6825ffe | Gyu-Ho Lee | 05 January 2017, 18:53:02 UTC | integration: use only digits in unix ports Fix https://github.com/coreos/etcd/issues/6959. | 05 January 2017, 20:34:54 UTC |
a42b399 | Anthony Romano | 05 January 2017, 19:28:33 UTC | Merge pull request #7094 from heyitsanthony/fix-duplicate-grant auth: use quorum get for GetUser/GetRole for mutable operations | 05 January 2017, 19:28:33 UTC |
5feb4e1 | Anthony Romano | 05 January 2017, 03:04:08 UTC | Merge pull request #7103 from heyitsanthony/proxy-watch-close grpcproxy: tear down watch when client context is done | 05 January 2017, 03:04:08 UTC |
fd72ecf | Anthony Romano | 05 January 2017, 00:33:19 UTC | Merge pull request #7087 from sinsharat/make_etcd-runner_command_compliant etcd-runner: make command compliant | 05 January 2017, 00:33:19 UTC |
e179225 | Anthony Romano | 05 January 2017, 00:23:27 UTC | grpcproxy: tear down watch when client context is done If client closes but all watch streams are not canceled, the outstanding watch will wait until it is canceled, causing watch server to potentially wait forever to close. Fixes #7102 | 05 January 2017, 00:23:27 UTC |
154f268 | Anthony Romano | 04 January 2017, 21:44:49 UTC | Merge pull request #7001 from heyitsanthony/etcdctl-doc etcdctl: tighten up output, reorganize README.md | 04 January 2017, 21:44:49 UTC |
10d3b81 | Gyu-Ho Lee | 04 January 2017, 20:09:29 UTC | Merge pull request #7093 from gyuho/member etcdserver: expose ErrMemberNotEnoughStarted | 04 January 2017, 20:09:29 UTC |
f9f691e | Anthony Romano | 04 January 2017, 00:34:36 UTC | auth: use quorum get for GetUser/GetRole for mutable operations GetUser would not propagate to the minority node, causing TestCtlV2GetRoleUser to run CreateUser instead of UpdateUser. Instead, use quorum get to fetch the current state of auth. Fixes #7069 | 04 January 2017, 19:55:07 UTC |
729dcd5 | Xiang Li | 04 January 2017, 18:47:44 UTC | Merge pull request #7090 from vimalk78/fix-comactor-resume-leadr-change#7040 etcdserver: resume compactor only if leader | 04 January 2017, 18:47:44 UTC |
559a82f | Anthony Romano | 04 January 2017, 18:32:07 UTC | Merge pull request #7097 from heyitsanthony/benchmark-verbose benchmark: enable grpc error logging on stderr | 04 January 2017, 18:32:07 UTC |
40ae83b | Xiang Li | 04 January 2017, 18:16:48 UTC | Merge pull request #7099 from overvenus/patch-1 docs: fix recovery example in recovery.md | 04 January 2017, 18:16:48 UTC |
37501e2 | Xiang Li | 04 January 2017, 17:13:11 UTC | Merge pull request #7092 from xiang90/fix_raft raft: use status to test node stop | 04 January 2017, 17:13:11 UTC |
7aeddf6 | Neil Shen | 04 January 2017, 11:41:15 UTC | docs: fix recovery example in recovery.md | 04 January 2017, 11:41:15 UTC |
d0f301a | sharat | 03 January 2017, 11:33:07 UTC | etcd-runner:add flags in watcher for hardcoded values | 04 January 2017, 09:47:53 UTC |
b8444d4 | Anthony Romano | 04 January 2017, 08:20:35 UTC | benchmark: enable grpc error logging on stderr Lets you see connection errors (e.g., if tls is misconfigured) | 04 January 2017, 08:26:43 UTC |
5fac6b8 | vimalk78 | 03 January 2017, 19:22:21 UTC | etcdserver: resume compactor only if leader | 03 January 2017, 23:31:14 UTC |
2b5f9e1 | Gyu-Ho Lee | 03 January 2017, 23:14:22 UTC | etcdserver: expose ErrNotEnoughStartedMembers Fix https://github.com/coreos/etcd/issues/7072. | 03 January 2017, 23:23:06 UTC |
fc8cd44 | Xiang Li | 03 January 2017, 23:01:44 UTC | raft: use status to test node stop n.Tick() is async. It can be racy when running with n.Stop(). n.Status() is sync and has a feedback mechnism internally. So there wont be any race between n.Status() and n.Stop() call. | 03 January 2017, 23:18:48 UTC |
61064a7 | Gyu-Ho Lee | 03 January 2017, 18:34:13 UTC | Merge pull request #7085 from gyuho/raft-example-snapshot raftexample: load snapshot when opening WAL | 03 January 2017, 18:34:13 UTC |
5cb6dd2 | sharat | 03 January 2017, 05:50:56 UTC | etcd-runner: make command compliant | 03 January 2017, 09:13:58 UTC |
0af1679 | Gyu-Ho Lee | 31 December 2016, 00:26:27 UTC | raftexample: load snapshot when opening WAL Fix https://github.com/coreos/etcd/issues/7056. Previously we don't load snapshot when replaying WAL. | 31 December 2016, 01:28:57 UTC |
24601ca | Anthony Romano | 30 December 2016, 20:51:31 UTC | Merge pull request #7084 from heyitsanthony/watch-proxy-leak integration: wait for watch proxy to finish on client close | 30 December 2016, 20:51:31 UTC |
7544139 | Anthony Romano | 30 December 2016, 19:52:39 UTC | integration: defer clus.Terminate in watch tests Common pattern was defer cancel(), but clus.Terminate() at the end of the test. This appears to lead to a deadlock that is only released once the context times out, causing inflated test times. | 30 December 2016, 20:34:04 UTC |
9b5eb1a | Anthony Romano | 30 December 2016, 19:24:23 UTC | grpcproxy, etcdmain, integration: return done channel with WatchServer Makes it possible to synchronously close the watch server. Fixes #7078 | 30 December 2016, 20:09:48 UTC |
29e14dd | Gyu-Ho Lee | 30 December 2016, 18:14:12 UTC | Merge pull request #7081 from gyuho/timeout-rafthttp rafthttp: bump up timeout in pipeline test | 30 December 2016, 18:14:12 UTC |
cbb6ede | fanmin shi | 30 December 2016, 18:13:07 UTC | Merge pull request #7067 from fanminshi/rework_coverage_unit_integration coverage: rework coverage for unit and integration tests | 30 December 2016, 18:13:07 UTC |
d25f9fe | Gyu-Ho Lee | 30 December 2016, 17:46:16 UTC | rafthttp: bump up timeout in pipeline test Fix https://github.com/coreos/etcd/issues/6283. The timeout is too short. It could take more than 10ms to send when the buffer gets full after 'pipelineBufSize' of requests. | 30 December 2016, 17:46:16 UTC |
74e7614 | fanmin shi | 30 December 2016, 00:38:48 UTC | testutil: whitelist thread created by go cover | 30 December 2016, 01:19:27 UTC |
d9a3472 | fanmin shi | 27 December 2016, 23:15:57 UTC | coverage: rework code coverage for unit and integration tests | 30 December 2016, 01:19:03 UTC |
0dce29a | fanmin shi | 29 December 2016, 22:37:46 UTC | Merge pull request #7077 from fanminshi/consistent_naming etcdserver: consistent naming in raftReadyHandler | 29 December 2016, 22:37:46 UTC |
8242049 | fanmin shi | 29 December 2016, 22:37:25 UTC | Merge pull request #7076 from fanminshi/fix_e2e_test e2e: unset ETCDCTL_API env var before running e2e tests | 29 December 2016, 22:37:25 UTC |
734dd75 | Gyu-Ho Lee | 29 December 2016, 19:29:45 UTC | Merge pull request #7075 from gyuho/version-pull e2e: poll '/version' in release upgrade tests | 29 December 2016, 19:29:45 UTC |
2a1bae0 | fanmin shi | 29 December 2016, 19:27:16 UTC | etcdserver: consistent naming in raftReadyHandler | 29 December 2016, 19:27:16 UTC |
b741452 | fanmin shi | 29 December 2016, 19:16:35 UTC | e2e: unset ETCDCTL_API env var before running u2e tests existing ETCDCTL_API env var causes e2e to fail some of its tests. ETCDCTL_API should not be set before e2e tests start. the tests themselves should set ETCDCTL_API properly. | 29 December 2016, 19:21:15 UTC |
4e1010c | Gyu-Ho Lee | 29 December 2016, 18:52:40 UTC | e2e: poll '/version' in release upgrade tests Fix https://github.com/coreos/etcd/issues/7065. | 29 December 2016, 18:52:40 UTC |
67c7560 | Anthony Romano | 29 December 2016, 00:30:08 UTC | Merge pull request #7070 from heyitsanthony/fix-lease-race lease: use atomics for accessing lease expiry | 29 December 2016, 00:30:08 UTC |
b5cde6b | Anthony Romano | 28 December 2016, 21:22:29 UTC | lease: use atomics for accessing lease expiry Demote was racing on expiry when LeaseTimeToLive called Remaining. Replace with intrinsics since the ordering isn't important, but torn writes are bad. | 28 December 2016, 23:44:14 UTC |
1643ed5 | Anthony Romano | 28 December 2016, 23:41:00 UTC | Merge pull request #7071 from heyitsanthony/bump-integration-timeout test: bump grpcproxy pass timeout to 15m | 28 December 2016, 23:41:00 UTC |
f876ccb | Anthony Romano | 28 December 2016, 22:56:57 UTC | test: bump grpcproxy pass timeout to 15m integration tests have a 15m timeout elsewhere. The lease stress tests seem to have pushed the running time over 10m on proxy CI, causing failures from timeout. | 28 December 2016, 22:56:57 UTC |
12d930b | Anthony Romano | 28 December 2016, 22:30:31 UTC | Merge pull request #7068 from heyitsanthony/fix-v2-health v2http: submit QGET in health endpoint if no progress | 28 December 2016, 22:30:31 UTC |
3519a97 | Xiang Li | 28 December 2016, 21:12:11 UTC | Merge pull request #7039 from mitake/benchmark-dialtimeout benchmark: a new option for configuring dial timeout | 28 December 2016, 21:12:11 UTC |
9690220 | Anthony Romano | 28 December 2016, 21:04:55 UTC | Merge pull request #7064 from heyitsanthony/fix-health-perms etcdctl: treat permission denied as healthy endpoint | 28 December 2016, 21:04:55 UTC |
e246356 | Anthony Romano | 28 December 2016, 00:34:11 UTC | v2http: submit QGET in health endpoint if no progress Removing the periodic SYNC calls broke the health endpoint since the raft index stops updating. Instead, don't bother monitoring the raft index; issue a QGET directly to get a consensus response. Fixes #6985 | 28 December 2016, 20:20:56 UTC |
46062ef | Anthony Romano | 28 December 2016, 02:44:33 UTC | e2e: test cluster-health | 28 December 2016, 20:20:55 UTC |
e63059e | Xiang Li | 28 December 2016, 20:03:53 UTC | Merge pull request #7030 from crandles/grpc-histograms etcdmain: add '--metrics' option | 28 December 2016, 20:03:53 UTC |
36b2d3f | crandl201 | 17 December 2016, 01:04:20 UTC | etcdmain: add --metrics flag for exposing histogram metrics this adds a new flag, --metrics, that can be used to enable extensive (histogram) metrics. Fixes #7024 | 28 December 2016, 18:04:52 UTC |
00e00f1 | Anthony Romano | 27 December 2016, 22:40:29 UTC | ctlv3: consider permission denied error to be healthy for endpoints Relaxes the permission expectations for endpoint health by noting: * permission denial on linearized reads is always through consensus * endpoint health means consensus with the cluster through the endpoint So, there's no need to require permission on a health check key in order to know whether the endpoint is healthy. Fixes #7057 | 28 December 2016, 17:13:27 UTC |
b940e0d | Xiang Li | 28 December 2016, 05:15:53 UTC | Merge pull request #7042 from petermattis/pmattis/resume-after-heartbeat-resp raft: resume paused followers on receipt of MsgHeartbeatResp | 28 December 2016, 05:15:53 UTC |
a662dde | Hitoshi Mitake | 20 December 2016, 05:37:01 UTC | benchmark: a new option for configuring dial timeout Current benchmark doesn't have an option for configuring dial timeout of gRPC. This commit adds --dial-timeout for the purpose. It is useful for stopping long sticking benchmarks. | 28 December 2016, 05:07:43 UTC |
407afc6 | Anthony Romano | 27 December 2016, 22:48:23 UTC | e2e: check etcdctl endpoint health is healthy if denied permission to key | 27 December 2016, 22:49:52 UTC |
c000848 | Gyu-Ho Lee | 27 December 2016, 20:36:48 UTC | Merge pull request #7054 from gyuho/err etcd-tester: remove unused err var from maxRev | 27 December 2016, 20:36:48 UTC |
db8b15b | Gyu-Ho Lee | 23 December 2016, 00:28:10 UTC | etcd-tester: remove unused err var from maxRev | 27 December 2016, 20:16:43 UTC |
89b18ff | fanmin shi | 27 December 2016, 19:26:15 UTC | Merge pull request #7015 from fanminshi/fix_lease_expired_too_soon lease: force leader to apply its pending committed index for lease op… | 27 December 2016, 19:26:15 UTC |
2faf72f | fanmin shi | 22 December 2016, 20:03:49 UTC | etcdserver: rework update committed index logic | 27 December 2016, 18:11:40 UTC |
17873f7 | Anthony Romano | 27 December 2016, 18:03:01 UTC | Merge pull request #7008 from heyitsanthony/fix-dns retry on resolution failure for advertised peer DNS check | 27 December 2016, 18:03:01 UTC |
d9b9821 | Xiang Li | 27 December 2016, 01:42:36 UTC | Merge pull request #7060 from hhkbp2/fix-pre-vote-tests raft: fix pre-vote tests | 27 December 2016, 01:42:36 UTC |
920b155 | Dylan.Wen | 26 December 2016, 06:31:59 UTC | raft: fix pre-vote tests | 26 December 2016, 06:31:59 UTC |
7b7feb4 | fanmin shi | 17 December 2016, 01:11:23 UTC | leasehttp: buffer error channel to prevent goroutine leak | 22 December 2016, 22:25:01 UTC |
fef4a79 | fanmin shi | 15 December 2016, 20:09:20 UTC | lease: force leader to apply its pending committed index for lease operations suppose a lease granting request from a follower goes through and followed by a lease look up or renewal, the leader might not apply the lease grant request locally. So the leader might not find the lease from the lease look up or renewal request which will result lease not found error. To fix this issue, we force the leader to apply its pending commited index before looking up lease. FIX #6978 | 22 December 2016, 22:24:38 UTC |
1a8e3ca | Gyu-Ho Lee | 22 December 2016, 21:12:38 UTC | Merge pull request #7053 from gyuho/typo etcd-tester: fix typo, add endpoint in logs | 22 December 2016, 21:12:38 UTC |
591bb5e | Gyu-Ho Lee | 22 December 2016, 20:51:27 UTC | etcd-tester: fix typo, add endpoint in logs | 22 December 2016, 20:51:27 UTC |
acbf0fa | Xiang Li | 20 December 2016, 17:14:27 UTC | Merge pull request #7041 from m1093782566/raft-safe raft: make memory storage set method thread safe | 20 December 2016, 17:14:27 UTC |
e625400 | Peter Mattis | 19 December 2016, 20:24:21 UTC | raft: resume paused followers on receipt of MsgHeartbeatResp Previously, paused followers were resumed upon sending a MsgHearbeat. Fixes #7037 | 20 December 2016, 13:22:09 UTC |
8151d4d | 杜军 | 20 December 2016, 10:48:52 UTC | raft: make memory storage set method thread safe | 20 December 2016, 10:48:52 UTC |
d62ce55 | Gyu-Ho Lee | 17 December 2016, 02:53:11 UTC | Merge pull request #7027 from gyuho/default-host embed: only override default advertised client URL if the client listen URL is 0.0.0.0 | 17 December 2016, 02:53:11 UTC |
e58287f | Gyu-Ho Lee | 16 December 2016, 23:01:53 UTC | embed: only override default advertised client URL if the client listen URL is 0.0.0.0 | 17 December 2016, 02:31:04 UTC |
af3451b | Gyu-Ho Lee | 16 December 2016, 23:54:49 UTC | Merge pull request #7018 from gyuho/why Documentation: add 'why.md' | 16 December 2016, 23:54:49 UTC |
bef87cc | Gyu-Ho Lee | 15 December 2016, 23:15:24 UTC | Documentation: add 'why.md' | 16 December 2016, 23:54:03 UTC |
f95f7a3 | Gyu-Ho Lee | 16 December 2016, 23:37:21 UTC | Merge pull request #7028 from gyuho/faq Documentation: add FAQs on membership operation | 16 December 2016, 23:37:21 UTC |
2f0e82a | Gyu-Ho Lee | 16 December 2016, 19:44:45 UTC | Documentation: add FAQs on membership operation Copy Anthony's answer from: https://github.com/coreos/etcd/issues/6103 https://github.com/coreos/etcd/issues/6114 | 16 December 2016, 23:13:40 UTC |
780d2f2 | Anthony Romano | 13 December 2016, 01:03:09 UTC | etcdctl: tighten up output, reorganize README.md Documentation was far too repetitive, making it a chore to read and make changes. All commands are now organized by functionality and all repetitive bits about return values and output are in a generalized subsections. etcdctl's output handling was missing a lot of commands. Similarly, in many cases an output format could be given but fail to report an error as expected. | 16 December 2016, 21:54:20 UTC |
531c306 | Anthony Romano | 16 December 2016, 19:38:22 UTC | Merge pull request #7023 from heyitsanthony/lease-freeze clientv3: fix lease "freezing" on unhealthy cluster | 16 December 2016, 19:38:22 UTC |
a375e91 | Anthony Romano | 16 December 2016, 08:28:17 UTC | clientv3: don't reset keepalive stream on grant failure Was triggering cancelation errors on outstanding KeepAlives if Grant had to retry. | 16 December 2016, 18:36:51 UTC |
46bd842 | Anthony Romano | 16 December 2016, 08:23:49 UTC | clientv3/integration: test lease grant/keepalive with/without failures | 16 December 2016, 18:36:51 UTC |
87b1d95 | Anthony Romano | 16 December 2016, 08:21:07 UTC | v3api, rpctypes: add ErrTimeoutDueToConnectionLost Lack of GRPC code was causing this to look like a halting error to the client. | 16 December 2016, 18:25:35 UTC |
d9e928d | Anthony Romano | 16 December 2016, 17:51:34 UTC | Merge pull request #7020 from heyitsanthony/etcdctl-migrate-warn etcdctl: warn when backend takes too long to open on migrate | 16 December 2016, 17:51:34 UTC |
1095773 | Xiang Li | 16 December 2016, 04:39:19 UTC | Merge pull request #7022 from hongchaodeng/master docs: explicitly set ETCDCTL_API=3 in recovery.md | 16 December 2016, 04:39:19 UTC |
fa733e1 | Hongchao Deng | 16 December 2016, 04:10:30 UTC | docs: explicitly set ETCDCTL_API=3 in recovery.md | 16 December 2016, 04:10:30 UTC |
e71ff36 | Anthony Romano | 16 December 2016, 02:54:31 UTC | etcdctl: warn when backend takes too long to open on migrate | 16 December 2016, 02:57:57 UTC |
52e3dc5 | Brandon Philips | 16 December 2016, 02:27:52 UTC | Documentation: minor fix nodes -> node | 16 December 2016, 02:27:52 UTC |
93e303e | Gyu-Ho Lee | 15 December 2016, 23:45:23 UTC | Merge pull request #7017 from gyuho/faq dev-guide: add limit.md | 15 December 2016, 23:45:23 UTC |
a1e572b | Gyu-Ho Lee | 15 December 2016, 23:12:04 UTC | dev-guide: add limit.md | 15 December 2016, 23:44:21 UTC |
5aeee91 | Anthony Romano | 15 December 2016, 20:43:17 UTC | Merge pull request #7006 from heyitsanthony/clusterid-split Documentation: FAQ entry for cluster ID mismatches | 15 December 2016, 20:43:17 UTC |
14c851c | Anthony Romano | 14 December 2016, 18:57:31 UTC | Documentation: FAQ entry for cluster ID mismatches | 15 December 2016, 19:27:24 UTC |
86a4384 | Xiang Li | 15 December 2016, 16:06:36 UTC | Merge pull request #7010 from dennwc/keepalive-exit-err clientv3: ensure KeepAlive channel is closed or error is returned | 15 December 2016, 16:06:36 UTC |
35fd5dc | Xiang Li | 15 December 2016, 16:04:31 UTC | Merge pull request #6903 from mitake/auth-member protect membership change RPCs with auth | 15 December 2016, 16:04:31 UTC |
b126e31 | Denys Smirnov | 15 December 2016, 14:06:27 UTC | clientv3: better error message for keep alive loop halt | 15 December 2016, 14:06:27 UTC |
d46b753 | Hitoshi Mitake | 28 November 2016, 08:18:34 UTC | e2e: test cases of protecting membership change with auth | 15 December 2016, 13:54:20 UTC |
86d7390 | Hitoshi Mitake | 28 November 2016, 05:30:51 UTC | auth, etcdserver: protect membership change operations with auth This commit protects membership change operations with auth. Only users that have root role can issue the operations. Implements https://github.com/coreos/etcd/issues/6899 | 15 December 2016, 13:54:20 UTC |
5183ce0 | Denys Smirnov | 15 December 2016, 00:43:41 UTC | clientv3: add test for keep alive loop exit case | 15 December 2016, 01:02:44 UTC |
e0bcd4d | Denys Smirnov | 05 December 2016, 17:22:15 UTC | clientv3: return error from KeepAlive if corresponding loop exits after recvKeepAliveLoop exits client might call KeepAlive adding request channel that will not be closed this fix makes sure that recvKeepAliveLoop is running before adding request to lessor's list and returns error otherwise Fixes #6922 | 15 December 2016, 01:02:35 UTC |
d8513ad | Anthony Romano | 15 December 2016, 00:06:32 UTC | Merge pull request #7007 from heyitsanthony/lease-close clientv3: close Lease on client Close | 15 December 2016, 00:06:32 UTC |
26a3e9a | Anthony Romano | 14 December 2016, 22:22:37 UTC | membership: retry for 30s on advertise url check | 14 December 2016, 23:56:22 UTC |
29c30b2 | Anthony Romano | 14 December 2016, 22:21:41 UTC | etcdserver: retry for 30s on advertise url check | 14 December 2016, 23:56:22 UTC |
13b05ae | Anthony Romano | 14 December 2016, 22:07:33 UTC | netutil: ctx-ize URLStringsEqual Handles the case where the DNS entry will only be set up after etcd starts. | 14 December 2016, 23:46:30 UTC |
246fb29 | Anthony Romano | 14 December 2016, 20:11:17 UTC | clientv3: close Lease on client Close Fixes #6987 | 14 December 2016, 20:11:17 UTC |
a9f72ee | Anthony Romano | 14 December 2016, 20:05:37 UTC | Merge pull request #7005 from heyitsanthony/fix-pprof embed: deep copy user handlers | 14 December 2016, 20:05:37 UTC |
8f88632 | Gyu-Ho Lee | 14 December 2016, 19:51:34 UTC | Merge pull request #6965 from gyuho/faq Documentation: add more FAQs (follower, leader, sys-require) | 14 December 2016, 19:51:34 UTC |
626df4d | Gyu-Ho Lee | 08 December 2016, 21:08:59 UTC | Documentation: add more FAQs (follower, leader, sys-require) | 14 December 2016, 19:36:07 UTC |
cc931a2 | Anthony Romano | 14 December 2016, 18:12:40 UTC | embed: deep copy user handlers Shallow copy of user handlers leads to a nil map assignment when enabling pprof. Since the map is being modified, it should probably be deep copied into the server context, which fixes the crash. | 14 December 2016, 18:17:32 UTC |