Automatic merge from submit-queue
Fix fake event recorder race
Event recorder should wait for some time to get all expected events, the event may be written by another goroutine that just have finished.
It should not slow down the test in most cases, only when there is a bug and expected event is not sent.
Fixes#26578
Using P2 to speed up merge and to prevent further flakes.
@kubernetes/sig-storage
Event recorder should wait for some time to get all expected events, the event
may be written by another goroutine that just have finished.
It should not slow down the test in most cases, only when there is a bug and
expected event is not sent.
Automatic merge from submit-queue
retry GetThirdPartyGroupVersions
GetThirdPartyGroupVersions() may return a "NotFound" error if a thirdparty group is deleted in the interim between the group-discovery and the resource-discovery. This is causing e2e flakes in all tests that run kubectl, because test/e2e/thirdparty.go is creating/deleting thirdparty groups.
Fix#26425
The e2e flakes will have the following pattern:
1. the test is calling kubectl
2. error message is `Error from server: the server could not find the requested resource`
3. in the apiserver log, you should see `GET /apis/company.com/v1: (518.944µs) 404 [[kubectl/v1.3.0 (linux/amd64) kubernetes/ae28564] 104.154.110.118:46043]`
For detail see [here](https://github.com/kubernetes/kubernetes/issues/26425#issuecomment-222844523)
cc @janetkuo @brendanburns
Automatic merge from submit-queue
Stabilize controller unit tests.
Remove test "5-1", it's flaky as it depends on order of execution of goroutines. When the controller starts, existing claim is enqueued as "initial sync event" and a new volume is enqueued to separate goroutine. It is not deterministic which goroutine processes its events first and there is no way how to tell that the claim event was processed.
Also, force resync of the controllers after the test to make sure all events are processed.
Fixes unit test flakes.
@kubernetes/sig-storage
Automatic merge from submit-queue
AWS: ELB proxy protocol support via annotation service.beta.kubernetes.io/aws-load-balancer-proxy-protocol
This is a ~~work in progress~~ branch that adds support for the Proxy Protocol with Elastic Load Balancers. The proxy protocol is documented here: http://www.haproxy.org/download/1.5/doc/proxy-protocol.txt. It allows us to pass the "real ip" address of a client to pods behind services.
As it stands now, we create an ELB policy on the load balancer that enables the proxy protocol. We then enumerate each node port assigned to the load balancer and add our newly created policy to it. The manual process is documented here: http://docs.aws.amazon.com/ElasticLoadBalancing/latest/DeveloperGuide/enable-proxy-protocol.html
Right now, I’m looking to get some feedback on the approach before I dive too much deeper in the code. More precisely, I have questions regarding the following:
1) Right now I just check that a certain annotation exists on the service regardless of what its value is. Assuming we’re going to enable this feature via an annotation, what is the expected experience? This decision likely depends on the answers to the next questions.
2) Right now the implementation enables the proxy protocol on every ELB backend. The actual ELB API expects you to add the policy for each configured backend. Do we want the ability to configure the proxy protocol on a per service port basis? For example, if a service exposes TCP 80 and 443, would we want the ability to only enable the proxy protocol on port 443? Does this overcomplicate the implementation? If we wanted to go this direction we could do something like ...
```
{
"service.beta.kubernetes.io/aws-load-balancer-proxy-protocol": "tcp:80,tcp:443"
}
```
3) I avoided this because I was concerned with scope creep and our organization doesn’t need it, but could/should our implementation be adjusted to just handle ELB policies in general? I hadn’t used the ELB API until I started working on this branch so I don’t know how realistic this is. I also don't know how common this use case is as our organization has used our own load balancing setup prior to Kubernetes. This page has a couple of examples at the bottom: http://docs.aws.amazon.com/cli/latest/reference/elb/create-load-balancer-policy.html
cc @justinsb
<!-- Reviewable:start -->
---
This change is [<img src="http://reviewable.k8s.io/review_button.svg" height="35" align="absmiddle" alt="Reviewable"/>](http://reviewable.k8s.io/reviews/kubernetes/kubernetes/24569)
<!-- Reviewable:end -->
Add ELB proxy protocol support via the annotation
"service.beta.kubernetes.io/aws-load-balancer-proxy-protocol". This
allows servers like Nginx and Haproxy to retrieve the real IP address of
a remote client.
With quotes, the service doesn't start for systemd 219 with the error
saying the path of the netns cannot be found.
This PR fixes the bug by removing the quotes surround the netns path.
Remove test "5-1", it's flaky as it depends on order of execution of
goroutines. When the controller starts, existing claim is enqueued as
"initial sync event" and a new volume is enqueued to separate goroutine.
It is not deterministic which goroutine processes its events first and
there is no way how to tell that the claim event was processed.
Also, force resync of the controllers after the test to make sure all
events are processed.
Automatic merge from submit-queue
daemonset handle DeletedFinalStateUnknown
During an e2e run in OpenShift we ran into the DS controller panic when handling `DeletedFinalStateUnknown`. This PR checks for `DeletedFinalStateUnknown` and queues the embedded object if it is a `DaemonSet`.
@mikedanese - would you mind taking a look?
@deads2k
```
panic: interface conversion: interface is cache.DeletedFinalStateUnknown, not *extensions.DaemonSet
goroutine 4369 [running]:
k8s.io/kubernetes/pkg/controller/daemon.func·005(0x2f8a0c0, 0xc20b559680)
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/controller/daemon/controller.go:160 +0x50
k8s.io/kubernetes/pkg/controller/framework.ResourceEventHandlerFuncs.OnDelete(0xc20a0ae090, 0xc20a0ae0a0, 0xc20a0ae0b0, 0x2f8a0c0, 0xc20b559680)
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/controller/framework/controller.go:178 +0x41
k8s.io/kubernetes/pkg/controller/framework.(*ResourceEventHandlerFuncs).OnDelete(0xc20b8ebf20, 0x2f8a0c0, 0xc20b559680)
<autogenerated>:25 +0xb5
k8s.io/kubernetes/pkg/controller/framework.func·001(0x2f8a280, 0xc20b5522e0, 0x0, 0x0)
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/controller/framework/controller.go:248 +0x4be
k8s.io/kubernetes/pkg/controller/framework.(*Controller).processLoop(0xc20bb727e0)
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/controller/framework/controller.go:122 +0x6f
k8s.io/kubernetes/pkg/controller/framework.*Controller.(k8s.io/kubernetes/pkg/controller/framework.processLoop)·fm()
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/controller/framework/controller.go:97 +0x27
k8s.io/kubernetes/pkg/util/wait.func·001()
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/util/wait/wait.go:66 +0x61
k8s.io/kubernetes/pkg/util/wait.JitterUntil(0xc209f8cfb8, 0x3b9aca00, 0x0, 0xc2080543c0)
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/util/wait/wait.go:67 +0x8f
k8s.io/kubernetes/pkg/util/wait.Until(0xc209f8cfb8, 0x3b9aca00, 0xc2080543c0)
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/util/wait/wait.go:47 +0x4a
k8s.io/kubernetes/pkg/controller/framework.(*Controller).Run(0xc20bb727e0, 0xc2080543c0)
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/controller/framework/controller.go:97 +0x1fb
created by k8s.io/kubernetes/pkg/controller/daemon.(*DaemonSetsController).Run
/data/src/github.com/openshift/origin/Godeps/_workspace/src/k8s.io/kubernetes/pkg/controller/daemon/controller.go:212 +0xae
```
https://ci.openshift.redhat.com/jenkins/job/test_pull_requests_origin_check/1002/artifact/origin/artifacts/test-cmd/logs/openshift.log
Automatic merge from submit-queue
ScheduledJob validation
@erictune while playing earlier today I've noticed `suspend` isn't a pointer which requires it to be set. Additionally the validation for job selectors is too strict in that it requires the selector to match produced pods, which doesn't make sense for SJ, I've changed it to being forbidden to set entirely.
[]()
Automatic merge from submit-queue
Reduce volume controller sync period
fixes#24236 and most probably also fixes#25294.
Needs #25881! With the cache, binder is not affected by sync period. Without the cache, binding of 1000 PVCs takes more than 5 minutes (instead of ~70 seconds).
15 seconds were chosen by fair 2d10 roll :-)
Automatic merge from submit-queue
volume controller: Add cache with the latest version of PVs and PVCs
When the controller binds a PV to PVC, it saves both objects to etcd. However, there is still an old version of these objects in the controller Informer cache. So, when a new PVC comes, the PV is still seen as available and may get bound to the new PVC. This will be blocked by etcd, still, it creates unnecessary traffic that slows everything down.
To make everything worse, when periodic sync with the old PVC is performed, this PVC is seen by the controller as Pending (while it's already Bound on etcd) and will be bound to a different PV. Writing to this PV won't be blocked by etcd, only subsequent write of the PVC fails. So, the controller will need to roll back the PV in another transaction(s). The controller can keep itself pretty busy this way.
Also, we save bound PVs (and PVCs) as two transactions - we save say PV.Spec first and then .Status. The controller gets "PV.Spec updated" event from etcd and tries to fix the Status, as it seems to the controller it's outdated. This write again fails - there already is a correct version in etcd.
As we can't influence the Informer cache, it is read-only to the controller, this patch introduces second cache in the controller, which holds latest and greatest version on PVs and PVCs to prevent these useless writes to etcd . It gets updated with events from etcd *and* after etcd confirms successful save of PV/PVC modified by the controller.
The cache stores only *pointers* to PVs/PVCs, so in ideal case it shares the actual object data with the informer cache. They will diverge only for a short time when the controller modifies something and the informer cache did not get update events yet.
@kubernetes/sig-storage
Automatic merge from submit-queue
Move shell completion generation into 'kubectl completion' command
Remove static shell completion scripts from the repo and add `completion` command to kubectl:
```bash
$ source <(kubectl completion bash)
```
or
```bash
$ source <(kubectl completion zsh)
```
This makes maintenance easier because no static scripts must be generated and committed anymore in the repo.
Moreover, kubectl is self-contained again for the user including the latest completion code. I am thinking about the use-case of updating kubectl via gcloud (or some package manager). The completion code is always in-sync, without the need to download a `contrib/completion/bash/kubectl` file from github.
Opinions are welcome /cc @eparis @nak3
Fixes https://github.com/openshift/origin/issues/5290
<!-- Reviewable:start -->
---
This change is [<img src="http://reviewable.k8s.io/review_button.svg" height="35" align="absmiddle" alt="Reviewable"/>](http://reviewable.k8s.io/reviews/kubernetes/kubernetes/23801)
<!-- Reviewable:end -->
Automatic merge from submit-queue
Kubelet: Cache image history to eliminate the performance regression
Fix https://github.com/kubernetes/kubernetes/issues/25057.
The image history operation takes almost 50% of cpu usage in kubelet performance test. We should cache image history instead of getting it from runtime everytime.
This PR cached image history in imageStatsProvider and added unit test.
@yujuhong @vishh
/cc @kubernetes/sig-node
Mark v1.3 because this is a relatively significant performance regression.
[]()
Automatic merge from submit-queue
kubectl: cast scale errors to actual errors when deleting
Fixes some of the deployment reaper timeouts in e2e
@kubernetes/deployment @soltysh
Automatic merge from submit-queue
Do not call NewFlannelServer() unless flannel overlay is enabled
Ref: #26093
This makes so kubelet does not warn the user that iptables isn't in PATH, although the user didn't enable the flannel overlay.
@vishh @freehan @bprashanth
