Lots of comments describing the heuristics, how it fits together and the
limitations.
In particular, we can't guarantee correct volume placement if the set of
zones is changing between allocating volumes.
Automatic merge from submit-queue
Deployment controller's cleanupUnhealthyReplicas should respect minReadySeconds
```release-note
Fixed an issue that Deployment may be scaled down further than allowed by maxUnavailable when minReadySeconds is set.
```
Fixes#26834
Detected by a flake in deployment rollover e2e test (the only test that specifies `minReadySeconds`).
cc @kubernetes/deployment @pwittrock
cc @mqliang who first added `cleanupUnhealthyReplicas` in deployment controller
[]()
Automatic merge from submit-queue
Reapply ScheduledJob tests (2ab885a53a)
Re-applied the ScheduledJob tests (#25737) which were reverted due to an integration test error in #27184.
The problem was in `TestBatchGroupBackwardCompatibility` which is testing backwards compatibility for storing jobs (`extensions/v1beta1` vs `batch/v1`), which is not needed for `batch/v2alpha1`. I've added a skip to aforementioned test for that group. See `test/integration/master_test.go` for the actual fix.
@caesarxuchao @mikedanese ptal
@piosz @jszczepkowski @erictune fyi
[]()
Automatic merge from submit-queue
GCE provider: Limit Filter calls to regexps rather than insane blobs
Filters can't exceed 4k, and GET requests against the GCE API are also limited, so these break down in different ways at different cluster counts. Fix it by introducing an advisory `node-instance-prefix` configuration in the GCE provider that can hint the `EnsureLoadBalancer`/`UpdateLoadBalancer code` (and the firewall creation/update code). If it's not there, or wrong (a hostname that's registered violates it), just ignore it and grab the whole project.
Fixes#27731
[]()
Filters can't exceed 4k, and GET requests against the GCE API are also
limited, so these break down in different ways at different cluster
counts. Fix it by introducing an advisory node-instance-prefix
configuration in the GCE provider that can hint the
EnsureLoadBalancer/UpdateLoadBalancer code (and the firewall
creation/update code). If it's not there, or wrong (a hostname that's
registered violates it), just ignore it and grab the whole project.
When kubelet starts a pod that refers to non-existing PV, PVC or Node, it
should clearly show that the requested element does not exist.
Previous "PersistentVolumeClaim 'default/ceph-claim-wm' is not in cache"
looks like random kubelet hiccup, while "PersistentVolumeClaim
'default/ceph-claim-wm' not found" suggests that the object may not exist at
all and it might be an user error.
Fixes#27523
Automatic merge from submit-queue
Remove pod mutation for volumes annotated with supplemental groups
Removes the pod mutation added in #20490 -- partially resolves#27197 from the standpoint of making the feature inactive in 1.3. Our plan is to make this work correctly in 1.4.
@kubernetes/sig-storage
Automatic merge from submit-queue
swap FIRSTSEEN/LASTSEEN columns in `kubectl get event -w`
```release-note
Show LASTSEEN, the sorting key, as the first column in `kubectl get event` output
```
Not having LASTSEEN as the first column can confuse users into thinking
that events are not delivered in order.
Fixes#27060
Automatic merge from submit-queue
Kubelet Volume Manager Wait For Attach Detach Controller and Backoff on Error
* Closes https://github.com/kubernetes/kubernetes/issues/27483
* Modified Attach/Detach controller to report `Node.Status.AttachedVolumes` on successful attach (unique volume name along with device path).
* Modified Kubelet Volume Manager wait for Attach/Detach controller to report success before proceeding with attach.
* Closes https://github.com/kubernetes/kubernetes/issues/27492
* Implemented an exponential backoff mechanism for for volume manager and attach/detach controller to prevent operations (attach/detach/mount/unmount/wait for controller attach/etc) from executing back to back unchecked.
* Closes https://github.com/kubernetes/kubernetes/issues/26679
* Modified volume `Attacher.WaitForAttach()` methods to uses the device path reported by the Attach/Detach controller in `Node.Status.AttachedVolumes` instead of calling out to cloud providers.
Modify attach/detach controller to keep track of volumes to report
attached in Node VolumeToAttach status.
Modify kubelet volume manager to wait for volume to show up in Node
VolumeToAttach status.
Implement exponential backoff for errors in volume manager and attach
detach controller
Automatic merge from submit-queue
rkt: Map kubelet's `--stage1-image` flag to rkt's `--stage1-name` flag.
This enables rkt to use cached stage1 image instead of unpacking the stage1 image every time for every pod.
After this change, users need to preload the stage1 images in order to enable rkt to find the stage1 image with the name specified by this flag.
Also, the cloud config is modified to pre-load the stage1 images.
cc @kubernetes/sig-rktnetes @kubernetes/sig-node
Automatic merge from submit-queue
clarify kubectl recursive flag description
Clarify the description of the recursive flag in `kubectl` so that it's more intuitive to the user
This should make it into v1.3 as the rest of the recursive feature PR's will be available in 1.3
Automatic merge from submit-queue
kubectl describe node is allocatable aware
`kubectl describe node` will render node.status.allocatable if present.
in addition, it will report allocated resources relative to node.status.allocatable if present instead of capacity.
old code was confusing if you setup system-reserved and kube-reserved as allocated resource percentages were relative to node capacity and not schedulable amount of resources.
this is a small but valuable usability improvement, so i think it would be good to make 1.3 milestone.
/cc @kubernetes/sig-node @kubernetes/rh-cluster-infra @kubernetes/kubectl @davidopp
Automatic merge from submit-queue
httplog: Increase stack size
The previous size, of 2KB, in practice always was filled mostly by
http server-releated stuff well above the panic itself, and truncated
before anything of real value was printed in some cases.
This increases the stack size so that panics are printed in full (well, except for really large ones).
cc @lavalamp
This enables rkt to use cached stage1 image instead of unpacking the
stage1 image every time for every pod.
After this change, users need to preload the stage1 images in order to
enable rkt to find the stage1 image with the name specified by this flag.
Hot attach of disk to a scsi controller will work only if the
controller type is lsilogic-sas or paravirtual.This patch filters
the existing controller for these types, if it doesn't find one it
creates a new scsi controller.
Automatic merge from submit-queue
Allow disabling of dynamic provisioning
Allow administrators to opt-out of dynamic provisioning. Provisioning is still on by default, which is the current behavior.
Per a conversation with @jsafrane, a boolean toggle was added and plumbed through into the controller. Deliberate disabling will simply return nil from `provisionClaim` whereas a misconfigured provisioner will continue on and generate error events for the PVC.
@kubernetes/rh-storage @saad-ali @thockin @abhgupta
Automatic merge from submit-queue
Fill PV.Status.Message with deleter/recycler errors.
Instead of empty `Message` `kubectl describe pv` now shows:
```
Name: nfs
Labels: <none>
Status: Failed
Claim: default/nfs
Reclaim Policy: Recycle
Access Modes: RWX
Capacity: 1Mi
Message: Recycler failed: Pod was active on the node longer than specified deadline
Source:
Type: NFS (an NFS mount that lasts the lifetime of a pod)
Server: 10.999.999.999
Path: /
ReadOnly: false
```
This is actually a regression since 1.2
@kubernetes/sig-storage
Automatic merge from submit-queue
Allow emitting PersistentVolume events.
Similarly to Nodes, PersistentVolumes are not in any namespace and we should
not block events on them. Currently, these events are rejected with
`Event "nfs.145841cf9c8cfaf0" is invalid: involvedObject.namespace: Invalid value: "": does not match involvedObject`
Automatic merge from submit-queue
add unit and integration tests for rbac authorizer
This PR adds lots of tests for the RBAC authorizer.
The plan over the next couple days is to add a lot more test cases.
Updates #23396
cc @erictune
Automatic merge from submit-queue
Remove EncodeToStream(..., []unversioned.GroupVersion)
Was not being used. Is a signature change and is necessary for post 1.3 work on Templates and other objects that nest objects.
Extracted from #26044
Automatic merge from submit-queue
AWS volumes: Use /dev/xvdXX names with EC2
We are using HVM style names, which cannot be paravirtual style names.
See
http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/device_naming.html
This also fixes problems introduced when moving volume mounting to KCM.
Fix#27534
Automatic merge from submit-queue
Logging for OutOfDisk when file system info is not available
#26566
1. Adding logs for file system info being not available.
2. Reporting outOfDisk when file system info is not available.
Automatic merge from submit-queue
ObjectMeta, ListMeta, and TypeMeta should implement their interfaces
Make unversioned.ListMeta implement List. Update all the *List types so they implement GetListMeta.
This helps avoid using reflection to get list information.
Remove all unnecessary boilerplate, move the interfaces to the right
places, and add a test that verifies that objects implement one, the
other, but never both.
@ncdc @lavalamp this supercedes #26964 with the boilerplate removed. Added tests
Automatic merge from submit-queue
Fix bug in isLikelyNotMountPoint function
In nsenter_mount.go/isLikelyNotMountPoint function, the returned output
from findmnt command misses the last letter. Modify the code to use
String.contains instead of string matching. fixes#26421fixes#25056fixes#22911
Automatic merge from submit-queue
Filter seccomp profile path from malicious .. and /
Without this patch with `localhost/<some-releative-path>` as seccomp profile one can load any file on the host, e.g. `localhost/../../../../dev/mem` which is not healthy for the kubelet.
/cc @jfrazelle
Unit tests depend on https://github.com/kubernetes/kubernetes/pull/26710.
Automatic merge from submit-queue
kubelet/kubenet: split hostport handling into separate module
This pulls the hostport functionality of kubenet out into a separate module so that it can be more easily tested and potentially used from other code (maybe CNI, maybe downstream consumers like OpenShift, etc). Couldn't find a mock iptables so I wrote one, but I didn't look very hard.
@freehan @thockin @bprashanth
Automatic merge from submit-queue
Revert revert of downward api node defaults
Reverts the revert of https://github.com/kubernetes/kubernetes/pull/27439Fixes#27062
@dchen1107 - who at Google can help debug why this caused issues with GKE infrastructure but not GCE merge queue?
/cc @wojtek-t @piosz @fgrzadkowski @eparis @pmorie
Automatic merge from submit-queue
Remove an empty line being output when exposing annotations and labels via downward api volume
The issue is that formatMap function (for annotations and labels) in pkg/fieldpath/fieldpath.go appends a "\n" after each key value pair which is correct for all pairs except the last pair because then a complete string is returned with a "\n" in the end. It is inconsistent with other strings (metadata.name, namespace and resources) being returned as they dont have "\n" in the end. These returned strings are processed by sortLines function in pkg/volume/downwardapi/downwardapi.go and the function finally appends "\n" to each string, but incorrectly outputs an empty line if there is an already "\n" in the end with the input string. To illustrate:
The sortLines works as follows: lets say the input string is : "a\nb\nc\n".
1. It splits them as "a", "b", "c", "" (note empty string in the end).
2. it sort them: "", "a", b", "c"
3. And then it appends "\n" again to each string: "\n", "a\n" ,"b\n", "c\n"
So we can see that it is erroneously creating an empty string in the beginning when the input string to sortLines has "\n" in the end. As I said above, it is not an issue with metadata.name, namespace and resources as their input strings are without \n" in the end.
So now, the output in the downward api volume, (using the example in http://kubernetes.io/docs/user-guide/downward-api/):
```
# cat /etc/annotations
zone="us-est-coast"
cluster="test-cluster1"
rack="rack-22"
```
After this patch, the output will be correct and without the erroneous empty line in the beginning.
I could think other ways to solve this but I found the way in this patch with minimal code changes.
@kubernetes/rh-cluster-infra
Automatic merge from submit-queue
refuse to create a firewall rule with no target tag
fixes#25145
This modification in gce.firewallObject() will return error when trying
to create or update firewall rule if no node tag can be found. Also add
unit test for this modification.
We had a long-lasting bug which prevented creation of volumes in
non-master zones, because the cloudprovider in the volume label
admission controller is not initialized with the multizone setting
(issue #27656).
This implements a simple workaround: if the volume is created with the
failure-domain zone label, we look for the volume in that zone. This is
more efficient, avoids introducing a new semantic, and allows users (and
the dynamic provisioner) to create volumes in non-master zones.
Fixes#27657
Long term we plan on integrating this into the scheduler, but in the
short term we use the volume name to place it onto a zone.
We hash the volume name so we don't bias to the first few zones.
If the volume name "looks like" a PetSet volume name (ending with
-<number>) then we use the number as an offset. In that case we hash
the base name.
Fixes#27256
Automatic merge from submit-queue
pkg/client/leaderelection: log err when retrieving endpoint
The leader election code currently suppresses errors when trying to retrieve an endpoint. This can lead to difficult to debug situations.
In the case of a mis-configured controller-manager or scheduler - where they fail to contact an apiserver - this currently leads to no log output in the default case, or `failed to renew lease foo/bar` in `--v=4`, which isn't very actionable.
Automatic merge from submit-queue
fix updatePod() of RS and RC controllers
Fix updatePod of replication controller manager and replica set controller to handle pod label updates that match no RC or RS.
Fix#27405
If the mount operation exceeds the timeout, it will return an error and the
pod worker will retry in the next sync (10s or less). Compared with the
original value (i.e., 10 minutes), this frees the pod worker sooner to process
pod updates, if there are any.
This commit adds a new volume manager in kubelet that synchronizes
volume mount/unmount (and attach/detach, if attach/detach controller
is not enabled).
This eliminates the race conditions between the pod creation loop
and the orphaned volumes loops. It also removes the unmount/detach
from the `syncPod()` path so volume clean up never blocks the
`syncPod` loop.
Similarly to Nodes, PersistentVolumes are not in any namespace and we should
not block events on them. Currently, these events are rejected with
'Event "nfs.145841cf9c8cfaf0" is invalid: involvedObject.namespace: Invalid value: "": does not match involvedObject'
Automatic merge from submit-queue
kubelet/rkt - treat pod container as the infra - only network stats
As no "container name" annotation was being applied to the pod as a whole, the rkt pod container didn't have a container name label. This means that in stat/summary it came up as a nameless container that belonged to the pod.
this was problematic as it caused double counting of container stats.
this adds a container name annotation to the pod level which will be overridden during label creation by annotations of the same name at the container level for the containers themselves.
stats/summary will do the right thing as it will treat it the same as the infra container, just get network stats from it.
Suppress #26759
cc @kubernetes/sig-node @kubernetes/rktnetes-maintainers
Automatic merge from submit-queue
Rbac api group make subject apiversion optional
This fixes the verification for the "apiVerion" field in the RBAC subject and makes it optional. This field isn't used and currently won't pass validation if it's filled.
```yml
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1alpha1
metadata:
name: admins
subject:
- kind: User
name: admin-user
# apiVersion: "entering anything here will fail validation"
roleRef:
kind: ClusterRole
name: admin
apiVersion: rbac.authorization.k8s.io/v1alpha1
```
Automatic merge from submit-queue
Quota uses old object provided via admission
fixes https://github.com/kubernetes/kubernetes/issues/26178
@sdminonne - fixes a bug in services not intercepting updates.
/cc @liggitt
In nsenter_mount.go/isLikelyNotMountPoint function, the returned output
from findmnt command misses the last letter. Modify the code to make sure
that output has the full target path. fix#26421#25056#22911
- replaces probeVolume with scsiHostRescan to scan hot attached disks
- fixes substring match of UUID returned from AttachDisk
- changes DetachDisk to take volumePath argument instead of diskID
- fixes delayed failure at mount rather than attach disk
- removes cloning of virtual disk in AttachDisk
The previous size, of 2KB, in practice always was filled completely by
http server-releated stuff well above the panic itself, and truncated
before anything of real value was printed.
This increases the stack size so that panics are printed in full.
This sets AttachOptions.CommandName dynamically depending on the corba Command
hierarchy. If the root command is named e.g. "oc" (for the OpenShift cli) this
will result in "oc attach" instead of the static "kubectl attach" before this
patch.
Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1341450
Automatic merge from submit-queue
Let kubelet log the DeletionTimestamp if it's not nil in update
This helps to debug if it's the kubelet to blame when a pod is not deleted.
Example output:
```
SyncLoop (UPDATE, "api"): "redis-master_default(c6782276-2dd4-11e6-b874-64510650ab1c):DeletionTimestamp=2016-06-08T23:58:12Z"
```
ref #26290
cc @Random-Liu
Automatic merge from submit-queue
Update reason_cache.go, Get method operate lru cache not threadsafe
The reason_cache wrapped lru cache , lru cache modies linked list even for a get, should use WLock for both read and write
Automatic merge from submit-queue
Fix docker api version in kubelet
There are two variables `dockerv110APIVersion` and `dockerV110APIVersion` with
the same purpose, but different values. Remove the incorrect one and fix usage
in the file.
/cc @dchen1107 @Random-Liu
Automatic merge from submit-queue
processor listener: fix locking in pop()
Currently the lock in processorListener is used to guard pendingNotifications. But in pop, it also locks around on select chan. This will block the goroutine with lock acquired.
This PR changes the lock to guard the correct section only.
Automatic merge from submit-queue
pkg/kubectl: add resource printers for rbac api group
This PR adds the necessary kubectl printers for the rbac api group which we overlooked in previous PRs.
cc @erictune
Automatic merge from submit-queue
ResourceQuota BestEffort scope aligned with Pod level QoS
This aligns quota with the changes in kubelet and CLI.
So if quota allows 10 `BestEffort` pods, it will now track properly with what the user sees with changes in 1.3.
```
apiVersion: v1
kind: ResourceQuota
metadata:
name: best-effort
spec:
hard:
pods: "10"
scopes:
- BestEffort
```
/cc @vishh @kubernetes/rh-cluster-infra
Automatic merge from submit-queue
AWS: cache instances during service reload to avoid rate limiting on restart
Fixes#25610 by reducing redundant calls to DescribeInstances()
```release-note
* The AWS cloudprovider will cache results from DescribeInstances() if the set of nodes hasn't changed
```
Also move int/stringSlicesEqual from servicecontroller.go to pkg/util/slice
Automatic merge from submit-queue
Extract interface for master endpoints reconciler.
Make the master endpoints reconciler an interface so its implementation can be overridden, if
desired.
xref #20975#26574
cc @kubernetes/sig-api-machinery @lavalamp @smarterclayton @pmorie @DirectXMan12 @wojtek-t @kubernetes/rh-cluster-infra
OpenShift needs to be able to use a discovery client against a different
prefix. Make LegacyPrefix optional and parameterizable to the client. No
change to existing interfaces.
Automatic merge from submit-queue
fix recursive & non-recursive kubectl get of generic output format
This PR fixes the issues with `kubectl get` in https://github.com/kubernetes/kubernetes/issues/26466
Changes made:
- fix printing when using the generic output format in both non-recursive & recurvise settings to ensure that errors are being shown
- add tests to check printing generic output in a **non-recursive** setting with non-existent pods
- clean up the **recursive** `kubectl get` tests
/cc @janetkuo
Automatic merge from submit-queue
Sets IgnoreUnknown=1 in CNI_ARGS
```release-note
release-note-none
```
K8 uses CNI_ARGS to pass pod namespace, name and infra container
id to the CNI network plugin. CNI logic will throw an error
if these args are not known to it, unless the user specifies
IgnoreUnknown as part of CNI_ARGS. This PR sets IgnoreUnknown=1
to prevent the CNI logic from erroring and blocking pod setup.
https://github.com/appc/cni/pull/158https://github.com/appc/cni/issues/126
Automatic merge from submit-queue
Listing pods only once when getting pods for RS in deployment
Fixes#26834
1. Avoid ranging over RSes and then `List` pods of each RS. Instead, `List` pods of the deployment once, and then filter pods of each RS.
2. Avoid using clientset to `List` pods in deployment controller. Use podStore instead. (TODO in some functions because the unit tests don't have podStore.)
@kubernetes/deployment
[]()
Since appc requires gid to be non-empty today (https://github.com/appc/spec/issues/623),
we have to error out when gid is empty instead of using the root gid.
Make unversioned.ListMeta implement List. Update all the *List types so they implement GetListMeta.
This helps avoid using reflection to get list information.
Remove all unnecessary boilerplate, move the interfaces to the right
places, and add a test that verifies that objects implement one, the
other, but never both.
Automatic merge from submit-queue
rkt: Replace 'journalctl' with rkt's GetLogs() API.
This replaced the `journactl` shell out with rkt's GetLogs() API.
Fixes#26997
To make this fully work, we need rkt to have this patch #https://github.com/coreos/rkt/pull/2763
cc @kubernetes/sig-node @euank @alban @iaguis @jonboulle
Automatic merge from submit-queue
AWS: support mixed plaintext/encrypted ports in ELBs via service.beta.kubernetes.io/aws-load-balancer-ssl-ports annotation
Fixes#26268
Implements the second SSL ELB annotation, per #24978
`service.beta.kubernetes.io/aws-load-balancer-ssl-ports=*` (comma-separated list of port numbers or e.g. `https`)
If not specified, all ports are secure (SSL or HTTPS).
Automatic merge from submit-queue
rkt: Do not run rkt pod inside a pre-created netns when network plugin is no-op
This fixed a panic where the returned pod network status is nil. (Fix#26540)
Also this makes lkvm stage1 able to run inside a user defined network, where the network name needs to be 'rkt.kubernetes.io'. A temporal solution to solve the network issue for lkvm stage1.
Besides, I fixed minor issues such as passing the wrong pod UID when cleaning up the netns file.
/cc @euank @pskrzyns @jellonek @kubernetes/sig-node
I tested with no networkplugin locally, works fine.
As a reminder, we need to document this in the release.https://github.com/kubernetes/kubernetes/issues/26201
This fixed a panic where the returned pod network status is nil.
Also this makes lkvm stage1 able to run inside a user defined
network, where the network name needs to be 'rkt.kubernetes.io'.
Also fixed minor issues such as passing the wrong pod UID, ignoring
logging errors.
Automatic merge from submit-queue
rkt: Fix incomplete selinux context string when the option is partial.
Fix "EmptyDir" e2e tests failures caused by #https://github.com/kubernetes/kubernetes/pull/24901
As mentioned in https://github.com/kubernetes/kubernetes/pull/24901#discussion_r61372312
We should apply the selinux context of the rkt data directory (/var/lib/rkt) when users do not specify all the selinux options.
Due to my fault, the change was missed during rebase, thus caused the regression.
After applying this PR, the e2e tests passed.
```
$ go run hack/e2e.go -v -test --test_args="--ginkgo.dryRun=false --ginkgo.focus=EmptyDir"
...
Ran 19 of 313 Specs in 199.319 seconds
SUCCESS! -- 19 Passed | 0 Failed | 0 Pending | 294 Skipped PASS
```
BTW, the test is removed because the `--no-overlay=true` flag will only be there on non-coreos distro.
cc @euank @kubernetes/sig-node
Automatic merge from submit-queue
LBaaS v2 Support for Openstack Cloud Provider Plugin
Resolves#19774.
This work is based on Gophercloud support for LBaaS v2 currently in review (this will have to merge first):
https://github.com/rackspace/gophercloud/pull/575
These changes includes the addition of a new loadbalancer configuration option: **LBVersion**. If this configuration attribute is missing or anything other than "v2", lbaas v1 implementation will be used.
Justin Santa Barbara
