Automatic merge from submit-queue
Improve source ip preservation test, fail the test instead of panic.
From #31085.
The source IP preserve test starts to be flake again. Sending out this PR to get rid of panicing and log the unexpected output for future investigation.
@freehan
Automatic merge from submit-queue
fix loopback client config for integration tests
Wires an authenticator/authorizer pair much closer to what the actual master start does. Several pulls I have in progress require this.
@liggitt second commit wiring changes.
Automatic merge from submit-queue
[Client-gen] Let versioned client use versioned options
i.e., use v1.ListOptions, v1.DeleteOptions when possible.
Remove the extension/v1beta1.ListOptions, because it's exactly the same as v1.ListOptions, and is not referred throughout the code base. After its removal, I register v1.ListOptions during extensions/v1beta1 scheme registration.
First three commits are manual changes.
Fix#27753
cc @lavalamp
Automatic merge from submit-queue
Add a check for file size if the reading content returns empty
In order to debug the flaky tests for writing/reading files via
contains, this PR adds a check for file size if reading returns empty
content.
Automatic merge from submit-queue
Fix kubelet perf data to make it work again for perfdash.
Addresses https://github.com/kubernetes/kubernetes/pull/30333#issuecomment-248791257.
Add the "node" label back to fix kubelet perf dash. At least for now, we still need original perfdash to catch summary api performance regression.
/cc @coufon @yujuhong
Previously, we used the docker config digest (also called "image ID"
by Docker) for the value of the `ImageID` field in the container status.
This was not particularly useful, since the config manifest is not
what's used to identify the image in a registry, which uses the manifest
digest instead. Docker 1.12+ always populates the RepoDigests field
with the manifest digests, and Docker 1.10 and 1.11 populate it when
images are pulled by digest.
This commit changes `ImageID` to point to the the manifest digest when
available, using the prefix `docker-pullable://` (instead of
`docker://`)
Automatic merge from submit-queue
Heal the namespaceless ingresses in federation e2e.
For createIngressOrFail, it incorrectly returned the ingress passed in as an argument, which does not include the namespace, instead of the ingress returned from the create call (which does).
This in turn leads to errors in e2e tests like this:
INFO: Waiting for Ingress federated-ingress to acquire IP, error an empty namespace may not be set when a resource name is provided.
Self-applying LGTM label, as this is the same code that was LGTM'd by @nikhiljindal in #33502
Automatic merge from submit-queue
Delete evicted pet
If pet was evicted by kubelet - it will stuck in this state forever.
By analogy to regular pod we need to re-create pet so that it will
be re-scheduled to another node, so in order to re-create pet
and preserve consitent naming we will delete it in petset controller
and create after that.
fixes: https://github.com/kubernetes/kubernetes/issues/31098
Automatic merge from submit-queue
Deregister clusters during federated namespace e2e tear down.
This is causing other tests to leak resources.
cc @mwielgus @kubernetes/sig-cluster-federation
Automatic merge from submit-queue
Fix summary test
Issue was comparing an `unversioned.Time` rather than `time.Time`. I temporarily removed the `[Flaky]` tag so the PR builder will run the test. I will revert that change before submitting.
Automatic merge from submit-queue
fix kubectl taint e2e flake: add retries for removing taint
**What this PR does / why we need it**:
Why we need it: recent failures occurred in #29503 are caused by taints removing conflict on nodes, this PR is to fix it. (#33073 fixed taints updating conflict, but not taints removing.)
What this PR does: use `runKubectlRetryOrDie()` instead of `RunKubectlOrDie()` in all the places in "Kubectl taint" e2e tests.
**Which issue this PR fixes** : fixes part of #29503, (would like to keep this issue open for some days more to make sure no other failures occur)
**Special notes for your reviewer**: NONE
**Release note**: NONE
Automatic merge from submit-queue
Provide more detailed error message when pod fails to be success state
Return more detailed error message when it timeouts for waiting for pod to success state.
This PR extends #31895. #31895 is for running and ready state.
It is used for debugging #31561#32980
cc: @lavalamp
Automatic merge from submit-queue
pass loopback config to poststart hooks
Updates post start hooks to take a clientconfig with the new loopback credentials for bootstrapping.
@ericchiang This is a little bit of plumbing, but mainly auth I think.
Automatic merge from submit-queue
Add ECDSA support for service account tokens
Fixes#28180
```release-note
ECDSA keys can now be used for signing and verifying service account tokens.
```
Automatic merge from submit-queue
Increase timeout for federated ingress test.
Right now federated ingress e2e takes more than 1 minute, as we need to wait for the first clusters ingress to have an IP address allocated to it before creating the others. Sometimes this takes a while due to GCE loadbalancer backend delays.
Automatic merge from submit-queue
move core storage out of master.go
Moves the core resource creation out of master.go and makes it more congruent to the other storages.
WIP because I haven't run tests yet, but I figured I'd see what breaks in the morning.
Automatic merge from submit-queue
Change minion to node
Continuation of #1111
I tried to keep this PR down to just a simple search-n-replace to keep
things simple. I may have gone too far in some spots but its easy to
roll those back if needed - just let me know.
I avoided renaming `contrib/mesos/pkg/minion` because there's already
a `contrib/mesos/pkg/node` dir and fixing that will require a bit of work
due to a circular import chain that pops up. So I'm saving that for a
follow-on PR.
Signed-off-by: Doug Davis <dug@us.ibm.com>
Automatic merge from submit-queue
Rewrite summary e2e test to check metric sanity
Take two, forked from https://github.com/kubernetes/kubernetes/pull/28195
Adds a test library that extends the ginkgo matchers to check nested data structures. Then uses the new matcher library to thoroughly check the validity of every field in the summary metrics API. This approach is more flexible than the previous approach since it allows for different tests per-field, and is easier to add case-by-case exceptions. It also places the lower & upper bounds side-by-side, making the test much easier to read & reason about.
Most fields are expected to be within some bounds. This is not intended to be a performance test, so metric bounds are very loose. Rather, I'm looking to check that the values are sane to catch bugs like #27194Fixes#23411, https://github.com/kubernetes/kubernetes/issues/31989
/cc @kubernetes/sig-node
Automatic merge from submit-queue
Make the restart test restart the nodes without a mig rolling update.
This is one approach to fix#33113. I switched from using a mig rolling-update to just pushing the reset button on the nodes and then waiting for their boot IDs to change.
Contination of #1111
I tried to keep this PR down to just a simple search-n-replace to keep
things simple. I may have gone too far in some spots but its easy to
roll those back if needed.
I avoided renaming `contrib/mesos/pkg/minion` because there's already
a `contrib/mesos/pkg/node` dir and fixing that will require a bit of work
due to a circular import chain that pops up. So I'm saving that for a
follow-on PR.
I rolled back some of this from a previous commit because it just got
to big/messy. Will follow up with additional PRs
Signed-off-by: Doug Davis <dug@us.ibm.com>
Automatic merge from submit-queue
Fix TestCreateWithNonExistentOwner
Fix#30228
As https://github.com/kubernetes/kubernetes/issues/30228#issuecomment-248779567 described, the GC did delete the garbage, it's the test logic failed.
The test used to rely on `gc.QueuesDrained()`, which could return before the GC finished processing. It seems to be the only possible reason of the test failure. Hence, this PR changed the test to poll for the deletion of garbage.
Automatic merge from submit-queue
Use strongly-typed types.NodeName for a node name
We had another bug where we confused the hostname with the NodeName.
Also, if we want to use different values for the Node.Name (which is
an important step for making installation easier), we need to keep
better control over this.
A tedious but mechanical commit therefore, to change all uses of the
node name to use types.NodeName
Automatic merge from submit-queue
Refactor volume controller parameters into a structure
`persistentvolumecontroller.NewPersistentVolumeController` has 11 arguments now,
put them into a structure.
Also, rename `NewPersistentVolumeController` to `NewController`, `persistentvolume`
is already name of the package.
Fixes#30219
We had another bug where we confused the hostname with the NodeName.
To avoid this happening again, and to make the code more
self-documenting, we use types.NodeName (a typedef alias for string)
whenever we are referring to the Node.Name.
A tedious but mechanical commit therefore, to change all uses of the
node name to use types.NodeName
Also clean up some of the (many) places where the NodeName is referred
to as a hostname (not true on AWS), or an instanceID (not true on GCE),
etc.
Automatic merge from submit-queue
start creating controller SA roles. start with just one
This creates a clusterrole for the replicationcontroller controller. It also streamlines the rule creation code and I'll use this role as practice for wiring up RBAC rules.
@kubernetes/sig-auth
@ericchiang Jordan is ooto, mind taking a look?
persistentvolumecontroller.NewPersistentVolumeController has 11 arguments now,
put them into a structure.
Also, rename NewPersistentVolumeController to NewController, persistentvolume
is already name of the package.
Fixes#30219
Automatic merge from submit-queue
Added e2e framweork and a simple test for HA master.
Added e2e framweork for testing HA master. Added one simple e2e test for HA master that first grows and then shrinks GCE cluster.
Automatic merge from submit-queue
Corrected timeout on Downward API volume
The tests for Downward API volume were apparently created before the
default sync-frequency was set to 1 minute. As the current
implementation of Pod gives us no guarantee that the changes will be
seen in Downward API before `sync-frequency` we need the tests to
reflect this.
Fixes: #29633
Automatic merge from submit-queue
Pass the real cluster scoped service object to cleanup functions instead of passing the federation scoped object.
cc @kubernetes/sig-cluster-federation
Ref: Issue #31624
Automatic merge from submit-queue
Logging soak
Implements #24427
Needs
- #24471 so that it doesnt clog test outputs for scale
- builds on the utils function added in support of #22869
cc @timothysc @kubernetes/sig-testing
Automatic merge from submit-queue
Allow garbage collection to work against different API prefixes
The GC needs to build clients based only on Resource or Kind. Hoist the
restmapper out of the controller and the clientpool, support a new
ClientForGroupVersionKind and ClientForGroupVersionResource, and use the
appropriate one in both places.
Allows OpenShift to use the GC
Automatic merge from submit-queue
Speed up job's e2e when waiting for failure
**What this PR does / why we need it**:
Job controller synchronizes objects only when job itself or underlying pod changes. Or, when full resync is performed once 10 mins. This leads e2e test to unnecessarily wait that longer timeout, sometimes at least. I've added job modification action which triggers resync, if the job wasn't terminated within shorter period of time.
@ixdy ptal
@janetkuo @erictune fyi
Automatic merge from submit-queue
Fix node performance benchmark by using latest containervm image (docker 1.11.2)
Also add two more tests for resource tracking.
cc/ @Random-Liu @coufon
Automatic merge from submit-queue
Staging 1.5 client
Created the 1.5 folder and remove the 1.4 folder in the staging area in the master branch.
Content of kubernetes/client-go/1.4 will be pulled from the kubernetes/kubernetes 1.4 branch (https://github.com/kubernetes/contrib/pull/1719)
Automatic merge from submit-queue
remove storage related fields from genericapiserver
Removes `StorageFactory` and `StorageDecorator` from from `genericapiserver` since both constructs are related to building a `RESTStorage`, which should be provided fully formed (or via factory func) to a truly generic API server.
I found this while trying to move the creation API routes earlier.
The GC needs to build clients based only on Resource or Kind. Hoist the
restmapper out of the controller and the clientpool, support a new
ClientForGroupVersionKind and ClientForGroupVersionResource, and use the
appropriate one in both places.
Automatic merge from submit-queue
Node E2E: Change the disk eviction test to pull images again after the test.
Fixes https://github.com/kubernetes/kubernetes/issues/32022#issuecomment-248677706.
This PR changes the disk eviction test to pull test images again in `AfterEach`, because images may be evicted during the test.
@yujuhong
/cc @kubernetes/sig-node
Automatic merge from submit-queue
Make node E2E tests more transparent
Add some logging and minor code reorg to make the node E2E tests a little more transparent and understandable.
Automatic merge from submit-queue
Integration tests: Migrate scheduler perf to the integration suite, s…
*Problem*
issues like https://github.com/kubernetes/kubernetes/pull/32384/ happen because we dont compile the integration tests as part of the build
*Solution*
This pr will make it so scheduler_perf is included in the integration tests, but it will run in short mode that only makes sure compilation happens . it also removes the over generalized test/component directory which only has one entry
```
test/integration/replicationcontroller
test/integration/scheduler
test/integration/scheduler
test/integration/scheduler_perf
test/integration/scheduler_perf
test/integration/secrets
test/integration/serviceaccount
test/integration/storageclasses
```
The tests for Downward API volume were apparently created before the
default sync-frequency was set to 1 minute. As the current
implementation of Pod gives us no guarantee that the changes will be
seen in Downward API before `sync-frequency` we need the tests to
reflect this.
Fixes: #29633
Automatic merge from submit-queue
Refactor cert utils into one pkg, add funcs from bootkube for kubeadm to use
**What this PR does / why we need it**:
We have ended-up with rather incomplete and fragmented collection of utils for handling certificates. It may be worse to consider using `cfssl` for doing all of these things, but for now there is some functionality that we need in `kubeadm` that we can borrow from bootkube. It makes sense to move the utils from bookube into core, as discussed in #31221.
**Special notes for your reviewer**: I've taken the opportunity to review names of existing funcs and tried to make some improvements in that area (with help from @peterbourgon).
**Release note**:
```release-note
NONE
```
Automatic merge from submit-queue
Minor Ingress tests cleanup, that includes service shard and GCE resource cleanups in underlying clusters.
Follow up for #32810.
cc @kubernetes/sig-cluster-federation
Automatic merge from submit-queue
Allow secure access to apiserver from Admission Controllers
* Allow options.InsecurePort to be set to 0 to switch off insecure access
* In NewSelfClient, Set the TLSClientConfig to the cert and key files
if InsecurePort is switched off
* Mint a bearer token that allows the client(s) created in NewSelfClient
to talk to the api server
* Add a new authenticator that checks for this specific bearer token
Fixes#13598
Automatic merge from submit-queue
Update the containervm image to the latest one (container-v1-3-v20160…
Node e2e is running with old containervm image which only has docker 1.9.1. This pr fixed such issue.
Automatic merge from submit-queue
Dumping federation events if federation e2e test failed
Updating the e2e framework to dump events in federation control plane if a federation e2e test failed.
This should help in debugging https://github.com/kubernetes/kubernetes/issues/32733
cc @kubernetes/sig-cluster-federation
Spawn pet set with single replica and simple pod. They will have
conflicting hostPort definitions, and spawned on the same node.
As the result pet set pod, it will be created after simple pod, will be
in Failed state. Pet set controller will try to re-create it. After
verifying that pet set pod failed and was recreated atleast once, we will
remove pod with conflicting hostPort and wait until pet set pod will be in
running state.
Change-Id: I5903f5881f8606c696bd390df58b06ece33be88a
Automatic merge from submit-queue
Move generated openAPI specs out of genericapiserver and make it configurable
A follow up to #31468
Generated OpenAPI Spec does not belong to genericapiserver package. A new package "generated" created in hope of all generated codes goes into it in future. Openapi folder of that package contains generated definitions and generic API server will accept the definition map as a configuration parameter.
Reference: #13414
Prior to this, we would approve eviction as long as the current state of
the pods matched the budget. The new version requires that after the
eviction, the pods would still match the budget.
Also update tests to match.
Automatic merge from submit-queue
Stop referring to loop variable inside closure.
Prior to this, I was actually running the same (last) test 5 times in a row. :-(
Fixes#33137
Automatic merge from submit-queue
Fix the properties file for node e2e cri validation.
I fixed this locally before, but accidentally missed in the PR. Sorry about that.
This time, I've tried myself, it should work.
@yujuhong
Automatic merge from submit-queue
Bump up GCI version.
```release-note
Upgrading Container-VM base image for k8s on GCE. Brief changelog as follows:
- Fixed performance regression in veth device driver
- Docker and related binaries are statically linked
- Fixed the issue of systemd being oom-killable
```
Fixes#32596
This needs a cherrypick into v1.4 release branch because it is fixing v1.4 release blocking issues. This patch is easy and safe to rollback in case of emergencies.
@vishh can you please review?
Fixes#32596 and many other issues.
cc/ @kubernetes/goog-image FYI
Automatic merge from submit-queue
Viper direct bindings to TestContext struct with hierarchichal suppor…
Part of #31453 to support hierarchichal parameters. This one does so for density, paves way for other tests as well.
Brief changelog compared to gci-dev-54-8743-3-0:
- Fixed performance regression in veth device driver
- Docker and related binaries are statically linked
- Fixed the issue of systemd being oom-killable
- Updated built-in kubelet version to 1.3.7
- add ethtool and ebtables binaries expected by kubelet
Fixes#32596
Automatic merge from submit-queue
separate RESTStorage by API group
This demonstrates how we could start separating the description of `RESTStorage` from `pkg/master`. Each API group owns constructing its own RESTStorage. Since `pkg/registry` is where we place all packages that create RESTStorage, it seems reason to split the package by `pkg/registry/<api group>/storage` to create RESTStorage and `pkg/registry/<api group>/<resource>` to support each individual resource.
The interface to construct this RESTStorage is dependent upon some `genericapiserver` types now, but we could (should?) move those items to a different location.
@kubernetes/sig-api-machinery
@lavalamp can you weigh in on how I've split these packages? I think we agree that we need better segregation, but we haven't spoken about how to spell it.
* Allow options.InsecurePort to be set to 0 to switch off insecure access
* In NewSelfClient, Set the TLSClientConfig to the cert and key files
if InsecurePort is switched off
* Mint a bearer token that allows the client(s) created in NewSelfClient
to talk to the api server
* Add a new authenticator that checks for this specific bearer token
Fixes#13598
Automatic merge from submit-queue
Node E2E: Add image white list
This is part of #29081. Fixes#29155.
As is discussed with @yujuhong in #29155, it is difficult to maintain the prepull image list if it is not enforced.
This PR added an image white list in the test framework, only images in the white list could be used in the test. If the image is not in the white list, the test will fail with reason:
```
Image "XXX" is not in the white list, consider adding it to CommonImageWhiteList in test/e2e/common/util.go or NodeImageWhiteList in test/e2e_node/image_list.go
```
Notice that if image pull policy is `PullAlways`, the image is not necessary to be in the white list or prepulled, because the test expects the image to be pulled during the test.
Currently, the image white list is only enabled in node e2e, because the image puller in e2e test is not integrated with the image white list yet.
/cc @kubernetes/sig-node
Automatic merge from submit-queue
Node E2E: Add image pull retry in image pulling test.
Fixes#29259, #28047.
This test added image pull retry in image pulling node e2e test. It will retry for 3 times until test successes.
This should be able to make the image pulling test less flaky.
@yujuhong
Automatic merge from submit-queue
Implement cleanup methods to cleanup underlying cluster resources in federated service tests.
cc @kubernetes/sig-cluster-federation @colhom
Automatic merge from submit-queue
add GenericAPIServer posthooks for initialization
Adds the concept of a `PostStartHook` to the `GenericAPIServer` to allow post-server start hooks. This gives a standard location to perform post-start bootstrapping tasks. The common case usage we have downstream are security related bootstrapping tasks that are performed on the "empty etcd" initialization cases. The RBAC authorizer is a good example of this in kube. It needs a location to create default policies to start a server which is capable of being accessed.
Kube is also likely to hit this for things like PSP and breaking the monolithic controller user into separate, scoped service accounts.
@kubernetes/sig-api-machinery for the `GenericAPIServer` bits
@kubernetes/sig-auth for the particular clusterrole bootstrapping. I've only done one to start, but I suspect we'll start making more and probably introduce a binding so that the RBAC super-user doesn't remain a special case forever.
Automatic merge from submit-queue
Make container exec failures in e2e easier to debug
Makes container exec failures in e2e tests easier to debug. Found while chasing some SELinux bugs :)
@pwittrock I'm adding this to the 1.4 milestone because it makes e2e failures easier to debug.
Automatic merge from submit-queue
[kubelet] Fix oom-score-adj policy in kubelet
Fixes#32238
We have been having this regression since v1.3. It is critical for GKE/GCE deployments of k8s because docker daemon has a high likelihood of being OOM killed which will end up nuking all containers.
The reason for moving from mnt to pid is that docker daemon moves itself into a new mnt namespace with systemd based deployments.
Automatic merge from submit-queue
Fix error message around gcloud calls in node e2e and gubernator
Fixes some janky error messages around gcloud calls.
Automatic merge from submit-queue
Pet Set Example for Cassandra
- updating cassandra to 3.7
- added pet set example
- adding pet set for Cassandra e2e tests
- changed service as we do not want a lb service, as we are running C*
- updated docs
cc @bgrant0607
cc @kubernetes/examples
We can probably close a couple of other open PR, since I did some other stuff.
Automatic merge from submit-queue
test: add/remove myself from tests appropriately
Added/removed myself from tests and run the pythong script that updates the csv
@fejta ptal
Automatic merge from submit-queue
Cleanup non-rest apiserver handlers
- rename MuxHelper -> PathRecorderMux
- move non-rest handlers into routes packages within genericapiserver and `pkg/routes` (those from master)
- move ui and logs handlers out of genericapiserver (they are
not generic)
- make version handler configurable (`config.EnableVersion`)
Automatic merge from submit-queue
Decrease timeout for namespace creation in test
If apiserver is unresponsive (e.g. because of crashloop or sth), we are wasting a lot of test time on retries.
Automatic merge from submit-queue
Add kubectl run ScheduledJob e2e test
**What this PR does / why we need it**:
This add another `kubectl run` e2e test, this time verifying proper creation of a ScheduledJob.
@janetkuo ptal
@deads2k that should give you more confidence when ditching manual clients
Automatic merge from submit-queue
change the error log for empty resource usage
This PR changes the error log for empty resource usage buffer for a container to be more clear. It happens when the container name is wrong, or cAdvisor somehow does not response.
Automatic merge from submit-queue
Add Viper parametrization as E2E config option.
do-not-merge
Fixes#18099 via viper rather than inis.
Wont build until we remove BurntSushi/ COPYING based deps from upstream viper.
I'll dig into those issues independently and update later, before pushing the updated godeps into this PR.
Kubernetes Submit Queue