47f5b7c3ad
Merge pull request #7876 from fanminshi/fix_7628
...
etcdserver: renaming db happens after snapshot persists to wal and snap files
2017-05-09 16:15:41 -07:00
dfdaf082c5
etcdserver: add a test to ensure renaming db happens before persisting wal and snap files
2017-05-09 14:00:22 -07:00
8b7b7222dd
etcdserver: renaming db happens after snapshot persists to wal and snap files
...
In the case that follower recieves a snapshot from leader
and crashes before renaming xxx.snap.db to db but after
snapshot has persisted to .wal and .snap, restarting
follower results loading old db, new .wal, and new .snap.
This will causes a index mismatch between snap metadata index
and consistent index from db.
This pr forces an ordering where saving/renaming db must
happen after snapshot is persisted to wal and snap file.
this guarantees wal and snap files are newer than db.
on server restart, etcd server checks if snap index > db consistent index.
if yes, etcd server attempts to load xxx.snap.db where xxx=snap index
if there is any and panic other wise.
FIXES #7628
2017-05-09 14:00:12 -07:00
a53a9e167e
Merge pull request #7898 from yudai/nit_remove_dup
...
v3rpc: remove duplicated error case for lease.ErrLeaseNotFound
2017-05-09 12:35:31 -07:00
b8875515a4
Merge pull request #7890 from yudai/keep_ka_loop_running
...
clientv3: Do no stop keep alive loop by server side errors
2017-05-09 11:00:21 -07:00
01a985eda5
Merge pull request #7897 from gyuho/bom
...
scripts: add 'BOM' update script
2017-05-09 10:52:42 -07:00
010ffc0692
v3rpc: remove duplicated error case for lease.ErrLeaseNotFound
2017-05-08 20:09:41 -07:00
8c9f01ef53
scripts: add 'BOM' update script
...
Need this script when we add external dependencies.
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-08 17:59:11 -07:00
aa85b0cea7
clientv3: Do no stop keep alive loop by server side errors
2017-05-08 15:47:34 -07:00
aac2292ab5
Merge pull request #7882 from heyitsanthony/srv-priority
...
gateway: DNS SRV priority
2017-05-08 14:17:04 -07:00
3a2e7653f2
Merge pull request #7879 from gyuho/http-server
...
embed: gracefully close peer handler
2017-05-08 14:00:45 -07:00
c232814003
etcdmain, tcpproxy: srv-priority policy
...
Adds DNS SRV weighting and priorities to gateway.
Partially addresses #4378
2017-05-08 11:35:18 -07:00
2655540481
Merge pull request #7892 from fanminshi/add_snashot_duration_metric
...
backend: add prometheus metric for large snapshot duration.
2017-05-08 11:22:51 -07:00
25eef5a6e4
Merge pull request #7893 from philips/readme-tagline
...
README: use the same tagline from github
2017-05-08 09:11:08 -07:00
7d21d6c894
embed: gracefully close peer handlers on shutdown
...
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-06 07:47:23 -07:00
af7d051019
Merge pull request #7885 from luedigernet/fix-TestEvent
...
Fix watch_test.go TestEvent
2017-05-05 23:31:59 -07:00
90af2ff302
README: use the same tagline from github
...
Just be consistent with the messaging and use of etcd
2017-05-05 18:07:26 -07:00
230106dd3c
backend: add prometheus metric for large snapshot duration.
...
FIXES #7878
2017-05-05 17:27:33 -07:00
8b081ce9b3
clientv3: check IsModify
...
Fix watch_test.go TestEvent
Prior to This fix the isModify case of the table driven test was never checked.
2017-05-05 19:39:59 +02:00
07ad18178d
pkg/srv: package for SRV utilities
...
Trying to decouple the v2 client from SRV code. Can't move
into discovery/ since that creates a circular dependency. So,
give up and move all the SRV code into a new package.
2017-05-05 09:27:59 -07:00
db6f45e939
Merge pull request #7830 from aaronlehmann/new-nodes-start-active
...
raft: Set the RecentActive flag for newly added nodes
2017-05-05 08:59:25 -07:00
1f8de1aab0
Merge pull request #7877 from fanminshi/warning_on_snapshotting
...
backend: print snapshotting duration warning every 30s
2017-05-04 18:03:47 -07:00
f7f30f2361
backend: print snapshotting duration warning every 30s
...
FIXES #7870
2017-05-04 16:41:03 -07:00
9451fa1f9c
raft: Add unit test TestAddNodeCheckQuorum
...
This test verifies that adding a node does not cause the leader to step
down until at least one full ElectionTick cycle elapses.
Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
2017-05-04 15:04:30 -07:00
c3b96f8a69
Merge pull request #7875 from yudai/compact_every_time
...
compactor: Make periodic compactor runs every hour
2017-05-04 13:24:27 -07:00
60dbad5a85
compactor: Make periodic compactor runs every hour
...
Closes #7868 .
2017-05-04 10:32:51 -07:00
505bf8c708
Merge pull request #7864 from gyuho/doc-link-fixes
...
*: run 'marker' in CI
2017-05-04 09:14:06 -07:00
2e32d2142d
Merge pull request #7869 from heyitsanthony/fix-lease-require-leader-test
...
clientv3/integration: drain keepalives before waiting for leader loss
2017-05-04 08:29:16 -07:00
282c6fd17d
Documentation: remove '[]' from '[DEPRECATED]'
...
To make 'marker' pass the tests
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-04 08:26:01 -07:00
c2959c998f
test: run 'marker' to find broken links
...
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-04 08:26:00 -07:00
e9a63473a0
scripts,travis: install 'marker' for CI tests
...
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-04 08:26:00 -07:00
7f05e220a4
Merge pull request #7874 from gyuho/scripts
...
integration/fixtures-expired: do not force 'rm'
2017-05-03 19:39:00 -07:00
4edbae4a91
integration/fixtures-expired: do not force 'rm'
...
To make gencerts.sh script safer.
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-03 18:45:44 -07:00
3b251b0ed3
Merge pull request #7871 from gyuho/fix-doc-2
...
*: fix broken links in markdown
2017-05-03 16:58:38 -07:00
4203320d04
*: fix other broken links in markdown
...
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-03 16:57:44 -07:00
feb930e357
Documentation/v3: fix broken links
...
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-03 16:57:38 -07:00
e4e057f8f7
Documentation/v2: fix broken links
...
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-03 15:37:53 -07:00
9fee35b02d
Merge pull request #7842 from heyitsanthony/fix-switch-race
...
clientv3: don't race on upc/downc/switch endpoints in balancer
2017-05-03 13:48:00 -07:00
f6d0dda187
clientv3/integration: drain keepalives before waiting for leader loss
...
500ms keepalive delay on proxy side causes client to sometimes send
a second keepalive since it waits more than 500ms for the first response.
Fixes #7658
2017-05-03 13:22:45 -07:00
8f40517adb
integration: close proxy's lease client
2017-05-03 13:22:24 -07:00
61c5a0c6ae
Merge pull request #7867 from gyuho/fix-tls-test
...
integration: clean up TLS reload tests, fix no-file while renaming
2017-05-03 12:43:41 -07:00
85fa594265
integration: clean up TLS reload tests, fix no-file while renaming
...
Fix https://github.com/coreos/etcd/issues/7865 .
It is also possible to have mis-matched key file
while renaming directories.
Signed-off-by: Gyu-Ho Lee <gyuhox@gmail.com>
2017-05-03 11:59:09 -07:00
c2d6a92b01
Merge pull request #7853 from gyuho/revert
...
Documentation/upgrades: revert KeepAlive interface change
2017-05-03 11:04:15 -07:00
24e85b2454
Merge pull request #7852 from heyitsanthony/revert-lease-err-ka
...
Revert "Merge pull request #7732 from heyitsanthony/lease-err-ka"
2017-05-03 11:03:17 -07:00
27b3bf230b
Merge pull request #7863 from heyitsanthony/stm-apis
...
concurrency: provide old STM functions as deprecated
2017-05-03 10:19:13 -07:00
de2e959b27
Merge pull request #7856 from fanminshi/fix_consistent_index_update
...
etcdserver: apply() sets consistIndex for any entry type
2017-05-03 09:07:16 -07:00
31d5d610fc
concurrency: provide old STM functions as deprecated
...
semver
2017-05-03 02:07:01 -07:00
e33b10a666
etcdserver: add a test to ensure config change also update ConsistIndex
2017-05-02 16:51:40 -07:00
61abf25859
integration: close accepted connection on stopc path
...
Connection pausing added another exit condition in the listener
path, causing the bridge to leak connections instead of closing
them when signalled to close. Also adds some additional Close
paranoia.
Fixes #7823
2017-05-02 16:46:43 -07:00
43e5f892f6
clientv3: don't race on upc/downc/switch endpoints in balancer
...
If the balancer update notification loop starts with a downed
connection and endpoints are switched while the old connection is up,
the balancer can potentially wait forever for an up connection without
refreshing the connections to reflect the current endpoints.
Instead, fetch upc/downc together, only caring about a single transition
either from down->up or up->down for each iteration
Simple way to reproduce failures: add time.Sleep(time.Second) to the
beginning of the update notification loop.
2017-05-02 16:43:24 -07:00
fanmin shi