50f689f596
Merge pull request #34128 from spxtr/nodeinfo
...
Automatic merge from submit-queue
Do a kubectl get nodes after turning up the e2e cluster.
https://github.com/kubernetes/test-infra/issues/366
2016-10-13 05:37:02 -07:00
1cfbb446c8
Try to bring down cluster in case of failure
2016-10-12 10:54:04 +02:00
da786a8dc5
hack/e2e.go: listen for Interrupt signal and attempt cleanup if using --down
...
Signed-off-by: Jess Frazelle <me@jessfraz.com>
Signed-off-by: Jess Frazelle <acidburn@google.com>
2016-10-07 13:06:54 -07:00
d5e0b1cf84
Dump node info after cluster up.
2016-10-05 12:46:43 -07:00
ca58119af7
e2e.go / kops: Add --kops-kubernetes-version
...
This adds a flag to override the version we're deploying with kops for
e2es. (The kops experience is disconnected from the actual source tree
we're running tests from, which is similar to GKE.)
2016-10-04 14:01:31 -07:00
eb44add29c
e2e.go / kops: Add ssh-key option to override ssh key
...
By default, Jenkins stuffs the ssh key in
/workspace/.aws/kube_aws_rsa. Allow this to be overridden easily on
the command line.
2016-10-03 17:50:15 -07:00
547e152932
Make KubemarkTest not have XML output.
2016-09-27 16:22:23 -07:00
d905478e0a
e2e.go: Add -deployment, add a kops deployment method
...
This splits off all the bash stuff into an interface, and plumbs
through a separate interface to bring up a cluster using "kops"
instead. Right now it assumes kops == AWS.
2016-09-27 14:45:19 -07:00
52cfe25351
Un-xmlWrap the Test function.
2016-09-26 10:59:38 -07:00
3503c4f71a
Make e2e.go give us JUnit results.
2016-09-22 18:40:53 -07:00
4acebfc243
Revert "e2e: only tear down existing cluster if it exists"
2016-09-12 13:53:35 -07:00
af9a922d6e
e2e: only tear down existing cluster if it exists
2016-09-09 15:26:47 -07:00
08b6eaff92
Merge pull request #30321 from fejta/check-resources
...
Automatic merge from submit-queue
Convert bool to error, helper func for cd to skew
Switch from various functions returning `False` when something bad happens to returning an `error`
Encapsulate logic to switch to the skew directory inside chdirSkew
Also add a TODO for using `hyphen-flags` instead of `underscore_flags`
2016-09-09 00:48:19 -07:00
fd15911910
Convert bool to error, helper func for cd to skew
2016-08-29 10:15:57 -07:00
d8a81155c8
Only run federated test if FEDERATION == "true"
2016-08-23 13:14:58 -07:00
65233e4bf2
Merge pull request #30575 from soltysh/scheduledjobs_e2e
...
Automatic merge from submit-queue
Scheduledjobs e2e
@janetkuo resubmitted e2e for SJ, I've updated all scripts to consume `KUBE_RUNTIME_CONFIG` properly in 2nd commit, ptal
2016-08-19 23:42:32 -07:00
5e421cc447
Fix resource leak check in case of run that cleans up after failed one
2016-08-16 17:14:30 +02:00
fdf87d6015
Enable batch/v2alpha1 globally for e2e tests
2016-08-14 14:24:49 +02:00
f64c052858
Revert "Scheduledjob e2e"
2016-08-12 16:12:19 -07:00
ac1f8bc665
Merge pull request #26027 from soltysh/scheduledjob_e2e
...
Automatic merge from submit-queue
Scheduledjob e2e
@erictune last element of the scheduledjob puzzle. I think we'll iterate on this once we have all the puzzles in place. This is one of those things that will be allowed to merge after code freeze.
```release-note
* Introducing ScheduledJobs as described in [the proposal](https://github.com/kubernetes/kubernetes/blob/master/docs/proposals/scheduledjob.md ) as part of `batch/v2alpha1` version (experimental feature).
```
[]()
2016-08-12 13:27:32 -07:00
08fa3e8e3f
Enable batch/v2alpha1 globally for e2e tests
2016-08-11 13:06:39 +02:00
9c70b01b2a
hack/e2e.go: Make --dump run if --up fails, and if --down isn't specified
...
See https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/logs/kubernetes-e2e-aws/629 for why this would be useful.
(And the requirement for --down is just unnecessary AFAICT.)
2016-08-10 15:09:57 -07:00
7011d8826e
Move kubemark, skew, upgrade, leak logic into hack/e2e.go
2016-08-09 14:04:19 -07:00
be5e98fafd
Merge pull request #28959 from asalkeld/e2e-isup
...
Automatic merge from submit-queue
e2e.go --isup should also check the version
Just move the setenv up to include --isup
fixes #18667
2016-07-22 01:59:18 -07:00
c342d8fc12
Only run federation tests if FEDERATION==true.
...
This is the same test we make everywhere else. Only here were we
checking against the empty string.
2016-07-15 16:31:27 -07:00
5c3eea15eb
e2e.go --isup should also check the version
...
fixes #18667
2016-07-14 16:20:30 +02:00
ef0c9f0c5b
Remove "All rights reserved" from all the headers.
2016-06-29 17:47:36 -07:00
a092fd2dd4
Federation build and e2e test integration
...
Federation components are now buildable and e2e-testable via e2e.go.
2016-05-25 15:07:27 -07:00
7e9cae85fe
Add a flag that lets e2e tests be run against single-node clusters.
2016-05-17 09:18:20 -07:00
d70c516d4f
Remove log collection code in cluster/gce/util.sh.
...
Also update some docs to mention cluster/log-dump.sh.
2016-03-03 15:28:33 -08:00
930291588c
Remove PrepareVersion from hack/e2e.go.
2016-02-24 13:08:22 -08:00
2b756ca56d
Move E2E kube-up call into test-setup from e2e-internal.
2016-02-17 16:49:07 -08:00
98981ac3ec
Remove unused functions from hack/e2e.go.
2016-02-17 11:23:33 -08:00
2874f30c05
Enable Deployments by default
2016-02-12 12:29:14 -08:00
e67667f1b5
Minion->Node rename: comments and vars for e2e.go and e2e.sh
2015-12-01 16:00:47 -08:00
d61ed52799
Adding an e2e test for deployment
2015-09-25 17:14:31 -07:00
170962becb
Explicitly log how long each step of e2e setup/test/teardown takes.
2015-09-15 15:04:05 -07:00
9f77e49109
typofix - https://github.com/vlajos/misspell_fixer
2015-08-08 22:31:48 +01:00
ecce796ee8
Use Fatalf
2015-05-10 13:08:55 +09:00
129d7f8d26
Fix typo in e2e
...
Correct comment in e2e.go ’…at least 3 minions running’ -> ‘…at least minMinionCount
minions running’.
2015-05-04 04:42:07 -04:00
6b3a6e6b98
Make copyright ownership statement generic
...
Instead of saying "Google Inc." (which is not always correct) say "The
Kubernetes Authors", which is generic.
2015-05-01 17:49:56 -04:00
f9a3315367
Remove useless e2e noise
2015-04-07 17:52:32 -07:00
ef801875ab
Add --test_args flag to hack/e2e.go.
...
Add a flag to allow passing options to the Ginkgo test runner, for
example to pass --ginkgo.focus to run only particular specs.
2015-02-17 17:40:32 -08:00
70d37800b7
Simplify usage of os/exec
...
For now, keep the finishRunning() wrapper but use a straight cmd.Run()
call instead of the convoluted goroutine trying to catch signals.
It turns out that Unix process group handling is enough to interrupt
pending processes when stopping the run with something like a Ctrl+C
which should be enough.
Tested:
- Full e2e run with hack/e2e-test.sh, two tests failed but looks like
they've been failing before this change.
- Started a hack/e2e.go -v -build and interrupted it with Ctrl+C,
confirmed that build-release.sh was killed in the process.
2015-02-13 14:53:42 -08:00
43543faaa7
Simplify ValidateClusterSize() to use cmd.Output() directly
...
This is a first step towards getting rid of finishRunningWithOutputs and
using the native os/exec methods directly where possible.
2015-02-11 15:02:24 -08:00
55e8357c0f
Merge pull request #4339 from zmerlynn/issue_4177
...
Fix hack/e2e.go exit status for -test
2015-02-11 13:21:37 -08:00
f08f7e6972
Clean up output handling in hack/e2e.go
...
Dovetail stdout/stderr and remove escaping for TAP.
2015-02-11 13:12:48 -08:00
ab38a78d70
Test() was mis-tracked. Change the variable name / handling here, and
...
also just make it unset so that if it's ever uninitialized it's
obvious.
Fixes #4177
2015-02-11 13:11:03 -08:00
9257b58e68
Fix e2e-internal scripts for cluster-size and watch-events
...
Fixes #4340
2015-02-11 12:46:31 -08:00
162d7a00b3
Fix hack/e2e.go --ctl output
...
Export the KUBE_CONFIG_FILE environment variable pointing to the test
output when calling cluster/kubectl.sh directly.
Tested by running:
$ go run ./hack/e2e.go -v -ctl='version --match-server-version=false'
2015-02-06 14:36:35 -08:00
Kubernetes Submit Queue