Mirroristas/etcd
2
0
Fork 0
mirror of https://github.com/etcd-io/etcd.git synced 2024-09-27 06:25:44 +00:00
Code Issues Packages Projects Releases Wiki Activity
8,286 Commits 15 Branches 491 Tags 148 MiB
Commit Graph

1153 Commits

Author SHA1 Message Date
Xiang Li
6b23a8131f *: test gofmt with -s and fix reported issues 2015-08-21 18:52:16 -07:00
Yicheng Qin
8c0610d4f5 Merge pull request #3352 from yichengq/fix-name-url 
fix that etcd fails to start if using both IP and hostname when discovery srv
 2015-08-21 12:38:38 -07:00
Yicheng Qin
72462a72fb etcdserver: remove TODO to delete URLStringsEqual 
Discovery SRV supports to compare IP addresses with domain names,
so we need URLStringsEqual function.
 2015-08-21 09:52:17 -07:00
Yicheng Qin
8ea3d157c5 Revert "Revert "Treat URLs have same IP address as same"" 
This reverts commit 3153e635d5799f1ff8715fd4b7c07e0283820acd.

Conflicts:
	etcdserver/config.go
 2015-08-21 09:41:13 -07:00
Xiang Li
11a689d063 etcdserver/auth: cache auth enable result 2015-08-20 23:05:00 -07:00
Yicheng Qin
bcb4d5d53e Merge pull request #3311 from yichengq/request-timeout 
extend hardcoded timeout for globally-deployed etcd cluster
 2015-08-17 17:00:24 -07:00
Yicheng Qin
1375ef8985 etcdserver: remove getVersion timeout 
The request can still time out because we have set dial timeout and
read/write timeout. It increases timeout expectation from 1s to 5s,
but it makes it workable in globally-deployer cluster.
 2015-08-17 16:50:40 -07:00
Xiang Li
d487cf6b63 etcdhttp:write etcderror for all errors in keyhandler 2015-08-17 15:51:29 -07:00
Yicheng Qin
c530385d6d Merge pull request #3313 from yichengq/internal-timeout 
etcdserver: use ReqTimeout only
 2015-08-17 15:05:46 -07:00
Xiang Li
af6d1d3d95 Merge pull request #3310 from xiang90/http_err 
*: key handler should write auth error as etcd error
 2015-08-17 14:57:19 -07:00
Yicheng Qin
2d5b95c49f etcdserver: use ReqTimeout only 
We cannot refer RTT value from heartbeat interval, so CommitTimeout
is invalid. Remove it and use ReqTimeout instead.
 2015-08-17 14:54:25 -07:00
Xiang Li
87f061bab2 *: key handler should write auth error as etcd error 2015-08-17 14:45:45 -07:00
Xiang Li
15e03d801f etcdserver: add version enforcement when setting cluster version 2015-08-17 11:12:39 -07:00
Xiang Li
f199a484af *: only print out major.minor version for cluster version 2015-08-15 08:30:06 -07:00
Xiang Li
bbcb38189c Merge pull request #3302 from xiang90/v 
etcdserver: better version detection log output
 2015-08-14 16:14:55 -07:00
Xiang Li
0076ab154b etcdserver: better version detection log output 
Fix https://github.com/coreos/etcd/issues/3288
 2015-08-14 16:08:33 -07:00
Xiang Li
dd56b7e05e Merge pull request #3299 from xiang90/txn 
initial support for txn
 2015-08-14 16:05:16 -07:00
Xiang Li
9233fff48f etcdserver: support txn 2015-08-14 11:45:31 -07:00
Xiang Li
46865fa5a5 etcdserverpb: update proto 2015-08-14 11:45:07 -07:00
Yicheng Qin
c229e6e655 etcdserver: improve error message when timeout due to leader fail 2015-08-13 15:46:21 -07:00
Yicheng Qin
ceb27b1c48 etcdhttp: add auth capability in 2.2 2015-08-13 14:49:10 -07:00
Yicheng Qin
0fdb77aea2 etcdserver: go back to marshal request in 2.1 way 
It fixes the problem that 2.1 cannot roll upgrade to 2.2 smoothly
because 2.1 cannot understand the bytes marshalled at 2.2.
 2015-08-13 13:41:52 -07:00
Yicheng Qin
27170e67b9 etcdserver: specify timeout caused by leader election 
Before this PR, the timeout caused by leader election returns:

```
14:45:37 etcd2 | 2015-08-12 14:45:37.786349 E | etcdhttp: got unexpected
response error (etcdserver: request timed out)
```

After this PR:

```
15:52:54 etcd1 | 2015-08-12 15:52:54.389523 E | etcdhttp: etcdserver:
request timed out, possibly due to leader down
```
 2015-08-12 16:53:18 -07:00
Yicheng Qin
c3d4d11402 etcdhttp: adjust request timeout based on config 
It uses heartbeat interval and election timeout to estimate the
expected request timeout.

This PR helps etcd survive under high roundtrip-time environment,
e.g., globally-deployed cluster.
 2015-08-12 09:22:59 -07:00
Yicheng Qin
5a91937367 etcdserver: adjust commit timeout based on config 
It uses heartbeat interval and election timeout to estimate the
commit timeout for internal requests.

This PR helps etcd survive under high roundtrip-time environment,
e.g., globally-deployed cluster.
 2015-08-11 21:09:03 -07:00
Xiang Li
a718329ad3 Merge pull request #3248 from xiang90/v3 
initial v3 demo
 2015-08-10 13:59:03 -07:00
Brandon Philips
fb1951204c etcdserver: move atomics to make etcd work on arm64 
Follow the simple rule in the atomic package:

"On both ARM and x86-32, it is the caller's responsibility to arrange
for 64-bit alignment of 64-bit words accessed atomically. The first word
in a global variable or in an allocated struct or slice can be relied
upon to be 64-bit aligned."

Tested on a system with /proc/cpuinfo reporting:

processor       : 0
model name      : ARMv7 Processor rev 1 (v7l)
Features        : swp half thumb fastmult vfp edsp thumbee neon vfpv3
tls vfpv4 idiva idivt vfpd32 lpae evtstrm
CPU implementer : 0x41
CPU architecture: 7
CPU variant     : 0x0
CPU part        : 0xc0d
CPU revision    : 1
 2015-08-08 18:11:41 -07:00
Xiang Li
9ff7075ce8 etcdserver: use v3server interface 2015-08-08 10:39:04 -07:00
Xiang Li
f004b4dac7 *: etcdserver supports v3 demo 2015-08-08 05:58:29 -07:00
Xiang Li
82afadbcc6 etcdserverpb: update proto 2015-08-08 05:31:35 -07:00
Xiang Li
845c51fedd *: fix typos vaild->valid 2015-08-07 10:57:11 -07:00
Yicheng Qin
f03f048232 Merge pull request #3184 from yichengq/fast-bootstrap 
etcdserver: tick ElectionTicks before starting when bootstrap new cluster
 2015-08-06 15:54:40 -07:00
Yicheng Qin
21f5b885f2 etcdserver: fast election timeout when bootstrap cluster 
The behavior accelarates the happen of the first-time leader election,
so the cluster could elect its leader fast. Technically, it could
help to reduce `electionMs - heartbeatMs` wait time for the first leader election.

Main usage:
1. Quick start for the local cluster when setting a little longer
election timeout
2. Quick start for the global cluster, which sets election timeout to
its maximum 50s.
 2015-08-06 15:44:26 -07:00
Yicheng Qin
a637e86372 Merge pull request #3220 from yichengq/fix-auth-check 
etcdhttp: fix access check for multiple roles in auth
 2015-08-06 15:09:04 -07:00
Xiang Li
58503817ec etcdserver: internal request union 2015-08-05 07:47:10 -07:00
Yicheng Qin
18169e896c etcdhttp: fix access check for multiple roles in auth 
Check access for multiple roles should go through all roles.
 2015-08-04 14:31:07 -07:00
Xiang Li
2b8abeb093 *: remove migration related stuff from 2.2 2015-08-01 19:37:20 +08:00
Barak Michener
dd1a8fe330 etcdhttp: Improve test coverage surrounding auth 2015-07-30 14:21:08 -04:00
Xiang Li
80b794dccc Merge pull request #3185 from xiang90/add_debug_endpoint 
etcdhttp: add config/local/debug endpoint
 2015-07-30 08:46:07 +08:00
Xiang Li
4e31df2c2b etcdhttp: add config/local/log endpoint 
PUT on the endpoint sets the GlobalDebugLevel to json level value.
The action overwrites the origianl log level setting from
users. We need to write doc to warn this.
 2015-07-30 08:35:01 +08:00
Yicheng Qin
6fc9dbfe56 Merge pull request #3114 from yichengq/clean-raft-init 
etcdserver: clean up start and stop logic of raft
 2015-07-27 14:19:25 -07:00
Yicheng Qin
7696dd3280 etcdserver: clean up start and stop logic of raft 
kill TODO and make it more readable.
 2015-07-27 13:24:26 -07:00
Xiang Li
53a77fa519 *: tnx -> txn 2015-07-24 23:21:09 +08:00
Yicheng Qin
b7892b20c1 etcdserver: rename defaultPublishRetryInterval -> defaultPublishTimeout 
This makes code more readable and reasonable.
 2015-07-23 10:09:28 -07:00
Yicheng Qin
5be545b872 Merge pull request #3077 from yichengq/fix-test-sync 
etcdserver: init raft internal var early
 2015-07-10 14:44:52 -07:00
Xiang Li
2fb8347d36 etcdserver: add rpc proto 2015-06-29 20:00:09 -07:00
Xiang Li
581ef05bab *: resolve proto warnings 2015-06-29 18:39:46 -07:00
Xiang Li
13f44e4b79 *: update generated proto code 2015-06-29 16:45:25 -07:00
Yicheng Qin
7f95780bfb etcdserver: init raft internal var early 
Its `stopped`/`done` should be created always before being used
in defer in server loop.

It fixes the race detected when running TestSyncTrigger.
 2015-06-29 15:34:15 -07:00
Yicheng Qin
2e41b4f9e1 etcdserver/auth: fix return value when creating root user 
Before:

```
$ curl http://127.0.0.1:4001/v2/auth/users/root -XPUT -d '{"user": "root",
"password": "root"}'
{"user":"root","roles":null}
```

After:

```
{"user":"root","roles":["root"]}
```
 2015-06-27 23:16:54 -07:00