97453a618b
*: rename kube-etcd-controller -> etcd-operator
2016-10-31 10:39:57 -07:00
ad86da1639
e2e: add disaster recovery test for all pods down
2016-10-28 16:57:10 -07:00
277adc3de5
k8s: delete pod with 0 grace period
2016-10-28 11:41:35 -07:00
fc823816ae
e2e: fix waitBackupPodUp
2016-10-28 11:21:57 -07:00
9851b8bbc8
k8sutil: add const BackupPodSelectorAppField
2016-10-28 10:57:52 -07:00
74d7e77952
e2e: add waitBackupPodUp
2016-10-28 10:48:42 -07:00
35b2ff98f8
e2e: add TODO comment for member recovery race
2016-10-27 19:46:39 -07:00
2002cb15f5
e2e: add etcd pods info and failed pod logs
2016-10-26 19:41:54 -07:00
aad86a3b7e
remove readiness probe and fix flake
...
We are removing readiness probe because there is a loop problem
- readiness failed due to etcd couldn't
bootstrap
- etcd couldn't bootstrap due to it can't communicate with
others
- it can't communicate with others due to readiness failed.
Originally we added readiness to reduce flakes. It isn't necessary
since our reconcile loop can now retry on common temporary errors.
We uncovered the flakes are caused by network setup problem.
It took a while and etcd exited due to error like:
```
dial tcp: lookup ...: dial udp 10.0.0.10:53: i/o timeout
```
We just sleep a few seconds in the container before starting server.
2016-10-26 19:38:40 -07:00
318411cf8f
minor refactor
2016-10-20 14:54:25 -07:00
59df7767cc
add container option in getLogs
2016-10-13 20:49:26 -07:00
eca23460d7
*: add License header
2016-10-13 16:19:41 -07:00
f6c1a9e7e1
e2e: refactor makeEtcdCluster
2016-10-11 15:33:38 -07:00
50857c8dff
manage etcd cluster upgrade
2016-10-11 15:12:54 -07:00
25180631e6
add CleanupBackupIfDeleted backup policy option
2016-10-08 08:39:56 -07:00
f2affa2068
create spec/ pkg and move all specs into it
2016-10-05 15:19:44 -07:00
a51f859414
more info at the end of each e2e test
2016-10-04 22:30:01 -07:00
083da277a7
use etcdctl and readiness probe in each etcd pod
2016-10-04 13:25:00 -07:00
7c07bc1b49
e2e: print kube-etcd-controller logs
2016-10-02 10:21:08 -07:00
bed78cc116
use persistent storage in backup pod
2016-09-29 20:35:53 -07:00
e9b5cf4449
e2e: minor logging change
2016-09-26 15:35:47 -07:00
2ef6a22087
e2e: add TestDisasterRecovery
2016-09-26 15:20:12 -07:00
2b558e0b0f
e2e: add TestOneMemberRecovery
2016-09-26 14:18:25 -07:00
5b73693c6e
e2e: add test TestResizeCluster5to3
2016-09-25 09:50:27 -07:00
0624541c9c
code cleanup
2016-09-24 10:54:31 -07:00
a80cddf15d
e2e: new test TestResizeCluster3to5
2016-09-23 18:03:06 -07:00
91f009e333
e2e: support unique namespace each test ( #135 )
...
* e2e: support unique namespace each test
* use downward api to convey namespace
2016-09-23 16:09:00 -07:00
063a47c5b5
e2e test
2016-09-14 13:27:08 -07:00
Xiang Li
Hongchao Deng
Hongchao Deng