f5753f2f51
*: support lease Attach
...
Now we can attach keys to leases. And revoking the lease removes all
the attached keys of that lease.
2016-01-09 11:01:58 -08:00
a7287b6374
Merge pull request #4161 from gyuho/typo_in_benchmark
...
tools/benchmark/cmd: print error out to stderr
2016-01-07 18:50:09 -08:00
187064187c
tools/benchmark: report standard deviation
2016-01-07 15:30:23 -08:00
cc39cde319
tools/benchmark/cmd: print error out to stderr
2016-01-07 12:55:26 -08:00
f9af744be3
lease: move storage.NoLease to lease package
2016-01-07 10:33:44 -08:00
5dd3f91903
*: make backend outside kv
...
KV and lease will share the same backend. Thus we need to make
backend outside KV.
2016-01-05 19:55:29 -08:00
4336278b44
*: support put with lease
2016-01-04 15:54:06 -08:00
ac330bb7c9
*: update watch related proto
...
1. Add watch/cancel request
2. Add necessary fields in response to return watch error
3. Add watch_id into watch response
2016-01-01 10:22:21 -08:00
a74147384d
Merge pull request #4070 from mitake/storage-bench
...
tools: a new tool for benchmarking storage backends
2015-12-31 22:26:46 -08:00
454865bd67
tools: a new tool for benchmarking storage backends
...
Current etcd repository has a test for benchmarking a storage backend
in storage/kvstore_bench_test.go. However, it is hard to test various
parameters (e.g. batch interval, a number of keys, etc) with the test.
This commit adds a new benchmarking subcommand "storage" to
tools/benchmark. It will encourage analysis of storage backends with
various parameter and complex workloads.
Exmaple usage:
$ ./benchmark storage put
total: 9.894173792s
average: 9.894173ms
minimum latency: 6.596991ms
maximum latency: 29.455695ms
2016-01-01 15:18:54 +09:00
1689bb3f02
tools/benchmark: support puts on sequential keys over a bounded keyspace
...
This patch makes it possible to use benchmark to generate n keys and
make random updates to only those n keys.
2015-12-30 16:47:24 -08:00
b97f78d356
tools/benchmark: ignore the binary "benchmark"
2015-12-30 16:02:52 +09:00
8e728afa62
tools/benchmark: support connecting to several endpoints
...
--endpoints is comma separated but gRPC blocks forever on comma
separated lists. Instead, round-robin select endpoints when
creating new connections.
2015-12-28 10:22:33 -08:00
382103af60
tools/benchmark: stream results into reports
...
Reports depended on writing all results to a large buffered channel and
reading from that synchronously. Similarly, requests were buffered the
same way which can take significant memory on big request strings. Instead,
have reports stream in results as they're produced then print when the
results channel closes.
2015-12-23 11:24:35 -08:00
4f5f999847
tools/functional-test: add docker support
...
Commit adds docker bits to run functional tester within containers.
requires:
- docker 1.9 (networking)
- docker-compose
2015-12-18 15:56:18 +01:00
4289871cef
tools/function-tester: fix shadowed variables
...
Fixes for https://github.com/coreos/etcd/issues/3954 .
2015-12-12 09:38:20 -08:00
8d4073d078
Merge pull request #3945 from xiang90/new_watch_bench
...
tools/benchmark: add watch subcommand.
2015-12-03 16:56:02 -08:00
c6430b3292
tools/benchmark: add watch subcommand.
...
Watch command run benchmark tests for watch releated operations:
1. watch keys 2. sending events to watchers.
To test 2, this test also puts keys into etcd cluster to trigger
event sending.
Besides the benchmark results showed at client side, the tester
can also monitor the server-side mem/cpu usage and also check
the metrics of slow watchers. If there are a lot of slow watchers,
it means etcd server is over-loaded by watchers.
2015-12-03 15:56:17 -08:00
93240bd0f0
Merge pull request #3933 from ngaut/fix-shadow-variables
...
tools: Fix shadow variables and incorrect printing format.
2015-12-01 08:55:22 -08:00
faff00d19e
tools: rewrite benchmark tool
2015-11-30 11:52:43 -08:00
232439191c
tools: Fix shadow variables and incorrect printing format.
2015-11-30 20:13:29 +08:00
dfc7cc7a62
tools/etcd-tester: extend timeout for stresser
...
Extend the timeout from 1s to defaultRequestTimeout 5s.
The 1s may bring unwanted burden to the target member. If the member is
busy at recovering, it has limited bandwidth for client requests. A
short timeout at client side will retry quickly while keeping the
on-going connections. Thus, etcd will queue lots of requests and
connections and takes long time to clear them. This finally causes the
timeout of member health check.
This problem is a general one that how etcd handles amounts of requests
at the same time in a good way. We don't plan to address it at current
stage.
2015-11-16 11:47:08 -08:00
1a3f7f7fa4
*: rename etcd service to kv service in gRPC
2015-11-04 10:05:49 -08:00
e1b2e7245b
tools/etcd-top: add copyright header
2015-11-01 18:19:32 -08:00
00557e96af
tools: add etcd-top
2015-11-01 18:07:27 -08:00
c4b3ad72d9
Merge pull request #3544 from xiang90/bench
...
v3benchmark: add put benchmark
2015-09-17 15:10:13 -07:00
f69582e1a2
v3benchmark: add put benchmark
2015-09-17 14:48:07 -07:00
3a2700141e
Merge pull request #3539 from xiang90/bench
...
godep: use github.com/cheggaaa/pb
2015-09-15 16:12:34 -07:00
38dd680f2e
godep: use github.com/cheggaaa/pb
2015-09-15 16:08:07 -07:00
8bb50635ce
Merge pull request #3538 from xiang90/bench
...
benchmarkv3: refactoring the main logic
2015-09-15 15:58:32 -07:00
4deb12fbbb
benchmarkv3: refactoring the main logic
2015-09-15 15:57:38 -07:00
7848ac3979
*: add missing license headers
2015-09-15 14:09:01 -07:00
2782418923
Godep: fixed missing dependencies
2015-09-04 04:51:44 +00:00
3632a1b9b1
*: add initial read benchmark for etcd v3
...
It includes the initial read benchmark for etcd v3.
This is the first step to give some rough thoughts. I haven't digged
deeper to answer some questions, including why its performance is not
better than HTTP + json, why one put will cause performance downgrade.
2015-08-25 07:50:18 -07:00
2b8abeb093
*: remove migration related stuff from 2.2
2015-08-01 19:37:20 +08:00
e71d43b58e
tools/functional-testing: add help message for flags
...
Help users to understand what these flags are for.
2015-04-29 13:59:55 -07:00
eafdd3b718
Merge pull request #2730 from yichengq/tester-key-param
...
main: parameterize stress key size and key suffix range
2015-04-27 17:02:36 -07:00
057d21cf79
main: parameterize stress key size and key suffix range
...
It faciliates tester to adjust the size of each request, the number of
keys in the store and the size of snapshot.
2015-04-27 16:46:56 -07:00
3127a3b659
tools/etcd-tester: wait longer for health
...
It dramatically reduce the probability that follower failed to catch up
the leader.
2015-04-21 17:55:24 -07:00
a9157ce6d3
build: do not build internal debugging tool
...
We are still playing around with the dump-log tool.
Stop building it publicly until we are happy with its
ux and functionality.
2015-03-31 11:45:12 -07:00
81750ab2d7
Merge pull request #2600 from yichengq/failure-isolate
...
tools/functional-tester: add isolate failures
2015-03-29 22:43:51 -07:00
04a62dd54b
tools/functional-tester: add isolate failures
2015-03-29 00:29:47 -07:00
097a56fe01
tools/etcd-agent: stop etcd only if it is running
...
Stop etcd only if it is running, and not report error when stopping etcd
which is not started.
2015-03-28 19:31:06 -07:00
d920c5b801
tools/etcd-tester: stress cluster using 50MB snapshot
2015-03-19 14:52:27 -07:00
46ebb83b90
tools/functional-tester/etcd-tester: report agent status
2015-03-13 15:29:57 -07:00
83bb02e320
tools/funcational-tester/etcd-agent: log the error for dubgging
2015-03-13 12:08:08 -07:00
e46beb75c8
tools/functional-tester/etcd-agent: add status rpc
2015-03-13 10:48:06 -07:00
462f32a81b
tools/functional-tester: add http status reporter
2015-03-12 12:49:48 -07:00
24a210ab20
tools/etcd-tester: add kill one member tests
2015-03-10 11:38:54 -07:00
ba20016f0f
tools/etcd-tester: reorganize failures
2015-03-05 21:14:41 -08:00
Xiang Li