github/kubernetes
1
0
Fork 0
mirror of https://github.com/k3s-io/kubernetes.git synced 2025-10-31 13:50:01 +00:00
Code Issues Projects Releases Wiki Activity
23,944 Commits 42 Branches 8,159 Tags
a6d96a04d05f546b16b4199bbc8aeda6102aedad
Commit Graph

500 Commits

Author SHA1 Message Date
Eric Tune
f5e6161e49 Merge pull request #11298 from mesosphere/fix-10776 
Fix deadlocks and race conditions in mesos master election notifier
 2015-07-15 13:55:17 -07:00
Dr. Stefan Schimanski
e98c8e7685 Fix deadlocks and race conditions in mesos master election notifier 
- n.node used the n.lock as underlaying locker. The service loop initially
  locked it, the Notify function tried to lock it before calling n.node.Signal,
  leading to a dead-lock.
- the go routine calling ChangeMaster was not synchronized with the Notify
  method. The former was triggering change events that the later never saw
  when the former's startup was faster that of Notify. Hence, not even a single
  event was noticed and not even a single start/stop call of the slow service
  was triggered.

This patch replaces the n.node condition object with a simple channel n.changed.
The service loop watches it.

Updating the notified private variables is still protected with n.lock against
races, but independently of the n.changed channel. Hence, the deadlock is gone.

Moreover, the startup of the Notify loop is synchronized with the go routine which
changes the master. Hence, the Notify loop will see the master changes.

Fixes #10776
 2015-07-15 21:45:53 +02:00
Eric Tune
3dad5a0652 Merge pull request #10835 from mesosphere/mesos-root-ca-file 
Add --root-ca-key code to Mesos controller-manager fork
 2015-07-14 12:16:49 -07:00
Dr. Stefan Schimanski
95c7dc8cb3 Re-enable mesos scheduler TestPlugin_LifeCycle test 2015-07-13 22:43:16 +02:00
Dr. Stefan Schimanski
143cf4b08d Use correct offer's hostname of test pods in mesos scheduler plugin tests 2015-07-13 22:41:23 +02:00
Dr. Stefan Schimanski
dd7345b25f Fix offer+pod races in mesos scheduler plugin test 
- Offers were reused and led to unexpected declining by the scheduler because
  the reused offer did not get a new expiration time.
- Pod scheduling and offer creation was not synchronized. When scheduling
  happened after aging of offers, the first issue was trigger. Because
  the mesos driver DeclineOffer was not mocked this lead to a test error.
 2015-07-13 22:41:23 +02:00
Dr. Stefan Schimanski
bf44f5df28 Add DeclineOffer return value to mock driver in mesos scheduler test 
Depending on timing the mesos scheduler might call DeclineOffer:

The default ttl of an offer in mesos scheduler is 5sec. If the tests run longer,
the old, unused offers are declined, leading to an mock error.

Probably fixes GoogleCloudPlatform/kubernetes#10795
 2015-07-13 22:41:23 +02:00
David Oppenheimer
089a703194 Disable TestPlugin_LifeCycle due to flakiness. 2015-07-10 22:14:16 -07:00
Dr. Stefan Schimanski
39b3af0fdc Add --root-ca-key code to Mesos' controller-manager fork 2015-07-07 18:19:47 +02:00
nikhiljindal
274792d7bb Stop exposing v1beta3 by default 2015-07-01 14:38:02 -07:00
Clayton Coleman
d8bb4552de Cloud provider should return an error 
Not fatal - makes cloud provider useful in methods that
can return error.
 2015-07-01 14:41:49 -04:00
Maxwell Forbes
712f303350 Merge pull request #9736 from sdminonne/bug_fix2 
To add validation for service ports when defined as string
 2015-06-25 19:37:04 -07:00
Maxwell Forbes
655179dcfb Merge pull request #10264 from mikedanese/ca-token 
add ca cert to token controller and all service accounts
 2015-06-25 09:56:35 -07:00
Maxwell Forbes
3afda5d566 Merge pull request #10312 from dchen1107/cleanup 
Take 2: Fix the race between configuring cbr0 and restarting static pods
 2015-06-24 17:59:50 -07:00
Maxwell Forbes
28946766a3 Merge pull request #9807 from krousey/container_manifest 
Removing ContainerManifest
 2015-06-24 17:55:29 -07:00
Mike Danese
56bde3342a add ca to token controller and all service accounts 2015-06-24 15:10:20 -07:00
Kris Rousey
d13421e084 Removing ContainerManifest 2015-06-24 11:31:34 -07:00
Dawn Chen
6ddfa512de Revert "Revert "Fix the race between configuring cbr0 and restarting static pods"" 
This reverts commit fd0a95dd12.
 2015-06-24 11:10:10 -07:00
Piotr Szczesniak
fd0a95dd12 Revert "Fix the race between configuring cbr0 and restarting static pods" 2015-06-24 09:56:49 +02:00
Jeff Lowdermilk
50d50a3cb8 Merge pull request #10211 from dchen1107/cleanup 
Fix the race between configuring cbr0 and restarting static pods
 2015-06-23 17:09:01 -07:00
Dawn Chen
23200d303f Fix several issues on running syncPods until network is configured. 
Also fixed unittests and compiling.
 2015-06-23 12:11:19 -07:00
Dr. Stefan Schimanski
9e0c9b4f5a Mesos: create static pod file source only for configured static pods 
The file source was created even when no static pods were configured.
In this case it was never marked as seen. As a consequence the kubelet
syncPods functions never deleted pods because it was too cautious due
an unseen pod source, leading to leaked pods.
 2015-06-23 12:25:21 +02:00
Jeff Lowdermilk
0d7de0991e Disable TestProc_doWithNestedXConcurrent 
This test is killing more than 50% of shippable builds. Disabling
to stop the madness.
 2015-06-22 15:39:43 -07:00
Salvatore Dario Minonne
4b13faa346 To add validation for service ports when defined as string (fixing issue #9734) 2015-06-22 17:21:51 +02:00
Justin Santa Barbara
97cafd20f6 NodeName != HostName: Fixes for contrib/mesos 2015-06-18 12:40:14 -07:00
Satnam Singh
e4f5529a2d Revert "Allow nodename to be != hostname, use AWS instance ID on AWS" 2015-06-18 11:27:55 -07:00
Justin Santa Barbara
77e1bd3f56 NodeName != HostName: Fixes for contrib/mesos 2015-06-17 00:40:43 -04:00
Dr. Stefan Schimanski
7abe12d6f4 Fix flaky mesos executor test 
The TestExecutorFrameworkMessage test sends a "task-lost:foo" message to the
executor in order to mark a pod as lost. For that the pod must be running first.
Otherwise, the executor code will send "TASK_FAILED" status updates, not "TASK_LOST".

Before this patch there was no synchronization between the pod startup and the
test case. Moreover, in order to startup a task a working apiserver URL must be
passed to the executor which was not the case either.

Fixes mesosphere/kubernetes-mesos#351
 2015-06-16 09:08:23 +02:00
Fabio Yeon
241e87cf9b Merge pull request #9077 from mesosphere/staticPodsUpstream 
Add static pod support to mesos scheduler and executor.
 2015-06-15 15:20:33 -07:00
Fabio Yeon
da02e3059a Merge pull request #9789 from mesosphere/plugin-test-race 
Fix mesos plugin-test race
 2015-06-15 13:04:12 -07:00
Joerg Schad
7af8bf6ed3 Add static pod support to mesos scheduler and executor. 
- the mesos scheduler gets a --static-pods-config parameter with a directory with
  pods specs. They are zipped and sent over to newly started mesos executors.
- the mesos executor receives the zipper static pod config via ExecutorInfo.Data
  and starts up the pods via the kubelet FileSource mechanism.
- both - the scheduler and the executor side - are fully unit tested
 2015-06-15 17:24:54 +00:00
Fabio Yeon
65ea6b914c Merge pull request #9785 from mesosphere/longer-timeout-for-proc-test 
increase timeout for nested concurrent test...
 2015-06-15 10:11:51 -07:00
Fabio Yeon
3dcf97a07c Merge pull request #9741 from mesosphere/fix-k8sm-342 
Fix kubernetes-mesos issue 342: pod readiness bug
 2015-06-15 10:11:06 -07:00
Fabio Yeon
18a3932585 Merge pull request #9265 from mesosphere/upstream-k8sm-cm-and-km 
Upstream Kubernetes Mesos controller manager and km binary
 2015-06-15 09:28:12 -07:00
James DeFelice
f956f23d37 move the container name fix to NewTestPod to avoid breakage in other unit tests in the future 2015-06-15 14:30:48 +00:00
Dr. Stefan Schimanski
615e4f99a7 Fix mesos plugin-test race 
ListWatch returned the internal list of the MockPodsListWatch object, leading
to a race. Fixes travis builds.

Fixes mesosphere/kubernetes-mesos#354
 2015-06-15 08:18:57 +02:00
James DeFelice
9f53405d42 increase timeout for nested concurrent test that is still intermittently timing out in travis 2015-06-15 02:25:27 +00:00
James DeFelice
5e4992aa5e fix broken TestExecutorLaunchAndKillTask unit test 2015-06-14 03:38:29 +00:00
James DeFelice
eca5a2ac88 lower container poll freq to something more sane 2015-06-13 22:23:43 +00:00
James DeFelice
c0b686e3be resolve pod readiness issue 
see https://github.com/mesosphere/kubernetes-mesos/issues/342
xref #9710
 2015-06-13 22:23:43 +00:00
David Oppenheimer
5e5c1d1097 Merge pull request #9767 from mesosphere/skip-exec-test-flake 
skip flakey unit test for now
 2015-06-13 14:31:21 -07:00
James DeFelice
270a4875b0 fix intermittent deadlock w/ nested concurrent action scheduling and 
give each delegate its own errOnce
 2015-06-13 15:19:01 +00:00
James DeFelice
b8b91fd169 skip flakey unit test for now 2015-06-13 04:25:19 +00:00
Mike Danese
5896ac9e07 skip GetHostIP call on standalone mode kubelet to fix spammy log statement 2015-06-12 11:37:53 -07:00
James DeFelice
1820114a2d Upstream controller manager and km binary 
Closes mesosphere/kubernetes-mesos#310
Depends on GoogleCloudPlatform/kubernetes#8882

- fix https://github.com/mesosphere/kubernetes-mesos/issues/336
- Fix comment typo
- Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/9265#commitcomment-11559038
- Add warning to k8s modules to also update mesos copies
- Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/9265#commitcomment-11558864
- Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/9265#commitcomment-11558855
- Add comments and TODO that hypercube and controllermanager need refactoring
 2015-06-12 14:03:00 +00:00
James DeFelice
6436c4a3bc additional comments as per code review 2015-06-11 13:47:14 +00:00
James DeFelice
768759d6f1 added km and k8sm-controller-manager binaries from PR#9265 2015-06-11 13:10:01 +00:00
James DeFelice
ee309f3cff add TODOs 2015-06-11 12:41:50 +00:00
James DeFelice
f445ead2da use common mesos_ prefix for metrics 2015-06-11 12:40:12 +00:00
James DeFelice
932c58a497 Kubernetes Mesos integration 
This commit includes the fundamental components of the Kubernetes Mesos
integration:

* Kubernetes-Mesos scheduler
* Kubernetes-Mesos executor
* Supporting libs

Dependencies and upstream changes are included in a separate commit for easy
review.

After this initial upstream, there'll be two PRs following.

* km (hypercube) and k8sm-controller-manager #9265
* Static pods support #9077

Fixes applied:

- Precise metrics subsystems definitions
  -  mesosphere/kubernetes-mesos#331
  - https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion_r31875232
  - https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion_r31875240
- Improve comments and add clarifications
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875208
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875226
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875227
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875228
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875239
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875243
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875234
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875256
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875255
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875251
- Clarify which Schedule function is actually called
  - Fixes https://github.com/GoogleCloudPlatform/kubernetes/pull/8882#discussion-diff-31875246
 2015-06-10 20:58:39 +00:00