| d01d611 | Xiang Li | 26 November 2014, 07:32:01 UTC | Merge pull request #1792 from xiang90/fix_raft_l raft: always write dummy entry to storage | 26 November 2014, 07:32:01 UTC |
| 39e6631 | Xiang Li | 26 November 2014, 07:27:40 UTC | raft: always write dummy entry to storage | 26 November 2014, 07:27:40 UTC |
| 7614aa5 | Yicheng Qin | 26 November 2014, 05:41:13 UTC | version: bump to alpha.4 | 26 November 2014, 05:41:13 UTC |
| 006da2f | Yicheng Qin | 26 November 2014, 02:46:23 UTC | Merge pull request #1789 from yichengq/234 rafthttp: only batch good MsgAppResp | 26 November 2014, 02:46:23 UTC |
| d5ceb26 | Yicheng Qin | 26 November 2014, 01:53:26 UTC | rafthttp: only batch good MsgAppResp A MsgAppResp with Reject set should be sent back to the leader as soon as possible instead of batching. | 26 November 2014, 01:53:26 UTC |
| dd57c1f | Yicheng Qin | 25 November 2014, 19:52:01 UTC | Merge pull request #1784 from yichengq/234 raft: not compact log if the compact index < first index of the log | 25 November 2014, 19:52:01 UTC |
| 4b43824 | Yicheng Qin | 25 November 2014, 19:28:25 UTC | raft: not compact log if the compact index < first index of the log It should ignore the compact operation instead of panic because the case that the log is restored from snapshot before executing compact is reasonable. | 25 November 2014, 19:51:20 UTC |
| 1a5333e | Yicheng Qin | 25 November 2014, 19:16:44 UTC | Merge pull request #1778 from yichengq/232 docs: add configuration doc | 25 November 2014, 19:16:44 UTC |
| 07ca99f | Yicheng Qin | 25 November 2014, 00:22:25 UTC | docs: add configuration doc | 25 November 2014, 19:09:29 UTC |
| aa2721e | Xiang Li | 25 November 2014, 16:59:16 UTC | Merge pull request #1781 from orls/fix-backup-snapshots Fix backup snapshots | 25 November 2014, 16:59:16 UTC |
| c46e304 | Xiang Li | 25 November 2014, 16:58:32 UTC | Merge pull request #1782 from orls/fix-forcenew-panic etcdserver: Fix panic when restoring from backup (-force-new-cluster) | 25 November 2014, 16:58:32 UTC |
| dbb6a75 | Owen Smith | 25 November 2014, 13:58:06 UTC | backup: Ensure dest snapshot dir always exists Otherwise the SaveSnap operation fails | 25 November 2014, 16:57:18 UTC |
| c67b937 | Owen Smith | 25 November 2014, 15:18:14 UTC | etcdserver: truncate WAL from correct index when forcing new cluster When loading from a backup with a snapshot and WAL, the length of WAL entries can be lower than the current index integer value, causing a panic when slicing off uncommitted entries. This looks for WAL entries higher than the current index before slicing. | 25 November 2014, 16:46:56 UTC |
| 9974bf0 | Owen Smith | 25 November 2014, 13:57:47 UTC | backup: Add fatal if dest snapshot failed Errors are currently silently swallowed | 25 November 2014, 14:08:20 UTC |
| e3cb3d6 | Xiang Li | 25 November 2014, 03:29:53 UTC | Merge pull request #1780 from xiang90/fix_raft_lead raft: always check leader changes in node run loop | 25 November 2014, 03:29:53 UTC |
| 9455119 | Xiang Li | 25 November 2014, 03:07:10 UTC | raft: always check leader changes in node run loop | 25 November 2014, 03:07:10 UTC |
| d67eea4 | Yicheng Qin | 25 November 2014, 01:44:25 UTC | Merge pull request #1779 from yichengq/233 rafthttp: limit the buffer for every read correctly | 25 November 2014, 01:44:25 UTC |
| 61ce494 | Yicheng Qin | 25 November 2014, 01:23:57 UTC | rafthttp: limit the buffer for every read correctly | 25 November 2014, 01:39:39 UTC |
| cfb96de | Yicheng Qin | 24 November 2014, 22:01:57 UTC | Merge pull request #1770 from yichengq/230 *: set read/write timeout for raft transport and listener | 24 November 2014, 22:01:57 UTC |
| 1e797c1 | Yicheng Qin | 24 November 2014, 21:32:09 UTC | rafthttp: limit the data size read from connection each time | 24 November 2014, 21:54:45 UTC |
| 3e55834 | Yicheng Qin | 24 November 2014, 07:37:54 UTC | *: set read/write timeout for raft transport and listener | 24 November 2014, 21:46:44 UTC |
| ad58122 | Yicheng Qin | 24 November 2014, 07:21:33 UTC | pkg/transport: fix dialer typo | 24 November 2014, 21:35:46 UTC |
| 400e573 | Yicheng Qin | 23 November 2014, 06:24:10 UTC | rafthttp: log start and stop of streaming | 24 November 2014, 21:35:46 UTC |
| bb640e3 | Xiang Li | 24 November 2014, 08:40:30 UTC | Merge pull request #1773 from xiang90/header pkg/transport: add copyright header | 24 November 2014, 08:40:30 UTC |
| c72221a | Xiang Li | 24 November 2014, 08:39:55 UTC | pkg/transport: add copyright header | 24 November 2014, 08:39:55 UTC |
| c6cbea4 | Xiang Li | 24 November 2014, 07:00:54 UTC | Merge pull request #1771 from xiang90/listener pkg/transport: add timeout dailer and timeout listener | 24 November 2014, 07:00:54 UTC |
| 35e6df6 | Xiang Li | 24 November 2014, 05:25:09 UTC | integration: longer request timeout for slow travis | 24 November 2014, 06:53:05 UTC |
| da1ff2d | Xiang Li | 24 November 2014, 05:12:02 UTC | pkg/transport: add timeout dailer and timeout listener | 24 November 2014, 05:18:34 UTC |
| 68e7986 | Xiang Li | 23 November 2014, 20:08:00 UTC | Merge pull request #1768 from xiang90/batcher rafthttp: add batcher | 23 November 2014, 20:08:00 UTC |
| 91bfead | Xiang Li | 22 November 2014, 00:42:12 UTC | rafthttp: add batcher After we enable streaming, there will be a large amount of msgAppResp due to high rate msgApp. We should batch msgAppResp in a meanful way. | 23 November 2014, 20:05:42 UTC |
| 66c30f2 | Xiang Li | 21 November 2014, 02:12:15 UTC | Merge pull request #1761 from xiang90/fix_raft raft: should not decrease match and next when handling out of order msgAppResp | 21 November 2014, 02:12:15 UTC |
| edcdffe | Xiang Li | 21 November 2014, 02:10:49 UTC | Merge pull request #1762 from xiang90/fix_import rafthttp: fix import | 21 November 2014, 02:10:49 UTC |
| 264a63b | Xiang Li | 21 November 2014, 02:05:45 UTC | rafthttp: fix import | 21 November 2014, 02:05:45 UTC |
| 063c5c7 | Xiang Li | 21 November 2014, 01:58:23 UTC | raft: should not decrease match and next when handling out of order msgAppResp | 21 November 2014, 01:58:23 UTC |
| c0fb1c8 | Brian Waldon | 21 November 2014, 00:26:14 UTC | Merge pull request #1755 from bcwaldon/golang.org-deps Switch to golang.org/x/net/context | 21 November 2014, 00:26:14 UTC |
| 5139257 | Barak Michener | 20 November 2014, 22:20:34 UTC | Merge pull request #1743 from barakmich/auto_upgrade etcdserver: autodetect v0.4 WALs and upgrade them to v0.5 automatically | 20 November 2014, 22:20:34 UTC |
| ce82a3e | Barak Michener | 20 November 2014, 21:49:34 UTC | Fix migration to allow snapshots to have the right IDs | 20 November 2014, 21:49:34 UTC |
| 53fbf0f | Xiang Li | 20 November 2014, 21:16:37 UTC | Merge pull request #1760 from xiang90/wal wal: add a bench for write entry | 20 November 2014, 21:16:37 UTC |
| 2d5ccf1 | Barak Michener | 20 November 2014, 20:37:15 UTC | add snapshotted integration test | 20 November 2014, 20:38:08 UTC |
| d1e7fee | Barak Michener | 20 November 2014, 18:17:47 UTC | fix test import loop | 20 November 2014, 20:38:08 UTC |
| 59a0c64 | Barak Michener | 19 November 2014, 20:33:37 UTC | fix import loop, add set to types, and fix comments | 20 November 2014, 20:38:08 UTC |
| 78ea333 | Barak Michener | 18 November 2014, 18:43:20 UTC | etcdserver: autodetect v0.4 WALs and upgrade them to v0.5 automatically | 20 November 2014, 20:38:08 UTC |
| d0dd205 | Xiang Li | 20 November 2014, 19:37:27 UTC | wal: add a bench for write entry | 20 November 2014, 20:07:17 UTC |
| aca195f | Yicheng Qin | 20 November 2014, 19:42:00 UTC | Merge pull request #1746 from yichengq/229 rafthttp: streaming server and client | 20 November 2014, 19:42:00 UTC |
| 9d53b94 | Yicheng Qin | 18 November 2014, 21:53:14 UTC | rafthttp: add streaming server and client | 20 November 2014, 19:34:50 UTC |
| da5538b | Brian Waldon | 20 November 2014, 18:06:16 UTC | dep: use vendored imports in codegangsta/cli | 20 November 2014, 18:19:13 UTC |
| 9a728a1 | Brian Waldon | 20 November 2014, 18:02:52 UTC | dep: bump golang.org/x/net/context Move from code.google.com/p/go.net/context to golang.org/x/net/context before bumping to latest. | 20 November 2014, 18:19:12 UTC |
| b50f331 | Xiang Li | 19 November 2014, 21:21:11 UTC | Merge pull request #1744 from xiang90/next raft: optimistically increase the next if the follower is already matched | 19 November 2014, 21:21:11 UTC |
| 12aaf04 | Brandon Philips | 19 November 2014, 20:13:21 UTC | Merge pull request #1752 from philips/include-etcd-migrate-in-releases scripts: build-release add etcd-migrate | 19 November 2014, 20:13:21 UTC |
| f08df9e | Brandon Philips | 19 November 2014, 20:13:08 UTC | Merge pull request #1753 from philips/fixup-build-docker scripts: build-docker tag and use ENTRYPOINT | 19 November 2014, 20:13:08 UTC |
| 00df131 | Brandon Philips | 19 November 2014, 20:11:27 UTC | scripts: build-docker tag and use ENTRYPOINT Use ENTRYPOINT so people can specify flags to etcd without providing the binary. Thanks to @hugod in IRC for pointing this out. | 19 November 2014, 20:11:27 UTC |
| d2e36a9 | Brandon Philips | 19 November 2014, 20:10:06 UTC | scripts: build-release add etcd-migrate this tool is only temporary for the alphas but make sure it makes it in. | 19 November 2014, 20:10:06 UTC |
| 893fb3b | Brandon Philips | 19 November 2014, 00:00:07 UTC | Merge pull request #1747 from proppy/master create .godir | 19 November 2014, 00:00:07 UTC |
| 03bacc1 | Johan Euphrosine | 18 November 2014, 23:01:57 UTC | create .godir | 18 November 2014, 23:01:57 UTC |
| 4c1fd07 | Xiang Li | 18 November 2014, 19:42:08 UTC | raft: optimistically increase the next if the follower is already matched This is useful since we want to pipeline the appendEntry requests. Without enabling optimistic increasing, the second pipelining appendEntry request will include the entries the first one has already sent out. We decrease the next directly to match if the leader receives a rejection for a matched follower. This happens if one pipelining request get lost and following ones arrives at the follower. | 18 November 2014, 21:41:38 UTC |
| e07ef69 | Xiang Li | 18 November 2014, 21:25:12 UTC | Merge pull request #1740 from xiang90/handleheartbeat raft: add handleHeartbeat | 18 November 2014, 21:25:12 UTC |
| f94ff96 | Yicheng Qin | 18 November 2014, 18:11:05 UTC | Merge pull request #1739 from yichengq/230 rafthttp: send takes raft message instead of bytes | 18 November 2014, 18:11:05 UTC |
| bd4cfa2 | Xiang Li | 18 November 2014, 00:58:49 UTC | raft: add handleHeartbeat handleHeartbeat commits to the commit index in the message. It never decreases the commit index of the raft state machine. | 18 November 2014, 16:34:06 UTC |
| 1a72143 | Yicheng Qin | 18 November 2014, 00:40:52 UTC | rafthttp: send takes raft message instead of bytes This gives streaming mechanism the chance to assemble and disassemble raft messages. | 18 November 2014, 06:39:53 UTC |
| 6cac631 | Xiang Li | 18 November 2014, 01:14:40 UTC | Merge pull request #1738 from yichengq/231 *: add rafthttp pkg into test list | 18 November 2014, 01:14:40 UTC |
| 04d4162 | Yicheng Qin | 18 November 2014, 00:45:54 UTC | *: add rafthttp pkg into test list | 18 November 2014, 00:45:54 UTC |
| 1635844 | Xiang Li | 18 November 2014, 00:26:07 UTC | Merge pull request #1737 from xiang90/include_commit raft: include commitIndex in heartbeat | 18 November 2014, 00:26:07 UTC |
| b93d87f | Xiang Li | 17 November 2014, 23:44:57 UTC | raft: include commitIndex in heartbeat | 18 November 2014, 00:19:28 UTC |
| a2c568a | Yicheng Qin | 18 November 2014, 00:14:59 UTC | Merge pull request #1669 from yichengq/215 *: add rafthttp as a separate package | 18 November 2014, 00:14:59 UTC |
| f24e214 | Yicheng Qin | 17 November 2014, 22:40:21 UTC | rafthttp: move server stats in raftHandler to etcdserver | 18 November 2014, 00:02:20 UTC |
| 5dc5f81 | Yicheng Qin | 17 November 2014, 00:23:47 UTC | *: etcdhttp.raftHandler -> rafthttp.RaftHandler | 17 November 2014, 23:52:24 UTC |
| 3fcc011 | Yicheng Qin | 16 November 2014, 20:12:57 UTC | etcdserver: rename sender.go -> sendhub.go | 17 November 2014, 23:35:15 UTC |
| 84fbf7a | Yicheng Qin | 16 November 2014, 18:21:05 UTC | *: etcdserver.sender -> rafthttp.Sender | 17 November 2014, 23:35:10 UTC |
| e04e463 | Xiang Li | 17 November 2014, 22:34:59 UTC | Merge pull request #1736 from xiang90/verify etcdserver: refactor ValidateClusterAndAssignIDs | 17 November 2014, 22:34:59 UTC |
| 0541f0a | Xiang Li | 17 November 2014, 22:08:45 UTC | etcdserver: refactor ValidateClusterAndAssignIDs | 17 November 2014, 22:23:37 UTC |
| 144db79 | Xiang Li | 17 November 2014, 21:35:26 UTC | Merge pull request #1729 from xiang90/test_rm integration: add integration test for remove member | 17 November 2014, 21:35:26 UTC |
| c26de66 | Xiang Li | 16 November 2014, 05:28:03 UTC | integration: add integration test for remove member | 17 November 2014, 21:28:09 UTC |
| ee8fbee | Xiang Li | 17 November 2014, 21:19:59 UTC | Merge pull request #1733 from xiang90/config integration: add test for member restart | 17 November 2014, 21:19:59 UTC |
| 800747e | Xiang Li | 17 November 2014, 19:21:16 UTC | integration: add test for member restart | 17 November 2014, 19:28:18 UTC |
| a817ca7 | Xiang Li | 15 November 2014, 00:13:32 UTC | Merge pull request #1726 from xiang90/fix_sender etcdserver: add buffer to the sender queue | 15 November 2014, 00:13:32 UTC |
| fd512ff | Yicheng Qin | 14 November 2014, 23:49:29 UTC | version: bump to alpha.3 | 14 November 2014, 23:51:01 UTC |
| 7c4b84a | Xiang Li | 14 November 2014, 23:18:16 UTC | etcdserver: add buffer to the sender queue | 14 November 2014, 23:18:16 UTC |
| ac5a282 | Xiang Li | 14 November 2014, 22:16:39 UTC | Merge pull request #1720 from xiang90/sender_stop *: gracefully stop etcdserver | 14 November 2014, 22:16:39 UTC |
| 8bf71d7 | Xiang Li | 14 November 2014, 07:03:34 UTC | *: gracefully stop etcdserver | 14 November 2014, 22:12:24 UTC |
| 4e251f8 | Barak Michener | 14 November 2014, 21:58:22 UTC | Merge pull request #1578 from barakmich/bcm_migrate migrate: Add a migration tool to go from etcd v0.4 -> v0.5 data directories | 14 November 2014, 21:58:22 UTC |
| 192f200 | Barak Michener | 29 October 2014, 20:27:07 UTC | Fix up migration tool, add snapshot migration Fixes all updates since bcwaldon sketched the original, with cleanup and into an acutal working state. The commit log follows: fix pb reference and remove unused file post rebase unbreak the migrate folder correctly detect node IDs fix snapshotting Fix previous broken snapshot Add raft log entries to the translation; fix test for all timezones. (Still in progress, but passing) Fix etcd:join and etcd:remove print more data when dumping the log Cleanup based on yichengq's comments more comments Fix the commited index based on the snapshot, if one exists detect nodeIDs from snapshot add initial tool documentation and match the semantics in the build script and main formalize migration doc rename function and clarify docs fix nil pointer fix the record conversion test add migration to test suite and fix govet | 14 November 2014, 21:46:08 UTC |
| 5ea1f2d | Brian Waldon | 07 October 2014, 22:06:51 UTC | etcd4: migration from v0.4 -> v0.5 | 14 November 2014, 20:57:26 UTC |
| c36abea | Brian Waldon | 08 October 2014, 01:34:15 UTC | etcdserver: export Member.StoreKey | 14 November 2014, 20:57:26 UTC |
| b6887e4 | Yicheng Qin | 14 November 2014, 20:18:41 UTC | Merge pull request #1719 from yichengq/228 etcdserver: recover snapshot before applying committed entries | 14 November 2014, 20:18:41 UTC |
| 77433ff | Yicheng Qin | 14 November 2014, 19:45:03 UTC | etcdserver: recover cluster when receiving newer snapshot | 14 November 2014, 20:11:21 UTC |
| dfaa729 | Yicheng Qin | 14 November 2014, 19:14:06 UTC | etcdserver: check and select committed entries to apply | 14 November 2014, 20:11:16 UTC |
| f6a7f96 | Yicheng Qin | 14 November 2014, 08:10:46 UTC | etcdserver: recover from snapshot before applying requests | 14 November 2014, 20:08:39 UTC |
| 7d0ffb3 | Yicheng Qin | 14 November 2014, 08:09:54 UTC | raft: not set applied when restored from snapshot applied is only updated by application level through Advance. | 14 November 2014, 20:08:39 UTC |
| 1f71988 | Xiang Li | 14 November 2014, 16:39:08 UTC | Merge pull request #1703 from jonboulle/proc *: fix Procfile | 14 November 2014, 16:39:08 UTC |
| 6f7fd89 | Xiang Li | 14 November 2014, 05:29:02 UTC | Merge pull request #1718 from yichengq/226 sender: support elegant stop | 14 November 2014, 05:29:02 UTC |
| 12dba7d | Yicheng Qin | 14 November 2014, 01:06:21 UTC | sender: support elegant stop | 14 November 2014, 01:44:36 UTC |
| e66bda9 | Xiang Li | 13 November 2014, 23:16:52 UTC | Merge pull request #1714 from xiang90/stop StopNotify | 13 November 2014, 23:16:52 UTC |
| 6a1fe00 | Xiang Li | 13 November 2014, 22:35:50 UTC | Merge pull request #1704 from xiang90/print_config *: print out configuration when necessary | 13 November 2014, 22:35:50 UTC |
| 11f392b | Yicheng Qin | 13 November 2014, 22:30:40 UTC | Merge pull request #1708 from yichengq/223 etcdserver: validate new node is not registered before in best effort | 13 November 2014, 22:30:40 UTC |
| b5d480f | Xiang Li | 13 November 2014, 22:16:48 UTC | etcdserver: add StopNotify | 13 November 2014, 22:16:48 UTC |
| 978d0f1 | Xiang Li | 13 November 2014, 22:08:59 UTC | etcdserver: fix TestDoProposalStopped test We start etcd server in this test without the cluster. Sometimes it panics when accessing the cluster. Most of the time it does not panic, since we can stop the server fast enough before applying the first configuration change entry. | 13 November 2014, 22:08:59 UTC |
| fb344bc | Xiang Li | 13 November 2014, 22:01:56 UTC | etcdserver: minor cleanup | 13 November 2014, 22:01:56 UTC |
| 813ff6b | Xiang Li | 13 November 2014, 21:58:07 UTC | Merge pull request #1713 from xiang90/stop etcdserver: fix server.Stop() | 13 November 2014, 21:58:07 UTC |
| ac907d7 | Yicheng Qin | 13 November 2014, 17:25:21 UTC | etcdserver: validate new node is not registered before in best effort | 13 November 2014, 21:56:11 UTC |
| 30dfdb0 | Xiang Li | 13 November 2014, 21:47:12 UTC | etcdserver: fix server.Stop() Stop should be idempotent. It should simply send a stop signal to the server. It is the server's responsibility to stop the go-routines and related components. | 13 November 2014, 21:47:12 UTC |
