Automatic merge from submit-queue
Enable shared PID namespace by default for docker pods
**What this PR does / why we need it**: This PR enables PID namespace sharing for docker pods by default, bringing the behavior of docker in line with the other CRI runtimes when used with docker >= 1.13.1.
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: ref #1615
**Special notes for your reviewer**: cc @dchen1107 @yujuhong
**Release note**:
```release-note
Kubernetes now shares a single PID namespace among all containers in a pod when running with docker >= 1.13.1. This means processes can now signal processes in other containers in a pod, but it also means that the `kubectl exec {pod} kill 1` pattern will cause the pod to be restarted rather than a single container.
```
Automatic merge from submit-queue
Delete "hard-coded" default value in flags usage.
**What this PR does / why we need it**:
Some flags of kubernetes components have "hard-coded" default values in their usage info. In fact, [pflag pkg](https://github.com/kubernetes/kubernetes/blob/master/vendor/github.com/spf13/pflag/flag.go#L602-L608) has already added a string `(default value)` automatically in the usage info if the flag is initialized. Then we don't need to hard-code the default value in usage info. After this PR, if we want to update the default value of a flag, we only need to update the flag where it is initialized. `pflag` will update the usage info for us. This will avoid inconsistency.
For example:
Before
```
kubelet -h
...
--node-status-update-frequency duration Specifies how often kubelet posts node status to master. Note: be cautious when changing the constant, it must work with nodeMonitorGracePeriod in nodecontroller. Default: 10s (default 10s)
...
```
After
```
kubelet -h
...
--node-status-update-frequency duration Specifies how often kubelet posts node status to master. Note: be cautious when changing the constant, it must work with nodeMonitorGracePeriod in nodecontroller. (default 10s)
...
```
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, fixes #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #
**Special notes for your reviewer**:
This PR doesn't delete some "hard-coded" default values because they are not explicitly initialized. We still need to hard-code them to give users friendly info.
```
--allow-privileged If true, allow containers to request privileged mode. [default=false]
```
**Release note**:
```release-note
None
```
Kubelet flags are not necessarily appropriate for the KubeletConfiguration
object. For example, this PR also removes HostnameOverride and NodeIP
from KubeletConfiguration. This is a preleminary step to enabling Nodes
to share configurations, as part of the dynamic Kubelet configuration
feature (#29459). Fields that must be unique for each node inhibit
sharing, because their values, by definition, cannot be shared.
Automatic merge from submit-queue
kubelet: change image-gc-high-threshold below docker dm.min_free_space
docker dm.min_free_space defaults to 10%, which "specifies the min free space percent in a thin pool require for new device creation to succeed....Whenever a new a thin pool device is created (during docker pull or during container creation), the Engine checks if the minimum free space is available. If sufficient space is unavailable, then device creation fails and any relevant docker operation fails." [1]
This setting is preventing the storage usage to cross the 90% limit. However, image GC is expected to kick in only beyond image-gc-high-threshold. The image-gc-high-threshold has a default value of 90%, and hence GC never triggers. If image-gc-high-threshold is set to a value lower than (100 - dm.min_free_space)%, GC triggers.
xref https://bugzilla.redhat.com/show_bug.cgi?id=1408309
```release-note
changed kubelet default image-gc-high-threshold to 85% to resolve a conflict with default settings in docker that prevented image garbage collection from resolving low disk space situations when using devicemapper storage.
```
@derekwaynecarr @sdodson @rhvgoyal
Automatic merge from submit-queue
Use realistic value for the memory example of kube-reserved and system-reserved
Use realistic value for the memory example of kube-reserved and system-reserved
Currently, kublet help shows the memory example of
kube-reserved and system-reserved as 150G. This 150G is not realistic
value and it leads misconfiguration or confusion. This patch changes
to example value as 500Mi.
Before(same with system-reserved):
```
--kube-reserved value A set of ResourceName=ResourceQuantity (e.g. cpu=200m,memory=150G) pairs that describe resources reserved for kubernetes system components. Currently only cpu and memory are supported. See http://releases.k8s.io/HEAD/docs/user-guide/compute-resources.md for more detail. [default=none]
```
After(same with system-reserved):
```
--kube-reserved value A set of ResourceName=ResourceQuantity (e.g. cpu=200m,memory=500Mi) pairs that describe resources reserved for kubernetes system components. Currently only cpu and memory are supported. See http://releases.k8s.io/HEAD/docs/user-guide/compute-resources.md for more detail. [default=none]
```
Automatic merge from submit-queue (batch tested with PRs 41919, 41149, 42350, 42351, 42285)
enable cgroups tiers and node allocatable enforcement on pods by default.
```release-note
Pods are launched in a separate cgroup hierarchy than system services.
```
Depends on #41753
cc @derekwaynecarr
Automatic merge from submit-queue
Allow multipe DNS servers as comma-seperated argument for kubelet --dns
This PR explores how kubectls "--dns" could be extended to specify multiple DNS servers for in-cluster PODs. Testing on the local libvirt-coreos cluster shows that multiple DNS server are injected without issues.
Specifying multiple DNS servers increases resilience against
- Packet drops
- Single server failure
I am debugging services that do 50+ DNS requests for a single incoming interactive request, thus highly increase the chance of a slowdown (+5s) due to a single packet drop. Switching to two DNS servers will reduce the impact of the issues (roughly +1s on glibc, 0s on musl, error-rate goes down to error-rate^2).
Note that there is no need to change any runtime related code as far as I know. In the case of "default" dns the /etc/resolv.conf is parsed and multiple DNS server are send to the backend anyway. This only adds the same capability for the clusterFirst case.
I've heard from @thockin that multiple DNS entries are somehow considered. I've no idea what was considered, though. This is what I would like to see for our production use, though.
```release-note
NONE
```
Automatic merge from submit-queue
Switch resourcequota controller to shared informers
Originally part of #40097
I have had some issues with this change in the past, when I updated `pkg/quota` to use the new informers while `pkg/controller/resourcequota` remained on the old informers. In this PR, both are switched to using the new informers. The issues in the past were lots of flakey test failures in the ResourceQuota e2es, where it would randomly fail to see deletions and handle replenishment. I am hoping that now that everything here is consistently using the new informers, there won't be any more of these flakes, but it's something to keep an eye out for.
I also think `pkg/controller/resourcequota` could be cleaned up. I don't think there's really any need for `replenishment_controller.go` any more since it's no longer running individual controllers per kind to replenish. It instead just uses the shared informer and adds event handlers to it. But maybe we do that in a follow up.
cc @derekwaynecarr @smarterclayton @wojtek-t @deads2k @sttts @liggitt @timothysc @kubernetes/sig-scalability-pr-reviews
This change makes kubelet to use the CRI implementation by default,
unless the users opt out explicitly by using --enable-cri=false.
For the rkt integration, the --enable-cri flag will have no effect
since rktnetes does not use CRI.
Also, mark the original --experimental-cri flag hidden and deprecated,
so that we can remove it in the next release.
Depending on an exact cluster setup multiple dns may make sense.
Comma-seperated lists of DNS server are quite common as DNS servers
are always plain IPs.
Automatic merge from submit-queue (batch tested with PRs 40251, 40171)
Mark --docker-exec-handler deprecated
We plan to drop support for the nsenter exec handler in the future. Marking this flag as deprecated to warn the users.
Automatic merge from submit-queue
Kubelet: Fix the description of MaxContainers kubelet flag.
Found this during code review.
The default number has been changed to `-1` and `1`. 82c488bd6e/pkg/apis/componentconfig/v1alpha1/defaults.go (L279-L285)
@yujuhong
/cc @saad-ali This PR fixed incorrect doc.
Provides an opt-in flag, --experimental-fail-swap-on (and corresponding
KubeletConfiguration value, ExperimentalFailSwapOn), which is false by default.
In order to be able to use new mounter library, this PR adds the
mounterPath flag to kubelet which passes the flag to the mount
interface. If flag is empty, mount uses default mount path.
Automatic merge from submit-queue
WIP: Remove the legacy networking mode
<!-- Thanks for sending a pull request! Here are some tips for you:
1. If this is your first time, read our contributor guidelines https://github.com/kubernetes/kubernetes/blob/master/CONTRIBUTING.md and developer guide https://github.com/kubernetes/kubernetes/blob/master/docs/devel/development.md
2. If you want *faster* PR reviews, read how: https://github.com/kubernetes/kubernetes/blob/master/docs/devel/faster_reviews.md
3. Follow the instructions for writing a release note: https://github.com/kubernetes/kubernetes/blob/master/docs/devel/pull-requests.md#release-notes
-->
**What this PR does / why we need it**:
Removes the deprecated configure-cbr0 flag and networking mode to avoid having untested and maybe unstable code in kubelet, see: #33789
**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, #<issue_number>, ...)` format, will close that issue when PR gets merged)*:
fixes#30589fixes#31937
**Special notes for your reviewer**: There are a lot of deployments who rely on this networking mode. Not sure how we deal with that: force switch to kubenet or just delete the old deployment?
But please review the code changes first (the first commit)
**Release note**:
<!-- Steps to write your release note:
1. Use the release-note-* labels to set the release note state (if you have access)
2. Enter your extended release note in the below block; leaving it blank means using the PR title as the release note. If no release note is required, just write `NONE`.
-->
```release-note
Removed the deprecated kubelet --configure-cbr0 flag, and with that the "classic" networking mode as well
```
PTAL @kubernetes/sig-network @kubernetes/sig-node @mikedanese
The new flag, if specified, and if --container-runtime=docker, switches
kubelet to use the new CRI implementation for testing. This is hidden flag
since the feature is still under heavy development and the flag may be changed
in the near future.
This refactor removes the legacy KubeletConfig object and adds a new
KubeletDeps object, which contains injected runtime objects and
separates them from static config. It also reduces NewMainKubelet to two
arguments: a KubeletConfiguration and a KubeletDeps.
Some mesos and kubemark code was affected by this change, and has been
modified accordingly.
And a few final notes:
KubeletDeps:
KubeletDeps will be a temporary bin for things we might consider
"injected dependencies", until we have a better dependency injection
story for the Kubelet. We will have to discuss this eventually.
RunOnce:
We will likely not pull new KubeletConfiguration from the API server
when in runonce mode, so it doesn't make sense to make this something
that can be configured centrally. We will leave it as a flag-only option
for now. Additionally, it is increasingly looking like nobody actually uses the
Kubelet's runonce mode anymore, so it may be a candidate for deprecation
and removal.
MTU selection is difficult, and if there is a transport such as IPSEC in
use may be impossible. So we allow specification of the MTU with the
network-plugin-mtu flag, and we pass this down into the network
provider.
Currently implemented by kubenet.
Automatic merge from submit-queue
Kubelet: add --container-runtime-endpoint and --image-service-endpoint
Flag `--container-runtime-endpoint` (overrides `--container-runtime`) is introduced to identify the unix socket file of the remote runtime service. And flag `--image-service-endpoint` is introduced to identify the unix socket file of the image service.
This PR is part of #28789 Milestone 0.
CC @yujuhong @Random-Liu
Automatic merge from submit-queue
Implement TLS bootstrap for kubelet using `--experimental-bootstrap-kubeconfig` (2nd take)
Ref kubernetes/features#43 (comment)
cc @gtank @philips @mikedanese @aaronlevy @liggitt @deads2k @errordeveloper @justinsb
Continue on the older PR https://github.com/kubernetes/kubernetes/pull/30094 as there are too many comments on that one and it's not loadable now.
Automatic merge from submit-queue
Unblock iterative development on pod-level cgroups
In order to allow forward progress on this feature, it takes the commits from #28017#29049 and then it globally disables the flag that allows these features to be exercised in the kubelet. The flag can be re-added to the kubelet when its actually ready.
/cc @vishh @dubstack @kubernetes/rh-cluster-infra
Add --bootstrap-kubeconfig flag to kubelet. If the flag is non-empty
and --kubeconfig doesn't exist, then the kubelet will use the bootstrap
kubeconfig to create rest client and generate certificate signing request
to request a client cert from API server.
Once succeeds, the result cert will be written down to
--cert-dir/kubelet-client.crt, and the kubeconfig will be populated with
certfile, keyfile path pointing to the result certificate file, key file.
(The key file is generated before creating the CSR).
New flag --container-runtime-endpoint (overrides --container-runtime)
is introduced to kubelet which identifies the unix socket file of
the remote runtime service. And new flag --image-service-endpoint is
introduced to kubelet which identifies the unix socket file of the
image service.
Automatic merge from submit-queue
syncNetworkUtil in kubelet and fix loadbalancerSourceRange on GCE
fixes: #29997#29039
@yujuhong Can you take a look at the kubelet part?
@girishkalele KUBE-MARK-DROP is the chain for dropping connections. Marked connection will be drop in INPUT/OUTPUT chain of filter table. Let me know if this is good enough for your use case.
Automatic merge from submit-queue
Add volume reconstruct/cleanup logic in kubelet volume manager
Currently kubelet volume management works on the concept of desired
and actual world of states. The volume manager periodically compares the
two worlds and perform volume mount/unmount and/or attach/detach
operations. When kubelet restarts, the cache of those two worlds are
gone. Although desired world can be recovered through apiserver, actual
world can not be recovered which may cause some volumes cannot be cleaned
up if their information is deleted by apiserver. This change adds the
reconstruction of the actual world by reading the pod directories from
disk. The reconstructed volume information is added to both desired
world and actual world if it cannot be found in either world. The rest
logic would be as same as before, desired world populator may clean up
the volume entry if it is no longer in apiserver, and then volume
manager should invoke unmount to clean it up.
Fixes https://github.com/kubernetes/kubernetes/issues/27653
Currently kubelet volume management works on the concept of desired
and actual world of states. The volume manager periodically compares the
two worlds and perform volume mount/unmount and/or attach/detach
operations. When kubelet restarts, the cache of those two worlds are
gone. Although desired world can be recovered through apiserver, actual
world can not be recovered which may cause some volumes cannot be cleaned
up if their information is deleted by apiserver. This change adds the
reconstruction of the actual world by reading the pod directories from
disk. The reconstructed volume information is added to both desired
world and actual world if it cannot be found in either world. The rest
logic would be as same as before, desired world populator may clean up
the volume entry if it is no longer in apiserver, and then volume
manager should invoke unmount to clean it up.
Automatic merge from submit-queue
[kubelet] Introduce --protect-kernel-defaults flag to make the tunable behaviour configurable
Let's make the default behaviour of kernel tuning configurable. The default behaviour is kept modify as has been so far.
Also provide a new --pod-manifest-path flag and deprecate the old
--config one.
This field holds the location of a manifest file or directory of manifest
files for pods the Kubelet is supposed to run. The name of the field
should reflect that purpose.
Move SystemReserved and KubeReserved into KubeletConfiguration struct
Convert int64 to int32 for some external type fields so they match internal ones
tLS* to tls* for JSON field names
Fix dependency on removed options.AutoDetectCloudProvider
Change floats in KubeletConfiguration API to ints
Update external KubeletConfiguration type
Add defaults for new KubeletConfiguration fields
Modify some defaults to match upstream settings
Add/rename some conversion functions
Updated codegen
Fixed typos
Mike Danese caught that s.NodeLabels wasn't allocated, fix on line 118
of cmd/kubelet/app/options/options.go.
Provide list of valid sources in comment for HostNetworkSources field
Automatic merge from submit-queue
[kubelet] Allow opting out of automatic cloud provider detection in kubelet. By default kubelet will auto-detect cloud providers
fixes#28231
This PR contains Kubelet changes to enable attach/detach controller control.
* It introduces a new "enable-controller-attach-detach" kubelet flag to
enable control by controller. Default enabled.
* It removes all references "SafeToDetach" annoation from controller.
* It adds the new VolumesInUse field to the Node Status API object.
* It modifies the controller to use VolumesInUse instead of SafeToDetach
annotation to gate detachment.
* There is a bug in node-problem-detector that causes VolumesInUse to
get reset every 30 seconds. Issue https://github.com/kubernetes/node-problem-detector/issues/9
opened to fix that.
Automatic merge from submit-queue
Use pause image depending on the server's platform when testing
Removed all pause image constant strings, now the pause image is chosen by arch. Part of the effort of making e2e arch-agnostic.
The pause image name and version is also now only in two places, and it's documented to bump both
Also removed "amd64" constants in the code. Such constants should be replaced by `runtime.GOARCH` or by looking up the server platform
Fixes: #22876 and #15140
Makes it easier for: #25730
Related: #17981
This is for `v1.3`
@ixdy @thockin @vishh @kubernetes/sig-testing @andyzheng0831 @pensu
This patch adds the --exit-on-lock-contention flag, which must be used
in conjunction with the --lock-file flag. When provided, it causes the
kubelet to wait for inotify events for that lock file. When an 'open'
event is received, the kubelet will exit.
Automatic merge from submit-queue
WIP v0 NVIDIA GPU support
```release-note
* Alpha support for scheduling pods on machines with NVIDIA GPUs whose kubelets use the `--experimental-nvidia-gpus` flag, using the alpha.kubernetes.io/nvidia-gpu resource
```
Implements part of #24071 for #23587
I am not familiar with the scheduler enough to know what to do with the scores. Mostly punting for now.
Missing items from the implementation plan: limitranger, rkt support, kubectl
support and docs
cc @erictune @davidopp @dchen1107 @vishh @Hui-Zhi @gopinatht
Implements part of #24071
I am not familiar with the scheduler enough to know what to do with the scores. Punting for now.
Missing items from the implementation plan: limitranger, rkt support, kubectl
support and user docs
Automatic merge from submit-queue
rkt: bump rkt version to 1.2.1
Upon bumping the rkt version, `--hostname` is supported. Also we now gets the configs from the rkt api service, so `stage1-image` is deprecated.
cc @yujuhong @Random-Liu
Automatic merge from submit-queue
Make kubelet use an arch-specific pause image depending on GOARCH
Related to: #22876, #22683 and #15140
@ixdy @pwittrock @brendandburns @mikedanese @yujuhong @thockin @zmerlynn
`--kubelet-cgroups` and `--system-cgroups` respectively.
Updated `--runtime-container` to `--runtime-cgroups`.
Cleaned up most of the kubelet code that consumes these flags to match
the flag name changes.
Signed-off-by: Vishnu kannan <vishnuk@google.com>
* Metrics will not be expose until they are hooked up to a handler
* Metrics are not cached and expose a dos vector, this must be fixed before release or the stats should not be exposed through an api endpoint
This is part of migrating kubelet configuration to the componentconfig api
group and is preliminary to retrofitting client configuration and
implementing full fledged API group mechinary.
Signed-off-by: Mike Danese <mikedanese@google.com>
Add `kube-reserved` and `system-reserved` flags for configuration
reserved resources for usage outside of kubernetes pods. Allocatable is
provided by the Kubelet according to the formula:
```
Allocatable = Capacity - KubeReserved - SystemReserved
```
Also provides a method for estimating a reasonable default for
`KubeReserved`, but the current implementation probably is low and needs
more tuning.
x1957