest etcd v3.2 and v3.3.
**clientv3-grpc1.14** --- Official client implementation, with [grpc-go v1.14.x](https://github.com/grpc/grpc-go/releases/tag/v1.14.0), which is used in latest etcd v3.4.
**Balancer** --- etcd client load balancer that implements retry and failover mechanism. etcd client should automatically balance loads between multiple endpoints.
**Endpoints** --- A list of etcd server endpoints that clients can connect to. Typically, 3 or 5 client URLs of an etcd cluster.
**Pinned endpoint** --- When configured with multiple endpoints, <= v3.3 client balancer chooses only one endpoint to establish a TCP connection, in order to conserve total open connections to etcd cluster. In v3.4, balancer round-robins pinned endpoints for every request, thus distributing loads more evenly.
**Client Connection** --- TCP connection that has been established to an etcd server, via gRPC Dial.
**Sub Connection** --- gRPC SubConn interface. Each sub-connection contains a list of addresses. Balancer creates a SubConn from a list of resolved addresses. gRPC ClientConn can map to multiple SubConn (e.g. example.com resolves to `10.10.10.1` and `10.10.10.2` of two sub-connections). etcd v3.4 balancer employs internal resolver to establish one sub-connection for each endpoint.
**Transient disconnect** --- When gRPC server returns a status error of [code Unavailable](https://godoc.org/google.golang.org/grpc/codes#Code).
## Client requirements
**Correctness** --- Requests may fail in the presence of server faults. However, it never violates consistency guarantees: global ordering properties, never write corrupted data, at-most once semantics for mutable operations, watch never observes partial events, and so on.
**Liveness** --- Servers may fail or disconnect briefly. Clients should make progress in either way. Clients should [never deadlock](https://github.com/etcd-io/etcd/issues/8980) waiting for a server to come back from offline, unless configured to do so. Ideally, clients detect unavailable servers with HTTP/2 ping and failover to other nodes with clear error messages.
**Effectiveness** --- Clients should operate effectively with minimum resources: previous TCP connections should be [gracefully closed](https://github.com/etcd-io/etcd/issues/9212) after endpoint switch. Failover mechanism should effectively predict the next replica to connect, without wastefully retrying on failed nodes.
**Portability** --- Official client should be clearly documented and its implementation be applicable to other language bindings. Error handling between different language bindings should be consistent. Since etcd is fully committed to gRPC, implementation should be closely aligned with gRPC long-term design goals (e.g. pluggable retry policy should be compatible with [gRPC retry](https://github.com/grpc/proposal/blob/master/A6-client-retries.md)). Upgrades between two client versions should be non-disruptive.
## Client overview
The etcd client implements the following components:
* balancer that establishes gRPC connections to an etcd cluster,
* API client that sends RPCs to an etcd server, and
* error handler that decides whether to retry a failed request or switch endpoints.
Languages may differ in how to establish an initial connection (e.g. configure TLS), how to encode and send Protocol Buffer messages to server, how to handle stream RPCs, and so on. However, errors returned from etcd server will be the same. So should be error handling and retry policy.
For example, etcd server may return `"rpc error: code = Unavailable desc = etcdserver: request timed out"`, which is transient error that expects retries. Or return `rpc error: code = InvalidArgument desc = etcdserver: key is not provided`, which means request was invalid and should not be retried. Go client can parse errors with `google.golang.org/grpc/status.FromError`, and Java client with `io.grpc.Status.fromThrowable`.
### clientv3-grpc1.0: Balancer Overview
`clientv3-grpc1.0` maintains multiple TCP connections when configured with multiple etcd endpoints. Then pick one address and use it to send all client requests. The pinned address is maintained until the client object is closed (see *Figure 1*). When the client receives an error, it randomly picks another and retries.
`clientv3-grpc1.0` opening multiple TCP connections may provide faster balancer failover but requires more resources. The balancer does not understand node’s health status or cluster membership. So, it is possible that balancer gets stuck with one failed or partitioned node.
### clientv3-grpc1.7: Balancer Overview
`clientv3-grpc1.7` maintains only one TCP connection to a chosen etcd server. When given multiple cluster endpoints, a client first tries to connect to them all. As soon as one connection is up, balancer pins the address, closing others (see **Figure 2**).
The pinned address is to be maintained until the client object is closed. An error, from server or client network fault, is sent to client error handler (see **Figure 3**).
The client error handler takes an error from gRPC server, and decides whether to retry on the same endpoint, or to switch to other addresses, based on the error code and message (see **Figure 4** and **Figure 5**).
Stream RPCs, such as Watch and KeepAlive, are often requested with no timeouts. Instead, client can send periodic HTTP/2 pings to check the status of a pinned endpoint; if the server does not respond to the ping, balancer switches to other endpoints (see **Figure 6**).
`clientv3-grpc1.7` balancer sends HTTP/2 keepalives to detect disconnects from streaming requests. It is a simple gRPC server ping mechanism and does not reason about cluster membership, thus unable to detect network partitions. Since partitioned gRPC server can still respond to client pings, balancer may get stuck with a partitioned node. Ideally, keepalive ping detects partition and triggers endpoint switch, before request time-out (see [issue #8673](https://github.com/etcd-io/etcd/issues/8673) and **Figure 7**).
`clientv3-grpc1.7` balancer maintains a list of unhealthy endpoints. Disconnected addresses are added to “unhealthy” list, and considered unavailable until after wait duration, which is hard coded as dial timeout with default value 5-second. Balancer can have false positives on which endpoints are unhealthy. For instance, endpoint A may come back right after being blacklisted, but still unusable for next 5 seconds (see **Figure 8**).
`clientv3-grpc1.0` suffered the same problems above.
Upstream gRPC Go had already migrated to new balancer interface. For example, `clientv3-grpc1.7` underlying balancer implementation uses new gRPC balancer and tries to be consistent with old balancer behaviors. While its compatibility has been maintained reasonably well, etcd client still [suffered from subtle breaking changes](https://github.com/grpc/grpc-go/issues/1649). Furthermore, gRPC maintainer recommends [not relying on the old balancer interface](https://github.com/grpc/grpc-go/issues/1942#issuecomment-375368665). In general, to get better support from upstream, it is best to be in sync with latest gRPC releases. And new features, such as retry policy, may not be backported to gRPC 1.7 branch. Thus, both etcd server and client must migrate to latest gRPC versions.
### clientv3-grpc1.14: Balancer Overview
`clientv3-grpc1.7` is so tightly coupled with old gRPC interface, that every single gRPC dependency upgrade broke client behavior. Majority of development and debugging efforts were devoted to fixing those client behavior changes. As a result, its implementation has become overly complicated with bad assumptions on server connectivities.
The primary goal of `clientv3-grpc1.14` is to simplify balancer failover logic; rather than maintaining a list of unhealthy endpoints, which may be stale, simply roundrobin to the next endpoint whenever client gets disconnected from the current endpoint. It does not assume endpoint status. Thus, no more complicated status tracking is needed (see *Figure 8* and above). Upgrading to `clientv3-grpc1.14` should be no issue; all changes were internal while keeping all the backward compatibilities.
Internally, when given multiple endpoints, `clientv3-grpc1.14` creates multiple sub-connections (one sub-connection per each endpoint), while `clientv3-grpc1.7` creates only one connection to a pinned endpoint (see *Figure 9*). For instance, in 5-node cluster, `clientv3-grpc1.14` balancer would require 5 TCP connections, while `clientv3-grpc1.7` only requires one. By preserving the pool of TCP connections, `clientv3-grpc1.14` may consume more resources but provide more flexible load balancer with better failover performance. The default balancing policy is round robin but can be easily extended to support other types of balancers (e.g. power of two, pick leader, etc.). `clientv3-grpc1.14` uses gRPC resolver group and implements balancer picker policy, in order to delegate complex balancing work to upstream gRPC. On the other hand, `clientv3-grpc1.7` manually handles each gRPC connection and balancer failover, which complicates the implementation. `clientv3-grpc1.14` implements retry in the gRPC interceptor chain that automatically handles gRPC internal errors and enables more advanced retry policies like backoff, while `clientv3-grpc1.7` manually interprets gRPC errors for retries.
Improvements can be made by caching the status of each endpoint. For instance, balancer can ping each server in advance to maintain a list of healthy candidates, and use this information when doing round-robin. Or when disconnected, balancer can prioritize healthy endpoints. This may complicate the balancer implementation, thus can be addressed in later versions.
Client-side keepalive ping still does not reason about network partitions. Streaming request may get stuck with a partitioned node. Advanced health checking service need to be implemented to understand the cluster membership (see [issue #8673](https://github.com/etcd-io/etcd/issues/8673) for more detail).
Currently, retry logic is handled manually as an interceptor. This may be simplified via [official gRPC retries](https://github.com/grpc/proposal/blob/master/A6-client-retries.md).
etcd is designed to reliably store infrequently updated data and provide reliable watch queries. etcd exposes previous versions of key-value pairs to support inexpensive snapshots and watch history events (“time travel queries”). A persistent, multi-version, concurrency-control data model is a good fit for these use cases.
Membership reconfiguration has been one of the biggest operational challenges. Let’s review common challenges.
A newly joined etcd member starts with no data, thus demanding more updates from leader until it catches up with leader’s logs. Then leader’s network is more likely to be overloaded, blocking or dropping leader heartbeats to followers. In such case, a follower may election-timeout to start a new leader election. That is, a cluster with a new member is more vulnerable to leader election. Both leader election and the subsequent update propagation to the new member are prone to causing periods of cluster unavailability (see **Figure 1** below).
What if network partition happens? It depends on leader partition. If the leader still maintains the active quorum, the cluster would continue to operate (see **Figure 2**).
What if the leader becomes isolated from the rest of the cluster? Leader monitors progress of each follower. When leader loses connectivity from the quorum it reverts back to follower which will affect the cluster availability (see **Figure 3**).
When a new node is added to 3 node cluster, the cluster size becomes 4 and the quorum size becomes 3. What if a new node had joined the cluster, and then network partition happens? It depends on which partition the new member gets located after partition. If the new node happens to be located in the same partition as leader’s, the leader still maintains the active quorum of 3. No leadership election happens, and no cluster availability gets affected (see **Figure 4**).
If the cluster is 2-and-2 partitioned, then neither of partition maintains the quorum of 3. In this case, leadership election happens (see **Figure 5**).
What if network partition happens first, and then a new member gets added? A partitioned 3-node cluster already has one disconnected follower. When a new member is added, the quorum changes from 2 to 3. Now, this cluster has only 2 active nodes out 4, thus losing quorum and starting a new leadership election (see **Figure 6**).
Since member add operation can change the size of quorum, it is always recommended to “member remove” first to replace an unhealthy node.
Adding a new member to a 1-node cluster changes the quorum size to 2, immediately causing a leader election when the previous leader finds out quorum is not active. This is because “member add” operation is a 2-step process where user needs to apply “member add” command first, and then starts the new node process (see **Figure 7**).
An even worse case is when an added member is misconfigured. Membership reconfiguration is a two-step process: “etcdctl member add” and starting an etcd server process with the given peer URL. That is, “member add” command is applied regardless of URL, even when the URL value is invalid. If the first step is applied with invalid URLs, the second step cannot even start the new etcd. Once the cluster loses quorum, there is no way to revert the membership change (see **Figure 8**).
Same applies to a multi-node cluster. For example, the cluster has two members down (one is failed, the other is misconfigured) and two members up, but now it requires at least 3 votes to change the cluster membership (see **Figure 9**).
As seen above, a simple misconfiguration can fail the whole cluster into an inoperative state. In such case, an operator need manually recreate the cluster with `etcd --force-new-cluster` flag. As etcd has become a mission-critical service for [Kubernetes](https://kubernetes.io), even the slightest outage may have significant impact on users. What can we better to make etcd such operations easier? Among other things, leader election is most critical to cluster availability: Can we make membership reconfiguration less disruptive by not changing the size of quorum? Can a new node be idle, only requesting the minimum updates from leader, until it catches up? Can membership misconfiguration be always reversible and handled in a more secure way (wrong member add command run should never fail the cluster)? Should an user worry about network topology when adding a new member? Can member add API work regardless of the location of nodes and ongoing network partitions?
## Raft learner
In order to mitigate such availability gaps in the previous section, [Raft §4.2.1](https://ramcloud.stanford.edu/~ongaro/thesis.pdf) introduces a new node state “Learner,” which joins the cluster as a **non-voting member** until it catches up to the leader’s logs.
## Features in v3.4
An operator should do the minimum amount of work possible to add a new learner node. `member add --learner` command to add a new learner, which joins cluster as a non-voting member but still receives all data from leader (see **Figure 10**).
When a learner has caught up with leader’s progress, the learner can be promoted to a voting member using the `member promote` API, which then counts towards the quorum (see **Figure 11**).
etcd server validates promote request to ensure its operational safety. Only after its log has caught up to leader’s can learner be promoted to a voting member (see **Figure 12**).
Learner only serves as a standby node until promoted: Leadership cannot be transferred to learner. Learner rejects client reads and writes (client balancer should not route requests to learner). Which means learner does not need issue Read Index requests to leader. Such limitation simplifies the initial learner implementation in v3.4 release (see **Figure 13**).
In addition, etcd limits the total number of learners that a cluster can have, and avoids overloading the leader with log replication. Learner never promotes itself. While etcd provides learner status information and safety checks, cluster operator must make the final decision whether to promote learner or not.
## Features in v3.5
**Make learner state only and default** --- Defaulting a new member state to learner will greatly improve membership reconfiguration safety, because learner does not change the size of quorum. Misconfiguration will always be reversible without losing the quorum.
**Make voting-member promotion fully automatic** --- Once a learner catches up to leader’s logs, a cluster can automatically promote the learner. etcd requires certain thresholds to be defined by the user, and once the requirements are satisfied, learner promotes itself to a voting member. From a user’s perspective, “member add” command would work the same way as today but with greater safety provided by learner feature.
**Make learner standby failover node** --- A learner joins as a standby node, and gets automatically promoted when the cluster availability is affected.
**Make learner read-only** --- A learner can serve as a read-only node that never gets promoted. In a weak consistency mode, learner only receives data from leader and never process writes. Serving reads locally without consensus overhead would greatly decrease the workloads to leader but may serve stale data. In a strong consistency mode, learner requests read index from leader to serve latest data, but still rejects writes.
## Learner vs. mirror maker
etcd implements “mirror maker” using watch API to continuously relay key creates and updates to a separate cluster. Mirroring usually has low latency overhead once it completes initial synchronization. Learner and mirroring overlap in that both can be used to replicate existing data for read-only. However, mirroring does not guarantee linearizability. During network disconnects, previous key-values might have been discarded, and clients are expected to verify watch responses for correct ordering. Thus, there is no ordering guarantee in mirror. Use mirror for minimum latency (e.g. cross data center) at the costs of consistency. Use learner to retain all historical data and its ordering.
## Appendix: learner implementation in v3.4
### Expose "Learner" node type to "MemberAdd" API
etcd client adds a flag to “MemberAdd” API for learner node. And etcd server handler applies membership change entry with `pb.ConfChangeAddLearnerNode` type. Once the command has been applied, a server joins the cluster with `etcd --initial-cluster-state=existing` flag. This learner node can neither vote nor count as quorum.
etcd server must not transfer leadership to learner, since it may still lag behind and does not count as quorum. etcd server limits the number of learners that cluster can have to one: the more learners we have, the more data the leader has to propagate. Clients may talk to learner node, but learner rejects all requests other than serializable read and member status API. This is for simplicity of initial implementation. In the future, learner can be extended as a read-only server that continuously mirrors cluster data. Client balancer must provide helper function to exclude learner node endpoint. Otherwise, request sent to learner may fail. Client sync member call should factor into learner node type. So should client endpoints update call.
`MemberList` and `MemberStatus` responses should indicate which node is learner.
### Add "MemberPromote" API
Internally in Raft, second `MemberAdd` call to learner node promotes it to a voting member. Leader maintains the progress of each follower and learner. If learner has not completed its snapshot message, reject promote request. Only accept promote request if and only if: The learner node is in a healthy state. The learner is in sync with leader or the delta is within the threshold (e.g. the number of entries to replicate to learner is less than 1/10 of snapshot count, which means it is less likely that even after promotion leader would not need send snapshot to the learner). All these logic are hard-coded in `etcdserver` package and not configurable.
## Reference
* Original GitHub issue ([issue #9161](https://github.com/etcd-io/etcd/issues/9161))
* Use case ([issue #3715](https://github.com/etcd-io/etcd/issues/3715))
* Use case ([issue #8888](https://github.com/etcd-io/etcd/issues/8888))
* Use case ([issue #10114](https://github.com/etcd-io/etcd/issues/10114))
The name "etcd" originated from two ideas, the unix "/etc" folder and "d"istributed systems. The "/etc" folder is a place to store configuration data for a single system whereas etcd stores configuration information for large scale distributed systems. Hence, a "d"istributed "/etc" is "etcd".
@@ -78,18 +76,18 @@ In theory, it’s possible to build these primitives atop any storage systems pr
For distributed coordination, choosing etcd can help prevent operational headaches and save engineering effort.
etcd uses [Prometheus][prometheus] for metrics reporting. The metrics can be used for real-time monitoring and debugging. etcd does not persist its metrics; if a member restarts, the metrics will be reset.
@@ -102,7 +99,7 @@ Abnormally high snapshot duration (`snapshot_save_total_duration_seconds`) indic
## Prometheus supplied metrics
The Prometheus client library provides a number of metrics under the `go` and `process` namespaces. There are a few that are particularly interesting.
The Prometheus client library provides a number of metrics under the `go` and `process` namespaces. There are a few that are particlarly interesting.
@@ -124,12 +122,12 @@ $ etcdctl role remove myrolename
## Enabling authentication
The minimal steps to enabling auth are as follows. The administrator can set up users and roles before or after enabling authentication, as a matter of preference.
The minimal steps to enabling auth are as follows. The administrator can set up users and roles before or after enabling authentication, as a matter of preference.
Make sure the root user is created:
```
$ etcdctl user add root
$ etcdctl user add root
Password of root:
```
@@ -169,10 +167,7 @@ $ etcdctl --user user --password password get foo
Otherwise, all `etcdctl` commands remain the same. Users and roles can still be created and modified, but require authentication by a user with the root role.
## Using TLS Common Name
As of version v3.2 if an etcd server is launched with the option `--client-cert-auth=true`, the field of Common Name (CN) in the client's TLS cert will be used as an etcd user. In this case, the common name authenticates the user and the client does not need a password. Note that if both of 1. `--client-cert-auth=true` is passed and CN is provided by the client, and 2. username and password are provided by the client, the username and password based authentication is prioritized. Note that this feature cannot be used with gRPC-proxy and gRPC-gateway. This is because gRPC-proxy terminates TLS from its client so all the clients share a cert of the proxy. gRPC-gateway uses a TLS connection internally for transforming HTTP request to gRPC request so it shares the same limitation. Therefore the clients cannot provide their CN to the server correctly. gRPC-proxy will cause an error and stop if a given cert has non empty CN. gRPC-proxy returns an error which indicates that the client has an non empty CN in its cert.
As of version v3.2 if an etcd server is launched with the option `--client-cert-auth=true`, the field of Common Name (CN) in the client's TLS cert will be used as an etcd user. In this case, the common name authenticates the user and the client does not need a password. Note that if both of 1. `--client-cert-auth=true` is passed and CN is provided by the client, and 2. username and password are provided by the client, the username and password based authentication is prioritized.
As of version v3.3 if an etcd server is launched with the option `--peer-cert-allowed-cn` filtering of CN inter-peer connections is enabled. Nodes can only join the etcd cluster if their CN match the allowed one.
See [etcd security page](https://github.com/etcd-io/etcd/blob/master/Documentation/op-guide/security.md) for more details.
## Notes on password strength
`etcdctl` command line interface and etcd API don't check a strength (length, coexistence of numbers and alphabets, etc) of the password during creating a new user or updating password of an existing user. An administrator needs to care about a requirement of password strength by themselves.
See [etcd security page](https://github.com/coreos/etcd/blob/master/Documentation/op-guide/security.md) for more details.
etcd is configurable through a configuration file, various command-line flags, and environment variables.
@@ -49,14 +47,14 @@ To start etcd automatically using custom settings at startup in Linux, using a [
+ env variable: ETCD_ELECTION_TIMEOUT
### --listen-peer-urls
+ List of URLs to listen on for peer traffic. This flag tells the etcd to accept incoming requests from its peers on the specified scheme://IP:port combinations. Scheme can be http or https. Alternatively, use `unix://<file-path>` or `unixs://<file-path>` for unix sockets.If 0.0.0.0 is specified as the IP, etcd listens to the given port on all interfaces. If an IP address is given as well as a port, etcd will listen on the given port and interface. Multiple URLs may be used to specify a number of addresses and ports to listen on. The etcd will respond to requests from any of the listed addresses and ports.
+ List of URLs to listen on for peer traffic. This flag tells the etcd to accept incoming requests from its peers on the specified scheme://IP:port combinations. Scheme can be either http or https.If 0.0.0.0 is specified as the IP, etcd listens to the given port on all interfaces. If an IP address is given as well as a port, etcd will listen on the given port and interface. Multiple URLs may be used to specify a number of addresses and ports to listen on. The etcd will respond to requests from any of the listed addresses and ports.
+ default: "http://localhost:2380"
+ env variable: ETCD_LISTEN_PEER_URLS
+ example: "http://10.0.0.1:2380"
+ invalid example: "http://example.com:2380" (domain name is invalid for binding)
### --listen-client-urls
+ List of URLs to listen on for client traffic. This flag tells the etcd to accept incoming requests from the clients on the specified scheme://IP:port combinations. Scheme can be either http or https. Alternatively, use `unix://<file-path>` or `unixs://<file-path>` for unix sockets. If 0.0.0.0 is specified as the IP, etcd listens to the given port on all interfaces. If an IP address is given as well as a port, etcd will listen on the given port and interface. Multiple URLs may be used to specify a number of addresses and ports to listen on. The etcd will respond to requests from any of the listed addresses and ports.
+ List of URLs to listen on for client traffic. This flag tells the etcd to accept incoming requests from the clients on the specified scheme://IP:port combinations. Scheme can be either http or https. If 0.0.0.0 is specified as the IP, etcd listens to the given port on all interfaces. If an IP address is given as well as a port, etcd will listen on the given port and interface. Multiple URLs may be used to specify a number of addresses and ports to listen on. The etcd will respond to requests from any of the listed addresses and ports.
+ default: "http://localhost:2379"
+ env variable: ETCD_LISTEN_CLIENT_URLS
+ example: "http://10.0.0.1:2379"
@@ -84,16 +82,6 @@ To start etcd automatically using custom settings at startup in Linux, using a [
+ default: 0
+ env variable: ETCD_QUOTA_BACKEND_BYTES
### --backend-batch-limit
+ BackendBatchLimit is the maximum operations before commit the backend transaction.
+ default: 0
+ env variable: ETCD_BACKEND_BATCH_LIMIT
### --backend-batch-interval
+ BackendBatchInterval is the maximum time before commit the backend transaction.
+ default: 0
+ env variable: ETCD_BACKEND_BATCH_INTERVAL
### --max-txn-ops
+ Maximum number of operations permitted in a transaction.
+ default: 128
@@ -184,7 +172,7 @@ To start etcd automatically using custom settings at startup in Linux, using a [
### --strict-reconfig-check
+ Reject reconfiguration requests that would cause quorum loss.
+ default: true
+ default: false
+ env variable: ETCD_STRICT_RECONFIG_CHECK
### --auto-compaction-retention
@@ -263,7 +251,6 @@ The security flags help to [build a secure etcd cluster][security].
+ Enable client cert authentication.
+ default: false
+ env variable: ETCD_CLIENT_CERT_AUTH
+ CN authentication is not supported by gRPC-gateway.
### --client-crl-file
+ Path to the client certificate revocation list file.
@@ -323,22 +310,9 @@ The security flags help to [build a secure etcd cluster][security].
+ default: none
+ env variable: ETCD_PEER_CERT_ALLOWED_CN
### --cipher-suites
+ Comma-separated list of supported TLS cipher suites between server/client and peers.
+ Specify 'zap' for structured logging or 'capnslog'.
+ default: capnslog
+ env variable: ETCD_LOGGER
@@ -346,8 +320,7 @@ The security flags help to [build a secure etcd cluster][security].
### --log-outputs
+ Specify 'stdout' or 'stderr' to skip journald logging even when running under systemd, or list of comma separated output targets.
+ default: default
+ env variable: ETCD_LOG_OUTPUTS
+ 'default' use 'stderr' config for v3.4 during zap logger migraion
+ env variable: ETCD_LOG_OUTPUT
### --debug
+ Drop the default log level to DEBUG for all subpackages.
@@ -366,7 +339,7 @@ For example, it may panic if other members in the cluster are still alive.
Follow the instructions when using these flags.
### --force-new-cluster
+ Force to create a new one-member cluster. It commits configuration changes forcing to remove all existing members in the cluster and add itself, but is strongly discouraged. Please review the [disaster recovery][recovery] documentation for preferred v3 recovery procedures.
+ Force to create a new one-member cluster. It commits configuration changes forcing to remove all existing members in the cluster and add itself. It needs to be set to [restore a backup][restore].
+ default: false
+ env variable: ETCD_FORCE_NEW_CLUSTER
@@ -380,24 +353,20 @@ Follow the instructions when using these flags.
+ Enable runtime profiling data via HTTP server. Address is at client URL + "/debug/pprof/"
+ default: false
+ env variable: ETCD_ENABLE_PPROF
### --metrics
+ Set level of detail for exported metrics, specify 'extensive' to include histogram metrics.
+ default: basic
+ env variable: ETCD_METRICS
### --listen-metrics-urls
+ List of additional URLs to listen on that will respond to both the `/metrics` and `/health` endpoints
+ default: ""
+ env variable: ETCD_LISTEN_METRICS_URLS
## Auth flags
@@ -406,24 +375,16 @@ Follow the instructions when using these flags.
+ For asymmetric algorithms ('RS', 'PS', 'ES'), the public key is optional, as the private key contains enough information to both sign and verify tokens.
+ Example option of JWT: '--auth-token jwt,pub-key=app.rsa.pub,priv-key=app.rsa,sign-method=RS512,ttl=10m'
+ default: "simple"
+ env variable: ETCD_AUTH_TOKEN
### --bcrypt-cost
+ Specify the cost / strength of the bcrypt algorithm for hashing auth passwords. Valid values are between 4 and 31.
+ default: 10
+ env variable: (not supported)
## Experimental flags
### --experimental-backend-bbolt-freelist-type
+ The freelist type that etcd backend(bboltdb) uses (array and map are supported types).
Failures are common in a large deployment of machines. A machine fails when its hardware or software malfunctions. Multiple machines fail together when there are power failures or network issues. Multiple kinds of failures can also happen at once; it is almost impossible to enumerate all possible failure cases.
etcd gateway is a simple TCP proxy that forwards network data to the etcd cluster. The gateway is stateless and transparent; it neither inspects client requests nor interferes with cluster responses. It does not terminate TLS connections, do TLS handshakes on behalf of its clients, or verify if the connection is secured.
etcd gateway is a simple TCP proxy that forwards network data to the etcd cluster. The gateway is stateless and transparent; it neither inspects client requests nor interferes with cluster responses.
The gateway supports multiple etcd server endpoints and works on a simple round-robin policy. It only routes to available endpoints and hides failures from its clients. Other retry policies, such as weighted round-robin, may be supported in the future.
The gateway supports multiple etcd server endpoints and works on a simple round-robin policy. It only routes to available enpoints and hides failures from its clients. Other retry policies, such as weighted round-robin, may be supported in the future.
## When to use etcd gateway
@@ -62,7 +60,7 @@ infra2.example.com. 300 IN A 10.0.1.12
Start the etcd gateway to fetch the endpoints from the DNS SRV entries with the command:
```bash
$ etcd gateway start --discovery-srv=example.com
$ etcd gateway --discovery-srv=example.com
2016-08-16 11:21:18.867350 I | tcpproxy: ready to proxy client requests to [...]
* Comma-separated list of etcd server targets for forwarding client connections.
* Default: `127.0.0.1:2379`
* Invalid example: `https://127.0.0.1:2379` (gateway does not terminate TLS). Note that the gateway does not verify the HTTP schema or inspect the requests, it only forwards requests to the given endpoints.
* Invalid example: `https://127.0.0.1:2379` (gateway does not terminate TLS)
* Path to the client TLS CA file for the etcd cluster to verify the endpoints returned from SRV discovery. Note that it is ONLY used for authenticating the discovered endpoints rather than creating connections for data transferring. The gateway never terminates TLS connections or create TLS connections on behalf of its clients.
* Path to the client TLS CA file for the etcd cluster. Used to authenticate endpoints.
The gRPC proxy is a stateless etcd reverse proxy operating at the gRPC layer (L7). The proxy is designed to reduce the total processing load on the core etcd cluster. For horizontal scalability, it coalesces watch and lease API requests. To protect the cluster against abusive clients, it caches key range requests.
@@ -196,7 +194,7 @@ $ ETCDCTL_API=3 etcdctl --endpoints=localhost:2379 get my-prefix/my-key
## TLS termination
Terminate TLS from a secure etcd cluster with the gRPC proxy by serving an unencrypted local endpoint.
Terminate TLS from a secure etcd cluster with the grpc proxy by serving an unencrypted local endpoint.
To try it out, start a single member etcd cluster with client https:
@@ -213,7 +211,7 @@ $ ETCDCTL_API=3 etcdctl --endpoints=http://localhost:2379 endpoint status
$ ETCDCTL_API=3 etcdctl --endpoints=https://localhost:2379 --cert=client.crt --key=client.key --cacert=ca.crt endpoint status
```
Next, start a gRPC proxy on `localhost:12379` by connecting to the etcd endpoint `https://localhost:2379` using the client certificates:
Next, start a grpc proxy on `localhost:12379` by connecting to the etcd endpoint `https://localhost:2379` using the client certificates:
@@ -225,28 +223,3 @@ Finally, test the TLS termination by putting a key into the proxy over http:
$ ETCDCTL_API=3 etcdctl --endpoints=http://localhost:12379 put abc def
# OK
```
## Metrics and Health
The gRPC proxy exposes `/health` and Prometheus `/metrics` endpoints for the etcd members defined by `--endpoints`. An alternative define an additional URL that will respond to both the `/metrics` and `/health` endpoints with the `--metrics-addr` flag.
```bash
$ etcd grpc-proxy start \
--endpoints https://localhost:2379 \
--metrics-addr https://0.0.0.0:4443 \
--listen-addr 127.0.0.1:23790 \
--key client.key \
--key-file proxy-server.key \
--cert client.crt \
--cert-file proxy-server.crt \
--cacert ca.pem \
--trusted-ca-file proxy-ca.pem
```
### Known issue
The main interface of the proxy serves both HTTP2 and HTTP/1.1. If proxy is setup with TLS as show in the above example, when using a client such as cURL against the listening interface will require explicitly setting the protocol to HTTP/1.1 on the request to return`/metrics` or `/health`. By using the `--metrics-addr` flag the secondary interface will not have this requirement.
etcd usually runs well with limited resources for development or testing purposes; it’s common to develop with etcd on a laptop or a cheap cloud machine. However, when running etcd clusters in production, some hardware guidelines are useful for proper administration. These suggestions are not hard rules; they serve as a good starting point for a robust production deployment. As always, deployments should be tested with simulated workloads before running in production.
@@ -14,7 +12,7 @@ All etcd maintenance manages storage resources consumed by the etcd keyspace. Fa
Higher `--snapshot-count` holds more Raft entries in memory until snapshot, thus causing [recurrent higher memory usage](https://github.com/kubernetes/kubernetes/issues/60589#issuecomment-371977156). Since leader retains latest Raft entries for longer, a slow follower has more time to catch up before leader snapshot. `--snapshot-count` is a tradeoff between higher memory usage and better availabilities of slow followers.
Since v3.2, the default value of `--snapshot-count` has [changed from from 10,000 to 100,000](https://github.com/etcd-io/etcd/pull/7160).
Since v3.2, the default value of `--snapshot-count` has [changed from from 10,000 to 100,000](https://github.com/coreos/etcd/pull/7160).
In performance-wise, `--snapshot-count` greater than 100,000 may impact the write throughput. Higher number of in-memory objects can slow down [Go GC mark phase `runtime.scanobject`](https://golang.org/src/runtime/mgc.go), and infrequent memory reclamation makes allocation slow. Performance varies depending on the workloads and system environments. However, in general, too frequent compaction affects cluster availabilities and write throughputs. Too infrequent compaction is also harmful placing too much pressure on Go garbage collector. See https://www.slideshare.net/mitakeh/understanding-performance-aspects-of-etcd-and-raft for more research results.
@@ -47,11 +45,11 @@ Error: rpc error: code = 11 desc = etcdserver: mvcc: required revision has been
$ etcd --auto-compaction-retention=1
```
[v3.0.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.0.md) and [v3.1.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.1.md) with `--auto-compaction-retention=10` run periodic compaction on v3 key-value store for every 10-hour. Compactor only supports periodic compaction. Compactor records latest revisions every 5-minute, until it reaches the first compaction period (e.g. 10-hour). In order to retain key-value history of last compaction period, it uses the last revision that was fetched before compaction period, from the revision records that were collected every 5-minute. When `--auto-compaction-retention=10`, compactor uses revision 100 for compact revision where revision 100 is the latest revision fetched from 10 hours ago. If compaction succeeds or requested revision has already been compacted, it resets period timer and starts over with new historical revision records (e.g. restart revision collect and compact for the next 10-hour period). If compaction fails, it retries in 5 minutes.
[v3.0.0](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.0.md) and [v3.1.0](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.1.md) with `--auto-compaction-retention=10` run periodic compaction on v3 key-value store for every 10-hour. Compactor only supports periodic compaction. Compactor records latest revisions every 5-minute, until it reaches the first compaction period (e.g. 10-hour). In order to retain key-value history of last compaction period, it uses the last revision that was fetched before compaction period, from the revision records that were collected every 5-minute. When `--auto-compaction-retention=10`, compactor uses revision 100 for compact revision where revision 100 is the latest revision fetched from 10 hours ago. If compaction succeeds or requested revision has already been compacted, it resets period timer and starts over with new historical revision records (e.g. restart revision collect and compact for the next 10-hour period). If compaction fails, it retries in 5 minutes.
[v3.2.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md) compactor runs [every hour](https://github.com/etcd-io/etcd/pull/7875). Compactor only supports periodic compaction. Compactor continues to record latest revisions every 5-minute. For every hour, it uses the last revision that was fetched before compaction period, from the revision records that were collected every 5-minute. That is, for every hour, compactor discards historical data created before compaction period. The retention window of compaction period moves to next hour. For instance, when hourly writes are 100 and `--auto-compaction-retention=10`, v3.1 compacts revision 1000, 2000, and 3000 for every 10-hour, while v3.2.x, v3.3.0, v3.3.1, and v3.3.2 compact revision 1000, 1100, and 1200 for every 1-hour. If compaction succeeds or requested revision has already been compacted, it resets period timer and removes used compacted revision from historical revision records (e.g. start next revision collect and compaction from previously collected revisions). If compaction fails, it retries in 5 minutes.
[v3.2.0](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.2.md) compactor runs [every hour](https://github.com/coreos/etcd/pull/7875). Compactor only supports periodic compaction. Compactor continues to record latest revisions every 5-minute. For every hour, it uses the last revision that was fetched before compaction period, from the revision records that were collected every 5-minute. That is, for every hour, compactor discards historical data created before compaction period. The retention window of compaction period moves to next hour. For instance, when hourly writes are 100 and `--auto-compaction-retention=10`, v3.1 compacts revision 1000, 2000, and 3000 for every 10-hour, while v3.2.x, v3.3.0, v3.3.1, and v3.3.2 compact revision 1000, 1100, and 1200 for every 1-hour. If compaction succeeds or requested revision has already been compacted, it resets period timer and removes used compacted revision from historical revision records (e.g. start next revision collect and compaction from previously collected revisions). If compaction fails, it retries in 5 minutes.
In [v3.3.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.3.md), [v3.3.1](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.3.md), and [v3.3.2](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.3.md), `--auto-compaction-mode=revision --auto-compaction-retention=1000` automatically `Compact` on `"latest revision" - 1000` every 5-minute (when latest revision is 30000, compact on revision 29000). For instance, `--auto-compaction-mode=periodic --auto-compaction-retention=72h` automatically `Compact` with 72-hour retention windown, for every 7.2-hour. For instance, `--auto-compaction-mode=periodic --auto-compaction-retention=30m` automatically `Compact` with 30-minute retention windown, for every 3-minute. Periodic compactor continues to record latest revisions for every 1/10 of given compaction period (e.g. 1-hour when `--auto-compaction-mode=periodic --auto-compaction-retention=10h`). For every 1/10 of given compaction period, compactor uses the last revision that was fetched before compaction period, to discard historical data. The retention window of compaction period moves for every 1/10 of given compaction period. For instance, when hourly writes are 100 and `--auto-compaction-retention=10`, v3.1 compacts revision 1000, 2000, and 3000 for every 10-hour, while v3.2.x, v3.3.0, v3.3.1, and v3.3.2 compact revision 1000, 1100, and 1200 for every 1-hour. Futhermore, when writes per minute are 1000, v3.3.0, v3.3.1, and v3.3.2 with `--auto-compaction-mode=periodic --auto-compaction-retention=30m` compact revision 30000, 33000, and 36000, for every 3-minute with more finer granularity.
In [v3.3.0](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.3.md), [v3.3.1](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.3.md), and [v3.3.2](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.3.md), `--auto-compaction-mode=revision --auto-compaction-retention=1000` automatically `Compact` on `"latest revision" - 1000` every 5-minute (when latest revision is 30000, compact on revision 29000). For instance, `--auto-compaction-mode=periodic --auto-compaction-retention=72h` automatically `Compact` with 72-hour retention windown, for every 7.2-hour. For instance, `--auto-compaction-mode=periodic --auto-compaction-retention=30m` automatically `Compact` with 30-minute retention windown, for every 3-minute. Periodic compactor continues to record latest revisions for every 1/10 of given compaction period (e.g. 1-hour when `--auto-compaction-mode=periodic --auto-compaction-retention=10h`). For every 1/10 of given compaction period, compactor uses the last revision that was fetched before compaction period, to discard historical data. The retention window of compaction period moves for every 1/10 of given compaction period. For instance, when hourly writes are 100 and `--auto-compaction-retention=10`, v3.1 compacts revision 1000, 2000, and 3000 for every 10-hour, while v3.2.x, v3.3.0, v3.3.1, and v3.3.2 compact revision 1000, 1100, and 1200 for every 1-hour. Futhermore, when writes per minute are 1000, v3.3.0, v3.3.1, and v3.3.2 with `--auto-compaction-mode=periodic --auto-compaction-retention=30m` compact revision 30000, 33000, and 36000, for every 3-minute with more finer granularity.
When `--auto-compaction-retention=10h`, etcd first waits 10-hour for the first compaction, and then does compaction every hour (1/10 of 10-hour) afterwards like this:
@@ -68,7 +66,7 @@ When `--auto-compaction-retention=10h`, etcd first waits 10-hour for the first c
Whether compaction succeeds or not, this process repeats for every 1/10 of given compaction period. If compaction succeeds, it just removes compacted revision from historical revision records.
In [v3.3.3](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.3.md), `--auto-compaction-mode=revision --auto-compaction-retention=1000` automatically `Compact` on `"latest revision" - 1000` every 5-minute (when latest revision is 30000, compact on revision 29000). Previously, `--auto-compaction-mode=periodic --auto-compaction-retention=72h` automatically `Compact` with 72-hour retention windown for every 7.2-hour. **Now, `Compact` happens, for every 1-hour but still with 72-hour retention window.** Previously, `--auto-compaction-mode=periodic --auto-compaction-retention=30m` automatically `Compact` with 30-minute retention windown for every 3-minute. **Now, `Compact` happens, for every 30-minute but still with 30-minute retention window.** Periodic compactor keeps recording latest revisions for every compaction period when given period is less than 1-hour, or for every 1-hour when given compaction period is greater than 1-hour (e.g. 1-hour when `--auto-compaction-mode=periodic --auto-compaction-retention=24h`). For every compaction period or 1-hour, compactor uses the last revision that was fetched before compaction period, to discard historical data. The retention window of compaction period moves for every given compaction period or hour. For instance, when hourly writes are 100 and `--auto-compaction-mode=periodic --auto-compaction-retention=24h`, `v3.2.x`, `v3.3.0`, `v3.3.1`, and `v3.3.2` compact revision 2400, 2640, and 2880 for every 2.4-hour, while `v3.3.3`*or later* compacts revision 2400, 2500, 2600 for every 1-hour. Furthermore, when `--auto-compaction-mode=periodic --auto-compaction-retention=30m` and writes per minute are about 1000, `v3.3.0`, `v3.3.1`, and `v3.3.2` compact revision 30000, 33000, and 36000, for every 3-minute, while `v3.3.3`*or later* compacts revision 30000, 60000, and 90000, for every 30-minute.
In [v3.3.3](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.3.md), `--auto-compaction-mode=revision --auto-compaction-retention=1000` automatically `Compact` on `"latest revision" - 1000` every 5-minute (when latest revision is 30000, compact on revision 29000). Previously, `--auto-compaction-mode=periodic --auto-compaction-retention=72h` automatically `Compact` with 72-hour retention windown for every 7.2-hour. **Now, `Compact` happens, for every 1-hour but still with 72-hour retention window.** Previously, `--auto-compaction-mode=periodic --auto-compaction-retention=30m` automatically `Compact` with 30-minute retention windown for every 3-minute. **Now, `Compact` happens, for every 30-minute but still with 30-minute retention window.** Periodic compactor keeps recording latest revisions for every compaction period when given period is less than 1-hour, or for every 1-hour when given compaction period is greater than 1-hour (e.g. 1-hour when `--auto-compaction-mode=periodic --auto-compaction-retention=24h`). For every compaction period or 1-hour, compactor uses the last revision that was fetched before compaction period, to discard historical data. The retention window of compaction period moves for every given compaction period or hour. For instance, when hourly writes are 100 and `--auto-compaction-mode=periodic --auto-compaction-retention=24h`, `v3.2.x`, `v3.3.0`, `v3.3.1`, and `v3.3.2` compact revision 2400, 2640, and 2880 for every 2.4-hour, while `v3.3.3`*or later* compacts revision 2400, 2500, 2600 for every 1-hour. Futhermore, when `--auto-compaction-mode=periodic --auto-compaction-retention=30m` and writes per minute are about 1000, `v3.3.0`, `v3.3.1`, and `v3.3.2` compact revision 30000, 33000, and 36000, for every 3-minute, while `v3.3.3`*or later* compacts revision 30000, 60000, and 90000, for every 30-minute.
Each etcd server provides local monitoring information on its client port through http endpoints. The monitoring data is useful for both system health checking and cluster debugging.
@@ -123,6 +121,8 @@ Access: proxy
Then import the default [etcd dashboard template][template] and customize. For instance, if Prometheus data source name is `my-etcd`, the `datasource` field values in JSON also need to be `my-etcd`.
etcd is designed to withstand machine failures. An etcd cluster automatically recovers from temporary failures (e.g., machine reboots) and tolerates up to *(N-1)/2* permanent failures for a cluster of N members. When a member permanently fails, whether due to hardware failure or disk corruption, it loses access to the cluster. If the cluster permanently loses more than *(N-1)/2* members then it disastrously fails, irrevocably losing quorum. Once quorum is lost, the cluster cannot reach consensus and therefore cannot continue accepting updates.
@@ -66,6 +64,6 @@ Now the restored etcd cluster should be available and serving the keyspace given
## Restoring a cluster from membership mis-reconfiguration with wrong URLs
Previously, etcd panics on [membership mis-reconfiguration with wrong URLs](https://github.com/etcd-io/etcd/issues/9173) (v3.2.15 or later returns [error early in client-side](https://github.com/etcd-io/etcd/pull/9174) before etcd server panic).
Previously, etcd panics on [membership mis-reconfiguration with wrong URLs](https://github.com/coreos/etcd/issues/9173) (v3.2.15 or later returns [error early in client-side](https://github.com/coreos/etcd/pull/9174) before etcd server panic).
Recommended way is restore from [snapshot](#snapshotting-the-keyspace). `--force-new-cluster` can be used to overwrite cluster membership while keeping existing application data, but is strongly discouraged because it will panic if other members from previous cluster are still alive. Make sure to save snapshot periodically.
etcd supports automatic TLS as well as authentication through client certificates for both clients to server as well as peer (server to server / cluster) communication.**Note that etcd doesn't enable [RBAC based authentication][auth] or the authentication feature in the transport layer by default to reduce friction for users getting started with the database. Further, changing this default would be a breaking change for the project which was established since 2013. An etcd cluster which doesn't enable security features can expose its data to any clients.**
etcd supports automatic TLS as well as authentication through client certificates for both clients to server as well as peer (server to server / cluster) communication.
To get up and running, first have a CA certificate and a signed key pair for one member. It is recommended to create and sign a new key pair for every member in a cluster.
@@ -126,7 +124,7 @@ And also the response from the server:
}
```
Specify cipher suites to block [weak TLS cipher suites](https://github.com/etcd-io/etcd/issues/8320).
Specify cipher suites to block [weak TLS cipher suites](https://github.com/coreos/etcd/issues/8320).
TLS handshake would fail when client hello is requested with invalid cipher suites.
@@ -242,9 +240,9 @@ When client authentication is enabled for an etcd member, the administrator must
## Notes for TLS authentication
Since [v3.2.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v320-2017-06-09), [TLS certificates get reloaded on every client connection](https://github.com/etcd-io/etcd/pull/7829). This is useful when replacing expiry certs without stopping etcd servers; it can be done by overwriting old certs with new ones. Refreshing certs for every connection should not have too much overhead, but can be improved in the future, with caching layer. Example tests can be found [here](https://github.com/coreos/etcd/blob/b041ce5d514a4b4aaeefbffb008f0c7570a18986/integration/v3_grpc_test.go#L1601-L1757).
Since [v3.2.0](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.2.md#v320-2017-06-09), [TLS certificates get reloaded on every client connection](https://github.com/coreos/etcd/pull/7829). This is useful when replacing expiry certs without stopping etcd servers; it can be done by overwriting old certs with new ones. Refreshing certs for every connection should not have too much overhead, but can be improved in the future, with caching layer. Example tests can be found [here](https://github.com/coreos/etcd/blob/b041ce5d514a4b4aaeefbffb008f0c7570a18986/integration/v3_grpc_test.go#L1601-L1757).
Since [v3.2.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v320-2017-06-09), [server denies incoming peer certs with wrong IP `SAN`](https://github.com/etcd-io/etcd/pull/7687). For instance, if peer cert contains any IP addresses in Subject Alternative Name (SAN) field, server authenticates a peer only when the remote IP address matches one of those IP addresses. This is to prevent unauthorized endpoints from joining the cluster. For example, peer B's CSR (with `cfssl`) is:
Since [v3.2.0](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.2.md#v320-2017-06-09), [server denies incoming peer certs with wrong IP `SAN`](https://github.com/coreos/etcd/pull/7687). For instance, if peer cert contains any IP addresses in Subject Alternative Name (SAN) field, server authenticates a peer only when the remote IP address matches one of those IP addresses. This is to prevent unauthorized endpoints from joining the cluster. For example, peer B's CSR (with `cfssl`) is:
```json
{
@@ -270,7 +268,7 @@ Since [v3.2.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v320
when peer B's actual IP address is `10.138.0.2`, not `10.138.0.27`. When peer B tries to join the cluster, peer A will reject B with the error `x509: certificate is valid for 10.138.0.27, not 10.138.0.2`, because B's remote IP address does not match the one in Subject Alternative Name (SAN) field.
Since [v3.2.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v320-2017-06-09), [server resolves TLS `DNSNames` when checking `SAN`](https://github.com/etcd-io/etcd/pull/7767). For instance, if peer cert contains only DNS names (no IP addresses) in Subject Alternative Name (SAN) field, server authenticates a peer only when forward-lookups (`dig b.com`) on those DNS names have matching IP with the remote IP address. For example, peer B's CSR (with `cfssl`) is:
Since [v3.2.0](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.2.md#v320-2017-06-09), [server resolves TLS `DNSNames` when checking `SAN`](https://github.com/coreos/etcd/pull/7767). For instance, if peer cert contains only DNS names (no IP addresses) in Subject Alternative Name (SAN) field, server authenticates a peer only when forward-lookups (`dig b.com`) on those DNS names have matching IP with the remote IP address. For example, peer B's CSR (with `cfssl`) is:
```json
{
@@ -282,7 +280,7 @@ Since [v3.2.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v320
when peer B's remote IP address is `10.138.0.2`. When peer B tries to join the cluster, peer A looks up the incoming host `b.com` to get the list of IP addresses (e.g. `dig b.com`). And rejects B if the list does not contain the IP `10.138.0.2`, with the error `tls: 10.138.0.2 does not match any of DNSNames ["b.com"]`.
Since [v3.2.2](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v322-2017-07-07), [server accepts connections if IP matches, without checking DNS entries](https://github.com/etcd-io/etcd/pull/8223). For instance, if peer cert contains IP addresses and DNS names in Subject Alternative Name (SAN) field, and the remote IP address matches one of those IP addresses, server just accepts connection without further checking the DNS names. For example, peer B's CSR (with `cfssl`) is:
Since [v3.2.2](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.2.md#v322-2017-07-07), [server accepts connections if IP matches, without checking DNS entries](https://github.com/coreos/etcd/pull/8223). For instance, if peer cert contains IP addresses and DNS names in Subject Alternative Name (SAN) field, and the remote IP address matches one of those IP addresses, server just accepts connection without further checking the DNS names. For example, peer B's CSR (with `cfssl`) is:
```json
{
@@ -293,9 +291,9 @@ Since [v3.2.2](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v322
],
```
when peer B's remote IP address is `10.138.0.2` and `invalid.domain` is a invalid host. When peer B tries to join the cluster, peer A successfully authenticates B, since Subject Alternative Name (SAN) field has a valid matching IP address. See [issue#8206](https://github.com/etcd-io/etcd/issues/8206) for more detail.
when peer B's remote IP address is `10.138.0.2` and `invalid.domain` is a invalid host. When peer B tries to join the cluster, peer A successfully authenticates B, since Subject Alternative Name (SAN) field has a valid matching IP address. See [issue#8206](https://github.com/coreos/etcd/issues/8206) for more detail.
Since [v3.2.5](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v325-2017-08-04), [server supports reverse-lookup on wildcard DNS `SAN`](https://github.com/etcd-io/etcd/pull/8281). For instance, if peer cert contains only DNS names (no IP addresses) in Subject Alternative Name (SAN) field, server first reverse-lookups the remote IP address to get a list of names mapping to that address (e.g. `nslookup IPADDR`). Then accepts the connection if those names have a matching name with peer cert's DNS names (either by exact or wildcard match). If none is matched, server forward-lookups each DNS entry in peer cert (e.g. look up `example.default.svc` when the entry is `*.example.default.svc`), and accepts connection only when the host's resolved addresses have the matching IP address with the peer's remote IP address. For example, peer B's CSR (with `cfssl`) is:
Since [v3.2.5](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.2.md#v325-2017-08-04), [server supports reverse-lookup on wildcard DNS `SAN`](https://github.com/coreos/etcd/pull/8281). For instance, if peer cert contains only DNS names (no IP addresses) in Subject Alternative Name (SAN) field, server first reverse-lookups the remote IP address to get a list of names mapping to that address (e.g. `nslookup IPADDR`). Then accepts the connection if those names have a matching name with peer cert's DNS names (either by exact or wildcard match). If none is matched, server forward-lookups each DNS entry in peer cert (e.g. look up `example.default.svc` when the entry is `*.example.default.svc`), and accepts connection only when the host's resolved addresses have the matching IP address with the peer's remote IP address. For example, peer B's CSR (with `cfssl`) is:
```json
{
@@ -306,9 +304,9 @@ Since [v3.2.5](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md#v325
],
```
when peer B's remote IP address is `10.138.0.2`. When peer B tries to join the cluster, peer A reverse-lookup the IP `10.138.0.2` to get the list of host names. And either exact or wildcard match the host names with peer B's cert DNS names in Subject Alternative Name (SAN) field. If none of reverse/forward lookups worked, it returns an error `"tls: "10.138.0.2" does not match any of DNSNames ["*.example.default.svc","*.example.default.svc.cluster.local"]`. See [issue#8268](https://github.com/etcd-io/etcd/issues/8268) for more detail.
when peer B's remote IP address is `10.138.0.2`. When peer B tries to join the cluster, peer A reverse-lookup the IP `10.138.0.2` to get the list of host names. And either exact or wildcard match the host names with peer B's cert DNS names in Subject Alternative Name (SAN) field. If none of reverse/forward lookups worked, it returns an error `"tls: "10.138.0.2" does not match any of DNSNames ["*.example.default.svc","*.example.default.svc.cluster.local"]`. See [issue#8268](https://github.com/coreos/etcd/issues/8268) for more detail.
[v3.3.0](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.3.md) adds [`etcd --peer-cert-allowed-cn`](https://github.com/etcd-io/etcd/pull/8616) flag to support [CN(Common Name)-based auth for inter-peer connections](https://github.com/etcd-io/etcd/issues/8262). Kubernetes TLS bootstrapping involves generating dynamic certificates for etcd members and other system components (e.g. API server, kubelet, etc.). Maintaining different CAs for each component provides tighter access control to etcd cluster but often tedious. When `--peer-cert-allowed-cn` flag is specified, node can only join with matching common name even with shared CAs. For example, each member in 3-node cluster is set up with CSRs (with `cfssl`) as below:
[v3.3.0](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.3.md) adds [`etcd --peer-cert-allowed-cn`](https://github.com/coreos/etcd/pull/8616) flag to support [CN(Common Name)-based auth for inter-peer connections](https://github.com/coreos/etcd/issues/8262). Kubernetes TLS bootstrapping involves generating dynamic certificates for etcd members and other system components (e.g. API server, kubelet, etc.). Maintaining different CAs for each component provides tighter access control to etcd cluster but often tedious. When `--peer-cert-allowed-cn` flag is specified, node can only join with matching common name even with shared CAs. For example, each member in 3-node cluster is set up with CSRs (with `cfssl`) as below:
```json
{
@@ -368,7 +366,7 @@ I | embed: serving client requests on 127.0.0.1:22379
I | embed: serving client requests on 127.0.0.1:2379
```
[v3.2.19](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.2.md) and [v3.3.4](https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.3.md) fixes TLS reload when [certificate SAN field only includes IP addresses but no domain names](https://github.com/etcd-io/etcd/issues/9541). For example, a member is set up with CSRs (with `cfssl`) as below:
[v3.2.19](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.2.md) and [v3.3.4](https://github.com/coreos/etcd/blob/master/CHANGELOG-3.3.md) fixes TLS reload when [certificate SAN field only includes IP addresses but no domain names](https://github.com/coreos/etcd/issues/9541). For example, a member is set up with CSRs (with `cfssl`) as below:
```json
{
@@ -380,7 +378,7 @@ I | embed: serving client requests on 127.0.0.1:2379
In Go, server calls `(*tls.Config).GetCertificate` for TLS reload if and only if server's `(*tls.Config).Certificates` field is not empty, or `(*tls.ClientHelloInfo).ServerName` is not empty with a valid SNI from the client. Previously, etcd always populates `(*tls.Config).Certificates` on the initial client TLS handshake, as non-empty. Thus, client was always expected to supply a matching SNI in order to pass the TLS verification and to trigger `(*tls.Config).GetCertificate` to reload TLS assets.
However, a certificate whose SAN field does [not include any domain names but only IP addresses](https://github.com/etcd-io/etcd/issues/9541) would request `*tls.ClientHelloInfo` with an empty `ServerName` field, thus failing to trigger the TLS reload on initial TLS handshake; this becomes a problem when expired certificates need to be replaced online.
However, a certificate whose SAN field does [not include any domain names but only IP addresses](https://github.com/coreos/etcd/issues/9541) would request `*tls.ClientHelloInfo` with an empty `ServerName` field, thus failing to trigger the TLS reload on initial TLS handshake; this becomes a problem when expired certificates need to be replaced online.
Now, `(*tls.Config).Certificates` is created empty on initial TLS client handshake, first to trigger `(*tls.Config).GetCertificate`, and then to populate rest of the certificates on every new TLS connection, even when client SNI is empty (e.g. cert only includes IPs).
@@ -426,14 +424,8 @@ Make sure to sign the certificates with a Subject Name the member's public IP ad
The certificate needs to be signed for the member's FQDN in its Subject Name, use Subject Alternative Names (short IP SANs) to add the IP address. The `etcd-ca` tool provides `--domain=` option for its `new-cert` command, and openssl can make [it][alt-name] too.
### Does etcd encrypt data stored on disk drives?
No. etcd doesn't encrypt key/value data stored on disk drives. If a user need to encrypt data stored on etcd, there are some options:
* Let client applications encrypt and decrypt the data
* Use a feature of underlying storage systems for encrypting stored data like [dm-crypt]
* etcd-maintainers are listed in https://github.com/etcd-io/etcd/blob/master/MAINTAINERS.
* etcd-maintainers are listed in https://github.com/coreos/etcd/blob/master/MAINTAINERS.
Experimental platforms appear to work in practice and have some platform specific code in etcd, but do not fully conform to the stable support policy. Unstable platforms have been lightly tested, but less than experimental. Unlisted architecture and operating system pairs are currently unsupported; caveat emptor.
title: Migrate applications from using API v2 to API v3
---
# Migrate applications from using API v2 to API v3
The data store v2 is still accessible from the API v2 after upgrading to etcd3. Thus, it will work as before and require no application changes. With etcd 3, applications use the new grpc API v3 to access the mvcc store, which provides more features and improved performance. The mvcc store and the old store v2 are separate and isolated; writes to the store v2 will not affect the mvcc store and, similarly, writes to the mvcc store will not affect the store v2.
This guide assumes operational knowledge of Amazon Web Services (AWS), specifically Amazon Elastic Compute Cloud (EC2). This guide provides an introduction to design considerations when designing an etcd deployment on AWS EC2 and how AWS specific features may be utilized in that context.
Startingwithversion0.1.2bothetcdandetcdctlhavebeenportedtoFreeBSDandcan beinstalledeitherviapackagesorportssystem.Theirversionshavebeenrecently updatedto0.2.0sonowetcd and etcdctl can be enjoyedonFreeBSD10.0(RC4as ofnow)and9.x,wheretheyhavebeentested.Theymightalsoworkwheninstalledfrom portsonearlierversionsofFreeBSD,butit is untested; caveat emptor.
This document tracks people and use cases for etcd in production. By creating a list of production use cases we hope to build a community of advisors that we can reach out to with experience using various etcd applications, operation environments, and cluster sizes. The etcd development team may reach out periodically to check-in on how etcd is working in the field and update this list.
@@ -239,12 +237,3 @@ At [Branch][branch], we use kubernetes heavily as our core microservice platform
- *Environment*: Bare Metal
- *Backups*: None, all data is considered ephemeral.
## Transwarp
- *Application*: Transwarp Data Cloud, Transwarp Operating System, Transwarp Data Hub, Sophon
If any part of the etcd project has bugs or documentation mistakes, please let us know by [opening an issue][etcd-issue]. We treat bugs and mistakes very seriously and believe no issue is too small. Before creating a bug report, please check that an issue reporting the same problem does not already exist.
@@ -43,5 +41,5 @@ $ sudo journalctl -u etcd2
Due to an upstream systemd bug, journald may miss the last few log lines when its processes exit. If journalctl says etcd stopped without fatal or panic message, try `sudo journalctl -f -t etcd2` to get full log.
The etcd v3 API is designed to give users a more efficient and cleaner abstraction compared to etcd v2. There are a number of semantic and protocol changes in this new API. For an overview [see Xiang Li's video](https://youtu.be/J5AioGtEPeQ?t=211).
To prove out the design of the v3 API the team has also built [a number of example recipes](https://github.com/coreos/etcd/tree/master/contrib/recipes), there is a [video discussing these recipes too](https://www.youtube.com/watch?v=fj-2RY-3yVU&feature=youtu.be&t=590).
# Design
1. Flatten binary key-value space
2. Keep the event history until compaction
- access to old version of keys
- user controlled history compaction
3. Support range query
- Pagination support with limit argument
- Support consistency guarantee across multiple range queries
4. Replace TTL key with Lease
- more efficient/ low cost keep alive
- a logical group of TTL keys
5. Replace CAS/CAD with multi-object Txn
- MUCH MORE powerful and flexible
6. Support efficient watching with multiple ranges
7. RPC API supports the completed set of APIs.
- more efficient than JSON/HTTP
- additional txn/lease support
8. HTTP API supports a subset of APIs.
- easy for people to try out etcd
- easy for people to write simple etcd application
## Notes
### Request Size Limitation
The max request size is around 1MB. Since etcd replicates requests in a streaming fashion, a very large
request might block other requests for a long time. The use case for etcd is to store small configuration
values, so we prevent user from submitting large requests. This also applies to Txn requests. We might loosen
the size in the future a little bit or make it configurable.
## Protobuf Defined API
[api protobuf][api-protobuf]
[kv protobuf][kv-protobuf]
## Examples
### Put a key (foo=bar)
```
// A put is always successful
Put( PutRequest { key = foo, value = bar } )
PutResponse {
cluster_id = 0x1000,
member_id = 0x1,
revision = 1,
raft_term = 0x1,
}
```
### Get a key (assume we have foo=bar)
```
Get ( RangeRequest { key = foo } )
RangeResponse {
cluster_id = 0x1000,
member_id = 0x1,
revision = 1,
raft_term = 0x1,
kvs = {
{
key = foo,
value = bar,
create_revision = 1,
mod_revision = 1,
version = 1;
},
},
}
```
### Range over a key space (assume we have foo0=bar0… foo100=bar100)
The etcd v3 API is designed to give users a more efficient and cleaner abstraction compared to etcd v2. There are a number of semantic and protocol changes in this new API. For an overview [see Xiang Li's video](https://youtu.be/J5AioGtEPeQ?t=211).
To prove out the design of the v3 API the team has also built [a number of example recipes](https://github.com/coreos/etcd/tree/master/contrib/recipes), there is a [video discussing these recipes too](https://www.youtube.com/watch?v=fj-2RY-3yVU&feature=youtu.be&t=590).
# Design
1. Flatten binary key-value space
2. Keep the event history until compaction
- access to old version of keys
- user controlled history compaction
3. Support range query
- Pagination support with limit argument
- Support consistency guarantee across multiple range queries
4. Replace TTL key with Lease
- more efficient/ low cost keep alive
- a logical group of TTL keys
5. Replace CAS/CAD with multi-object Txn
- MUCH MORE powerful and flexible
6. Support efficient watching with multiple ranges
7. RPC API supports the completed set of APIs.
- more efficient than JSON/HTTP
- additional txn/lease support
8. HTTP API supports a subset of APIs.
- easy for people to try out etcd
- easy for people to write simple etcd application
## Notes
### Request Size Limitation
The max request size is around 1MB. Since etcd replicates requests in a streaming fashion, a very large
request might block other requests for a long time. The use case for etcd is to store small configuration
values, so we prevent user from submitting large requests. This also applies to Txn requests. We might loosen
the size in the future a little bit or make it configurable.
## Protobuf Defined API
[api protobuf][api-protobuf]
[kv protobuf][kv-protobuf]
## Examples
### Put a key (foo=bar)
```
// A put is always successful
Put( PutRequest { key = foo, value = bar } )
PutResponse {
cluster_id = 0x1000,
member_id = 0x1,
revision = 1,
raft_term = 0x1,
}
```
### Get a key (assume we have foo=bar)
```
Get ( RangeRequest { key = foo } )
RangeResponse {
cluster_id = 0x1000,
member_id = 0x1,
revision = 1,
raft_term = 0x1,
kvs = {
{
key = foo,
value = bar,
create_revision = 1,
mod_revision = 1,
version = 1;
},
},
}
```
### Range over a key space (assume we have foo0=bar0… foo100=bar100)
The default settings in etcd should work well for installations on a local network where the average network latency is low. However, when using etcd across multiple data centers or over networks with high latency, the heartbeat interval and election timeout settings may need tuning.
@@ -73,7 +71,7 @@ dropped MsgAppResp to 247ae21ff9436b2d since streamMsg's sending buffer is full
These errors may be resolved by prioritizing etcd's peer traffic over its client traffic. On Linux, peer traffic can be prioritized by using the traffic control mechanism:
```sh
```
tc qdisc add dev eth0 root handle 1: prio bands 3
tc filter add dev eth0 parent 1: protocol ip prio 1 u32 match ip sport 2380 0xffff flowid 1:1
tc filter add dev eth0 parent 1: protocol ip prio 1 u32 match ip dport 2380 0xffff flowid 1:1
In the general case, upgrading from etcd 2.3 to 3.0 can be a zero-downtime, rolling upgrade:
- one by one, stop the etcd v2.3 processes and replace them with etcd v3.0 processes
@@ -10,7 +8,7 @@ Before [starting an upgrade](#upgrade-procedure), read through the rest of this
### Upgrade checklists
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/etcd-io/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/coreos/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
#### Upgrade requirements
@@ -126,8 +124,8 @@ $ ETCDCTL_API=3 etcdctl endpoint health
## Known Issues
- etcd < v3.1 does not work properly if built with Go > v1.7. See [Issue 6951](https://github.com/etcd-io/etcd/issues/6951) for additional information.
- etcd < v3.1 does not work properly if built with Go > v1.7. See [Issue 6951](https://github.com/coreos/etcd/issues/6951) for additional information.
- If an error such as `transport: http2Client.notifyError got notified that the client transport was broken unexpected EOF.` shows up in the etcd server logs, be sure etcd is a pre-built release or built with (etcd v3.1+ & go v1.7+) or (etcd <v3.1 & go v1.6.x).
- Adding a v3 node to v2.3 cluster during upgrades is not supported and could trigger panics. See [Issue 7249](https://github.com/etcd-io/etcd/issues/7429) for additional information. Mixed versions of etcd members are only allowed during v3 migration. Finish upgrades before making any membership changes.
- Adding a v3 node to v2.3 cluster during upgrades is not supported and could trigger panics. See [Issue 7249](https://github.com/coreos/etcd/issues/7429) for additional information. Mixed versions of etcd members are only allowed during v3 migration. Finish upgrades before making any membership changes.
In the general case, upgrading from etcd 3.0 to 3.1 can be a zero-downtime, rolling upgrade:
- one by one, stop the etcd v3.0 processes and replace them with etcd v3.1 processes
@@ -10,7 +8,7 @@ Before [starting an upgrade](#upgrade-procedure), read through the rest of this
### Upgrade checklists
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/etcd-io/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/coreos/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
In the general case, upgrading from etcd 3.1 to 3.2 can be a zero-downtime, rolling upgrade:
- one by one, stop the etcd v3.1 processes and replace them with etcd v3.2 processes
@@ -10,7 +8,7 @@ Before [starting an upgrade](#upgrade-procedure), read through the rest of this
### Upgrade checklists
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/etcd-io/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/coreos/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
Highlighted breaking changes in 3.2.
@@ -18,7 +16,7 @@ Highlighted breaking changes in 3.2.
Higher `--snapshot-count` holds more Raft entries in memory until snapshot, thus causing [recurrent higher memory usage](https://github.com/kubernetes/kubernetes/issues/60589#issuecomment-371977156). Since leader retains latest Raft entries for longer, a slow follower has more time to catch up before leader snapshot. `--snapshot-count` is a tradeoff between higher memory usage and better availabilities of slow followers.
Since v3.2, the default value of `--snapshot-count` has [changed from from 10,000 to 100,000](https://github.com/etcd-io/etcd/pull/7160).
Since v3.2, the default value of `--snapshot-count` has [changed from from 10,000 to 100,000](https://github.com/coreos/etcd/pull/7160).
Previously, `grpc.ErrClientConnTimeout` error is returned on client dial time-outs. 3.2 instead returns `context.DeadlineExceeded` (see [#8504](https://github.com/etcd-io/etcd/issues/8504)).
Previously, `grpc.ErrClientConnTimeout` error is returned on client dial time-outs. 3.2 instead returns `context.DeadlineExceeded` (see [#8504](https://github.com/coreos/etcd/issues/8504)).
3.2.12 or later changes the function signatures of `clientv3` gRPC client wrapper. This change was needed to support [custom `grpc.CallOption` on message size limits](https://github.com/etcd-io/etcd/pull/9047).
3.2.12 or later changes the function signatures of `clientv3` gRPC client wrapper. This change was needed to support [custom `grpc.CallOption` on message size limits](https://github.com/coreos/etcd/pull/9047).
Before and after
@@ -172,7 +170,7 @@ Before and after
#### Changed `clientv3.Lease.TimeToLive` API
Previously, `clientv3.Lease.TimeToLive` API returned `lease.ErrLeaseNotFound` on non-existent lease ID. 3.2 instead returns TTL=-1 in its response and no error (see [#7305](https://github.com/etcd-io/etcd/pull/7305)).
Previously, `clientv3.Lease.TimeToLive` API returned `lease.ErrLeaseNotFound` on non-existent lease ID. 3.2 instead returns TTL=-1 in its response and no error (see [#7305](https://github.com/coreos/etcd/pull/7305)).
Before
@@ -214,7 +212,7 @@ clientv3yaml.NewConfig
3.2 now rejects domains names for `--listen-peer-urls` and `--listen-client-urls` (3.1 only prints out warnings), since domain name is invalid for network interface binding. Make sure that those URLs are properly formated as `scheme://IP:port`.
See [issue #6336](https://github.com/etcd-io/etcd/issues/6336) for more contexts.
See [issue #6336](https://github.com/coreos/etcd/issues/6336) for more contexts.
In the general case, upgrading from etcd 3.2 to 3.3 can be a zero-downtime, rolling upgrade:
- one by one, stop the etcd v3.2 processes and replace them with etcd v3.3 processes
@@ -10,13 +8,13 @@ Before [starting an upgrade](#upgrade-procedure), read through the rest of this
### Upgrade checklists
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/etcd-io/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/coreos/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
Highlighted breaking changes in 3.3.
#### Changed value type of `etcd --auto-compaction-retention` flag to `string`
Changed `--auto-compaction-retention` flag to [accept string values](https://github.com/etcd-io/etcd/pull/8563) with [finer granularity](https://github.com/etcd-io/etcd/issues/8503). Now that `--auto-compaction-retention` accepts string values, etcd configuration YAML file `auto-compaction-retention` field must be changed to `string` type. Previously, `--config-file etcd.config.yaml` can have `auto-compaction-retention: 24` field, now must be `auto-compaction-retention: "24"` or `auto-compaction-retention: "24h"`. If configured as `--auto-compaction-mode periodic --auto-compaction-retention "24h"`, the time duration value for `--auto-compaction-retention` flag must be valid for [`time.ParseDuration`](https://golang.org/pkg/time/#ParseDuration) function in Go.
Changed `--auto-compaction-retention` flag to [accept string values](https://github.com/coreos/etcd/pull/8563) with [finer granularity](https://github.com/coreos/etcd/issues/8503). Now that `--auto-compaction-retention` accepts string values, etcd configuration YAML file `auto-compaction-retention` field must be changed to `string` type. Previously, `--config-file etcd.config.yaml` can have `auto-compaction-retention: 24` field, now must be `auto-compaction-retention: "24"` or `auto-compaction-retention: "24h"`. If configured as `--auto-compaction-mode periodic --auto-compaction-retention "24h"`, the time duration value for `--auto-compaction-retention` flag must be valid for [`time.ParseDuration`](https://golang.org/pkg/time/#ParseDuration) function in Go.
```diff
# etcd.config.yaml
@@ -59,7 +57,7 @@ func (e *EtcdServer) Start() error {
#### Added `embed.Config.LogOutput` struct
**Note that this field has been renamed to `embed.Config.LogOutputs` in `[]string` type in v3.4. Please see [v3.4 upgrade guide](https://github.com/etcd-io/etcd/blob/master/Documentation/upgrades/upgrade_3_4.md) for more details.**
**Note that this field has been renamed to `embed.Config.LogOutputs` in `[]string` type in v3.4. Please see [v3.4 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_4.md) for more details.**
From v3.3, gRPC server logs are disabled by default.
**Note that `embed.Config.SetupLogging` method has been deprecated in v3.4. Please see [v3.4 upgrade guide](https://github.com/etcd-io/etcd/blob/master/Documentation/upgrades/upgrade_3_4.md) for more details.**
**Note that `embed.Config.SetupLogging` method has been deprecated in v3.4. Please see [v3.4 upgrade guide](https://github.com/coreos/etcd/blob/master/Documentation/upgrades/upgrade_3_4.md) for more details.**
#### Changed raw gRPC client wrapper function signatures
3.3 changes the function signatures of `clientv3` gRPC client wrapper. This change was needed to support [custom `grpc.CallOption` on message size limits](https://github.com/etcd-io/etcd/pull/9047).
3.3 changes the function signatures of `clientv3` gRPC client wrapper. This change was needed to support [custom `grpc.CallOption` on message size limits](https://github.com/coreos/etcd/pull/9047).
Previously, `grpc.ErrClientConnTimeout` error is returned on client dial time-outs. 3.3 instead returns `context.DeadlineExceeded` (see [#8504](https://github.com/etcd-io/etcd/issues/8504)).
Previously, `grpc.ErrClientConnTimeout` error is returned on client dial time-outs. 3.3 instead returns `context.DeadlineExceeded` (see [#8504](https://github.com/coreos/etcd/issues/8504)).
Before
@@ -369,52 +367,6 @@ After
docker pull gcr.io/etcd-development/etcd:v3.3.0
```
### Upgrades to >= v3.3.14
[v3.3.14](https://github.com/etcd-io/etcd/releases/tag/v3.3.14) had to include some features from 3.4, while trying to minimize the difference between client balancer implementation. This release fixes ["kube-apiserver 1.13.x refuses to work when first etcd-server is not available" (kubernetes#72102)](https://github.com/kubernetes/kubernetes/issues/72102).
`grpc.ErrClientConnClosing` has been [deprecated in gRPC >= 1.10](https://github.com/grpc/grpc-go/pull/1854).
```diff
import (
+ "go.etcd.io/etcd/clientv3"
"google.golang.org/grpc"
+ "google.golang.org/grpc/codes"
+ "google.golang.org/grpc/status"
)
_, err := kvc.Get(ctx, "a")
-if err == grpc.ErrClientConnClosing {
+if clientv3.IsConnCanceled(err) {
// or
+s, ok := status.FromError(err)
+if ok {
+ if s.Code() == codes.Canceled
```
[The new client balancer](https://github.com/etcd-io/etcd/blob/master/Documentation/learning/design-client.md) uses an asynchronous resolver to pass endpoints to the gRPC dial function. As a result, [v3.3.14](https://github.com/etcd-io/etcd/releases/tag/v3.3.14) or later requires `grpc.WithBlock` dial option to wait until the underlying connection is up.
```diff
import (
"time"
"go.etcd.io/etcd/clientv3"
+ "google.golang.org/grpc"
)
+// "grpc.WithBlock()" to block until the underlying connection is up
In the general case, upgrading from etcd 3.3 to 3.4 can be a zero-downtime, rolling upgrade:
- one by one, stop the etcd v3.3 processes and replace them with etcd v3.4 processes
- after running all v3.4 processes, new features in v3.4 are available to the cluster
Before [starting an upgrade](#upgrade-procedure), read through the rest of this guide to prepare.
### Upgrade checklists
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/coreos/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
Highlighted breaking changes in 3.4.
#### Make `ETCDCTL_API=3 etcdctl` default
`ETCDCTL_API=3` is now the default.
```diff
etcdctl set foo bar
Error: unknown command "set" for "etcdctl"
-etcdctl set foo bar
+ETCDCTL_API=2 etcdctl set foo bar
bar
ETCDCTL_API=3 etcdctl put foo bar
OK
-ETCDCTL_API=3 etcdctl put foo bar
+etcdctl put foo bar
```
#### Deprecated `etcd --ca-file` and `etcd --peer-ca-file` flags
`--ca-file` and `--peer-ca-file` flags are deprecated; they have been deprecated since v2.1.
v3.4 promotes `etcd_debugging_mvcc_db_total_size_in_bytes` Prometheus metrics to `etcd_mvcc_db_total_size_in_bytes`, in order to encourage etcd storage monitoring.
`etcd_debugging_mvcc_db_total_size_in_bytes` is still served in v3.4 for backward compatibilities. It will be completely deprecated in v3.5.
```diff
-etcd_debugging_mvcc_db_total_size_in_bytes
+etcd_mvcc_db_total_size_in_bytes
```
Note that `etcd_debugging_*` namespace metrics have been marked as experimental. As we improve monitoring guide, we will promote more metrics.
#### Deprecating `etcd --log-output` flag (now `--log-outputs`)
Rename [`etcd --log-output` to `--log-outputs`](https://github.com/coreos/etcd/pull/9624) to support multiple log outputs. **`etcd --logger=capnslog` does not support multiple log outputs.**
**`etcd --log-output`** will be deprecated in v3.5. **`etcd --logger=capnslog` will be deprecated in v3.5**.
```diff
-etcd --log-output=stderr
+etcd --log-outputs=stderr
+# to write logs to stderr and a.log file at the same time
+# only "--logger=zap" supports multiple writers
+etcd --logger=zap --log-outputs=stderr,a.log
```
v3.4 adds `etcd --logger=zap --log-outputs=stderr` support for structured logging and multiple log outputs. Main motivation is to promote automated etcd monitoring, rather than looking back server logs when it starts breaking. Future development will make etcd log as few as possible, and make etcd easier to monitor with metrics and alerts. **`etcd --logger=capnslog` will be deprecated in v3.5**.
#### Changed `log-outputs` field type in `etcd --config-file` to `[]string`
Now that `log-outputs` (old field name `log-output`) accepts multiple writers, etcd configuration YAML file `log-outputs` field must be changed to `[]string` type as below:
```diff
# Specify 'stdout' or 'stderr' to skip journald logging even when running under systemd.
-log-output: default
+log-outputs: [default]
```
#### Renamed `embed.Config.LogOutput` to `embed.Config.LogOutputs`
Renamed [**`embed.Config.LogOutput`** to **`embed.Config.LogOutputs`**](https://github.com/coreos/etcd/pull/9624) to support multiple log outputs. And changed [`embed.Config.LogOutput` type from `string` to `[]string`](https://github.com/coreos/etcd/pull/9579) to support multiple log outputs.
```diff
import "github.com/coreos/etcd/embed"
cfg := &embed.Config{Debug: false}
-cfg.LogOutput = "stderr"
+cfg.LogOutputs = []string{"stderr"}
```
#### v3.5 deprecates `capnslog`
**v3.5 will deprecate `etcd --log-package-levels` flag for `capnslog`**; `etcd --logger=zap --log-outputs=stderr` will the default. **v3.5 will deprecate `[CLIENT-URL]/config/local/log` endpoint.**
#### Deprecated `pkg/transport.TLSInfo.CAFile` field
#### Changed `embed.Config.SnapCount` to `embed.Config.SnapshotCount`
To be consistent with the flag name `etcd --snapshot-count`, `embed.Config.SnapCount` field has been renamed to `embed.Config.SnapshotCount`:
```diff
import "github.com/coreos/etcd/embed"
cfg := embed.NewConfig()
-cfg.SnapCount = 100000
+cfg.SnapshotCount = 100000
```
#### Changed `etcdserver.ServerConfig.SnapCount` to `etcdserver.ServerConfig.SnapshotCount`
To be consistent with the flag name `etcd --snapshot-count`, `etcdserver.ServerConfig.SnapCount` field has been renamed to `etcdserver.ServerConfig.SnapshotCount`:
```diff
import "github.com/coreos/etcd/etcdserver"
srvcfg := etcdserver.ServerConfig{
- SnapCount: 100000,
+ SnapshotCount: 100000,
```
#### Changed function signature in package `wal`
Changed `wal` function signatures to support structured logger.
```diff
import "github.com/coreos/etcd/wal"
+import "go.uber.org/zap"
+lg, _ = zap.NewProduction()
-wal.Open(dirpath, snap)
+wal.Open(lg, dirpath, snap)
-wal.OpenForRead(dirpath, snap)
+wal.OpenForRead(lg, dirpath, snap)
-wal.Repair(dirpath)
+wal.Repair(lg, dirpath)
-wal.Create(dirpath, metadata)
+wal.Create(lg, dirpath, metadata)
```
#### Deprecated `embed.Config.SetupLogging`
`embed.Config.SetupLogging` has been removed in order to prevent wrong logging configuration, and now set up automatically.
```diff
import "github.com/coreos/etcd/embed"
cfg := &embed.Config{Debug: false}
-cfg.SetupLogging()
```
#### Changed gRPC gateway HTTP endpoints (replaced `/v3beta` with `/v3`)
Before
```bash
curl -L http://localhost:2379/v3beta/kv/put \
-X POST -d '{"key": "Zm9v", "value": "YmFy"}'
```
After
```bash
curl -L http://localhost:2379/v3/kv/put \
-X POST -d '{"key": "Zm9v", "value": "YmFy"}'
```
Requests to `/v3beta` endpoints will redirect to `/v3`, and `/v3beta` will be removed in 3.5 release.
#### Deprecated container image tags
`latest` and minor version images tags are deprecated:
```diff
-docker pull gcr.io/etcd-development/etcd:latest
+docker pull gcr.io/etcd-development/etcd:v3.4.0
-docker pull gcr.io/etcd-development/etcd:v3.4
+docker pull gcr.io/etcd-development/etcd:v3.4.0
-docker pull gcr.io/etcd-development/etcd:v3.4
+docker pull gcr.io/etcd-development/etcd:v3.4.1
-docker pull gcr.io/etcd-development/etcd:v3.4
+docker pull gcr.io/etcd-development/etcd:v3.4.2
```
### Server upgrade checklists
#### Upgrade requirements
To upgrade an existing etcd deployment to 3.4, the running cluster must be 3.3 or greater. If it's before 3.3, please [upgrade to 3.3](upgrade_3_3.md) before upgrading to 3.4.
Also, to ensure a smooth rolling upgrade, the running cluster must be healthy. Check the health of the cluster by using the `etcdctl endpoint health` command before proceeding.
#### Preparation
Before upgrading etcd, always test the services relying on etcd in a staging environment before deploying the upgrade to the production environment.
Before beginning, [download the snapshot backup](../op-guide/maintenance.md#snapshot-backup). Should something go wrong with the upgrade, it is possible to use this backup to [downgrade](#downgrade) back to existing etcd version. Please note that the `snapshot` command only backs up the v3 data. For v2 data, see [backing up v2 datastore](../v2/admin_guide.md#backing-up-the-datastore).
#### Mixed versions
While upgrading, an etcd cluster supports mixed versions of etcd members, and operates with the protocol of the lowest common version. The cluster is only considered upgraded once all of its members are upgraded to version 3.4. Internally, etcd members negotiate with each other to determine the overall cluster version, which controls the reported version and the supported features.
#### Limitations
Note: If the cluster only has v3 data and no v2 data, it is not subject to this limitation.
If the cluster is serving a v2 data set larger than 50MB, each newly upgraded member may take up to two minutes to catch up with the existing cluster. Check the size of a recent snapshot to estimate the total data size. In other words, it is safest to wait for 2 minutes between upgrading each member.
For a much larger total data size, 100MB or more , this one-time process might take even more time. Administrators of very large etcd clusters of this magnitude can feel free to contact the [etcd team][etcd-contact] before upgrading, and we'll be happy to provide advice on the procedure.
#### Downgrade
If all members have been upgraded to v3.4, the cluster will be upgraded to v3.4, and downgrade from this completed state is **not possible**. If any single member is still v3.3, however, the cluster and its operations remains "v3.3", and it is possible from this mixed cluster state to return to using a v3.3 etcd binary on all members.
Please [download the snapshot backup](../op-guide/maintenance.md#snapshot-backup) to make downgrading the cluster possible even after it has been completely upgraded.
### Upgrade procedure
This example shows how to upgrade a 3-member v3.3 ectd cluster running on a local machine.
#### Step 1: check upgrade requirements
Is the cluster healthy and running v3.3.x?
```bash
etcdctl --endpoints=localhost:2379,localhost:22379,localhost:32379 endpoint health
<<COMMENT
localhost:2379 is healthy: successfully committed proposal: took = 2.118638ms
localhost:22379 is healthy: successfully committed proposal: took = 3.631388ms
localhost:32379 is healthy: successfully committed proposal: took = 2.157051ms
COMMENT
curl http://localhost:2379/version
<<COMMENT
{"etcdserver":"3.3.5","etcdcluster":"3.3.0"}
COMMENT
curl http://localhost:22379/version
<<COMMENT
{"etcdserver":"3.3.5","etcdcluster":"3.3.0"}
COMMENT
curl http://localhost:32379/version
<<COMMENT
{"etcdserver":"3.3.5","etcdcluster":"3.3.0"}
COMMENT
```
#### Step 2: download snapshot backup from leader
[Download the snapshot backup](../op-guide/maintenance.md#snapshot-backup) to provide a downgrade path should any problems occur.
etcd leader is guaranteed to have the latest application data, thus fetch snapshot from leader:
When each etcd process is stopped, expected errors will be logged by other cluster members. This is normal since a cluster member connection has been (temporarily) broken:
```bash
10.237579 I | etcdserver: updating the cluster version from 3.0 to 3.3
10.238315 N | etcdserver/membership: updated the cluster version from 3.0 to 3.3
10.238451 I | etcdserver/api: enabled capabilities for version 3.3
^C21.192174 N | pkg/osutil: received interrupt signal, shutting down...
21.192459 I | etcdserver: 7339c4e5e833c029 starts leadership transfer from 7339c4e5e833c029 to 729934363faa4a24
21.192569 I | raft: 7339c4e5e833c029 [term 8] starts to transfer leadership to 729934363faa4a24
21.192619 I | raft: 7339c4e5e833c029 sends MsgTimeoutNow to 729934363faa4a24 immediately as 729934363faa4a24 already has up-to-date log
WARNING: 2018/05/17 12:45:21 grpc: addrConn.resetTransport failed to create client transport: connection error: desc="transport: Error while dialing dial tcp: operation was canceled"; Reconnecting to {localhost:2379 0 <nil>}
WARNING: 2018/05/17 12:45:21 grpc: addrConn.transportMonitor exits due to: grpc: the connection is closing
21.193589 I | raft: 7339c4e5e833c029 [term: 8] received a MsgVote message with higher term from 729934363faa4a24 [term: 9]
21.193626 I | raft: 7339c4e5e833c029 became follower at term 9
21.193651 I | raft: 7339c4e5e833c029 [logterm: 8, index: 9, vote: 0] cast MsgVote for 729934363faa4a24 [logterm: 8, index: 9] at term 9
21.193675 I | raft: raft.node: 7339c4e5e833c029 lost leader 7339c4e5e833c029 at term 9
21.194424 I | raft: raft.node: 7339c4e5e833c029 elected leader 729934363faa4a24 at term 9
21.292898 I | etcdserver: 7339c4e5e833c029 finished leadership transfer from 7339c4e5e833c029 to 729934363faa4a24 (took 100.436391ms)
21.292975 I | rafthttp: stopping peer 729934363faa4a24...
21.293206 I | rafthttp: closed the TCP streaming connection with peer 729934363faa4a24 (stream MsgApp v2 writer)
21.293225 I | rafthttp: stopped streaming with peer 729934363faa4a24 (writer)
21.293437 I | rafthttp: closed the TCP streaming connection with peer 729934363faa4a24 (stream Message writer)
21.293459 I | rafthttp: stopped streaming with peer 729934363faa4a24 (writer)
21.293514 I | rafthttp: stopped HTTP pipelining with peer 729934363faa4a24
21.293590 W | rafthttp: lost the TCP streaming connection with peer 729934363faa4a24 (stream MsgApp v2 reader)
21.293610 I | rafthttp: stopped streaming with peer 729934363faa4a24 (stream MsgApp v2 reader)
21.293680 W | rafthttp: lost the TCP streaming connection with peer 729934363faa4a24 (stream Message reader)
21.293700 I | rafthttp: stopped streaming with peer 729934363faa4a24 (stream Message reader)
21.293711 I | rafthttp: stopped peer 729934363faa4a24
21.293720 I | rafthttp: stopping peer b548c2511513015...
21.293987 I | rafthttp: closed the TCP streaming connection with peer b548c2511513015 (stream MsgApp v2 writer)
21.294063 I | rafthttp: stopped streaming with peer b548c2511513015 (writer)
21.294467 I | rafthttp: closed the TCP streaming connection with peer b548c2511513015 (stream Message writer)
21.294561 I | rafthttp: stopped streaming with peer b548c2511513015 (writer)
21.294742 I | rafthttp: stopped HTTP pipelining with peer b548c2511513015
21.294867 W | rafthttp: lost the TCP streaming connection with peer b548c2511513015 (stream MsgApp v2 reader)
21.294892 I | rafthttp: stopped streaming with peer b548c2511513015 (stream MsgApp v2 reader)
21.294990 W | rafthttp: lost the TCP streaming connection with peer b548c2511513015 (stream Message reader)
21.295004 E | rafthttp: failed to read b548c2511513015 on stream Message (context canceled)
21.295013 I | rafthttp: peer b548c2511513015 became inactive
21.295024 I | rafthttp: stopped streaming with peer b548c2511513015 (stream Message reader)
21.295035 I | rafthttp: stopped peer b548c2511513015
```
#### Step 4: restart the etcd server with same configuration
Restart the etcd server with same configuration but with the new etcd binary.
The new v3.4 etcd will publish its information to the cluster. At this point, cluster still operates as v3.3 protocol, which is the lowest common version.
> `{"level":"info","ts":1526586617.1649797,"caller":"api/capability.go:76","msg":"enabled capabilities for version","cluster-version":"3.3"}`
> `{"level":"info","ts":1526586617.2107732,"caller":"etcdserver/server.go:1770","msg":"published local member to cluster through raft","local-member-id":"7339c4e5e833c029","local-member-attributes":"{Name:s1 ClientURLs:[http://localhost:2379]}","request-path":"/0/members/7339c4e5e833c029/attributes","cluster-id":"7dee9ba76d59ed53","publish-timeout":7}`
Verify that each member, and then the entire cluster, becomes healthy with the new v3.4 etcd binary:
```bash
etcdctl endpoint health --endpoints=localhost:2379,localhost:22379,localhost:32379
<<COMMENT
localhost:32379 is healthy: successfully committed proposal: took = 2.337471ms
localhost:22379 is healthy: successfully committed proposal: took = 1.130717ms
localhost:2379 is healthy: successfully committed proposal: took = 2.124843ms
COMMENT
```
Un-upgraded members will log warnings like the following until the entire cluster is upgraded.
This is expected and will cease after all etcd cluster members are upgraded to v3.4:
```
:41.942121 W | etcdserver: member 7339c4e5e833c029 has a higher version 3.4.0
:45.945154 W | etcdserver: the local etcd version 3.3.5 is not up-to-date
```
#### Step 5: repeat *step 3* and *step 4* for rest of the members
When all members are upgraded, the cluster will report upgrading to 3.4 successfully:
Member 1:
> `{"level":"info","ts":1526586949.0920913,"caller":"api/capability.go:76","msg":"enabled capabilities for version","cluster-version":"3.4"}`
> `{"level":"info","ts":1526586949.0921566,"caller":"etcdserver/server.go:2272","msg":"cluster version is updated","cluster-version":"3.4"}`
In the general case, upgrading from etcd 3.4 to 3.5 can be a zero-downtime, rolling upgrade:
- one by one, stop the etcd v3.4 processes and replace them with etcd v3.5 processes
@@ -12,7 +10,7 @@ Before [starting an upgrade](#upgrade-procedure), read through the rest of this
### Upgrade checklists
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/etcd-io/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
**NOTE:** When [migrating from v2 with no v3 data](https://github.com/coreos/etcd/issues/9480), etcd server v3.2+ panics when etcd restores from existing snapshots but no v3 `ETCD_DATA_DIR/member/snap/db` file. This happens when the server had migrated from v2 with no previous v3 data. This also prevents accidental v3 data loss (e.g. `db` file might have been moved). etcd requires that post v3 migration can only happen with v3 data. Do not upgrade to newer v3 versions until v3.0 server contains v3 data.
Highlighted breaking changes in 3.5.
@@ -45,7 +43,7 @@ TODO(add more monitoring guides); v3.4 adds `etcd --logger=zap` support for stru
#### Deprecated in `etcd --log-output`
v3.4 renamed [`etcd --log-output` to `--log-outputs`](https://github.com/etcd-io/etcd/pull/9624) to support multiple log outputs.
v3.4 renamed [`etcd --log-output` to `--log-outputs`](https://github.com/coreos/etcd/pull/9624) to support multiple log outputs.
**`etcd --log-output` has been deprecated in v3.5.**
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.