This is a fork of planetlabs/draino
. This fork is public because our primary intention was to contribute to the original project. With time the original project was not maintained anymore (difficulties to have reviews, and changes not integrated). This code as diverged a lot compare to the original project. Since the initial design is blocking us for some evolutions, we will soon stop maintaining that public fork soon (before end of 2022). We will maybe come back with a new equivalent project that we be owned and maintained by datadog, with more feature in te domain of workload eviction and node replacement
Draino automatically drains Kubernetes nodes based on labels and node
conditions. Nodes that match all of the supplied labels and any of the
supplied node conditions will be tainted immediately and drained after a
configurable drain-buffer
time.
Draino is intended for use alongside the Kubernetes Node Problem Detector and Cluster Autoscaler. The Node Problem Detector can set a node condition when it detects something wrong with a node - for instance by watching node logs or running a script. The Cluster Autoscaler can be configured to delete nodes that are underutilised. Adding Draino to the mix enables autoremediation:
- The Node Problem Detector detects a permanent node problem and sets the corresponding node condition.
- Draino notices the node condition. When the node is eligible (priorities/filters) it taints the node to prevent new pods being scheduled there, and schedules a drain of the node.
- Once the node has been drained the Cluster Autoscaler will consider it underutilised. It will be eligible for scale down (i.e. termination) by the Autoscaler after a configurable period of time.
Usage:
[flags]
[command]
Available Commands:
completion Generate the autocompletion script for the specified shell
group
help Help about any command
log
node
version
Flags:
--candidate-emptydir-pods Evict pods with local storage, i.e. with emptyDir volumes. (default true)
--cloud-provider string cloud provider where the application/controller is running
--cloud-provider-project string cloud provider project where the application/controller is running. Only make sense for gcp
--config-name string Name of the draino configuration
--context string kubernetes context
--cordon-protected-pod-annotation strings Protect nodes hosting pods with this annotation from being candidate. May be specified multiple times. KEY[=VALUE]
--datacenter string datacenter where the application/controller is running
--debug Run with debug logging.
--do-not-cordon-pod-controlled-by strings Do not make candidate nodes hosting pods that are controlled by the designated kind, empty VALUE for uncontrolled pods, May be specified multiple times. kind[[.version].group]] examples: StatefulSets StatefulSets.apps StatefulSets.apps.v1 (default [,StatefulSet])
--do-not-evict-pod-controlled-by strings Do not evict pods that are controlled by the designated kind, empty VALUE for uncontrolled pods, May be specified multiple times: kind[[.version].group]] examples: StatefulSets StatefulSets.apps StatefulSets.apps.v1 (default [,StatefulSet,DaemonSet])
--drain-buffer duration Delay to respect between end of previous drain (success or error) and a new attempt within a drain-group. (default 10m0s)
--drain-buffer-configmap-name string The name of the configmap used to persist the drain-buffer values. Default will be draino-<config-name>-drain-buffer.
--drain-group-labels string Comma separated list of label keys to be used to form draining groups. KEY1,KEY2,...
--drain-rate-limit-burst int Maximum number of parallel drains within a timeframe (default 1)
--drain-rate-limit-qps float32 Maximum number of node drains per seconds per condition (default 0.016666668)
--drain-sim-rate-limit-ratio float32 Which ratio of the overall kube client rate limiting should be used by the drain simulation. 1.0 means that it will use the same. (default 0.7)
--dry-run Emit an event without tainting or draining matching nodes.
--duration-before-replacement duration Max duration we are waiting for a node with Completed drain status to be removed before asking for replacement. (default 1h0m0s)
--encoding string output logs; one of json, json-kube, console (default "json-kube")
--event-aggregation-period duration Period for event generation on kubernetes object. (default 15m0s)
--evict-emptydir-pods Evict pods with local storage, i.e. with emptyDir volumes.
--eviction-headroom duration Additional time to wait after a pod's termination grace period for it to have been deleted. (default 30s)
--exclude-sts-on-node-without-storage To ensure backward compatibility with draino v1, we have to exclude pod of STS running on node without local-storage (default true)
--excluded-pod-per-node-estimation int Estimation of the number of pods that should be excluded from nodes. Used to compute some event cache size. (default 5)
--group-runner-period duration Period for running the group runner (default 10s)
-h, --help help for this command
--informer-namespace string restricts the manager's cache to watch objects in the desired namespace Defaults to all namespaces
--informer-sync-period duration minimum frequency at which watched resources are reconciled (default 1h0m0s)
--klog-verbosity int32 Verbosity to run klog at (default 4)
--kube-address string kube apiserver address (optional)
--kube-client-burst int Burst to use in kube client (default 10)
--kube-client-qps float32 QPS to use in kube client (default 5)
--kube-cluster-name string name of the kubernetes cluster where the application/controller is running
--kube-config string kubeconfig file; empty defaults to in-cluster config
--kubeconfig string Path to kubeconfig file. Leave unset to use in-cluster config.
--leader-elect whether or not to use leader election when starting the manager (default true)
--leader-elect-id string name of the configmap that leader election will use for holding the leader lock
--leader-elect-lease duration non-leader candidates will wait to force acquire leadership. This is measured against time of last observed ack (default 15s)
--leader-elect-namespace string namespace in which the leader election configmap will be created
--leader-elect-renew duration acting master will retry refreshing leadership before giving up (default 10s)
--leader-elect-retry duration clients should wait between tries of actions (default 2s)
--leader-resource-lock string type of resource that leader election will use for holding the leader lock (default "configmaps")
--listen string Address at which to expose /metrics and /healthz. (default ":10002")
--log-development development mode for logs. This disables the sampling and allows for negative level (beyond Debug that is (-1))
--log-events Indicate if events sent to kubernetes should also be logged (default true)
--log-level string log level; one of debug, info, warn, error, dpanic, panic, fatal (default "info")
--log-stacktrace string log stacktrace; one of debug, info, warn, error, dpanic, panic, fatal (default "dpanic")
--master string Address of Kubernetes API server. Leave unset to use in-cluster config.
--max-drain-attempts-before-fail int Maximum number of failed drain attempts before giving-up on draining the node. (default 8)
--max-node-replacement-per-hour int Maximum number of nodes per hour for which draino can ask replacement. (default 2)
--max-notready-nodes strings Maximum number of NotReady nodes in the cluster. When exceeding this value draino stop taking actions. (Value|Value%)
--max-notready-nodes-period duration Polling period to check all nodes readiness (default 1m0s)
--max-pending-pods strings Maximum number of Pending Pods in the cluster. When exceeding this value draino stop taking actions. (Value|Value%)
--max-pending-pods-period duration Polling period to check volume of pending pods (default 1m0s)
--min-eviction-timeout duration Minimum time we wait to evict a pod. The pod terminationGracePeriod will be used if it is bigger. (default 8m0s)
--namespace string namespace where the application/controller is running
--no-legacy-node-handler Deactivate draino legacy node handler
--node-conditions stringArray Nodes for which any of these conditions are true will be tainted and drained.
--node-label strings (Deprecated) Nodes with this label will be eligible for tainting and draining. May be specified multiple times
--node-label-expr string Nodes that match this expression will be eligible for tainting and draining.
--opt-in-pod-annotation strings Pod filtering out is ignored if the pod holds one of these annotations. In a way, this makes the pod directly eligible for draino eviction. May be specified multiple times. KEY[=VALUE]
--pod-warmup-delay-extension duration Extra delay given to the pod to complete is warmup phase (all containers have passed their startProbes) (default 30s)
--pre-activity-default-timeout duration Default duration to wait, for a pre activity to finish, before aborting the drain. This can be overridden by an annotation. (default 10m0s)
--preprovisioning-by-default Set this flag to activate pre-provisioning by default for all nodes
--preprovisioning-check-period duration Period to check if a node has been preprovisioned (default 30s)
--preprovisioning-timeout duration Timeout for a node to be preprovisioned before draining (default 1h20m0s)
--protected-pod-annotation strings Protect pods with this annotation from eviction. May be specified multiple times. KEY[=VALUE]
--pvc-management-by-default PVC management is automatically activated for a workload that do not use eviction++
--reset-config-labels Reset the scope label on the nodes
--retry-backoff-delay duration Additional delay to add between retry schedules. (default 23m0s)
--scope-analysis-period duration Period to run the scope analysis and generate metric (default 5m0s)
--service-addr string http endpoint for the services (default "0.0.0.0:8484")
--service-shutdown-timeout duration shutdown timeout for service (default 15s)
--service-with-healthcheck Activate the healthcheck handlers (default true)
--service-with-metrics Activate the metrics handler (default true)
--service-with-profiling Activate the profiling handler (default true)
--short-lived-pod-annotation strings Pod that have a short live, just like job; we prefer let them run till the end instead of evicting them; node is cordon. May be specified multiple times. KEY[=VALUE]
--skip-drain Whether to skip draining nodes after tainting.
--storage-class-allows-pv-deletion strings Storage class for which persistent volume (and associated claim) deletion is allowed. May be specified multiple times.
--tracer-addr string tracer server address; empty to disable
--tracer-service-name string set a tracer default service name; optional
--wait-before-draining duration Time to wait between moving a node in candidate status and starting the actual drain. (default 30s)
Draino allows filtering the elligible set of nodes using --node-label
and --node-label-expr
.
The original flag --node-label
is limited to the boolean AND of the specified labels. To express more complex predicates, the new --node-label-expr
flag allows for mixed OR/AND/NOT logic via https://github.com/antonmedv/expr.
An example of --node-label-expr
:
(metadata.labels.region == 'us-west-1' && metadata.labels.app == 'nginx') || (metadata.labels.region == 'us-west-2' && metadata.labels.app == 'nginx')
It is possible to prevent eviction of pods that are under control of:
- daemonset
- statefulset
- Custom Resource
- ...
or not even on control of anything. For this, use the flag do-not-evict-pod-controlled-by
; it can be repeated. An empty value means that we block eviction on pods that are uncontrolled.
The value can be a kind
or a kind.group
or a kind.version.group
to designate the owner resource type. If the version
or/and the group
are omitted it acts as a wildcard (any version, any group). It is case-sensitive and must match the API Resource definition.
Example:
- --do-not-evict-controlled-by=StatefulSet
- --do-not-evict-controlled-by=DaemonSet
- --do-not-evict-controlled-by=ExtendedDaemonSet.v1alpha1.datadoghq.com
- --do-not-evict-controlled-by=
Draino is automatically built from master and pushed to the Docker Hub.
Builds are tagged planetlabs/draino:$(git rev-parse --short HEAD)
.
Note: As of September, 2020 we no longer publish planetlabs/draino:latest
in order to encourage explicit and pinned releases.
An example Kubernetes deployment manifest is provided.
Draino provides a simple healthcheck endpoint at /healthz
and Prometheus
metrics at /metrics
. The following metrics exist:
$ kubectl -n kube-system exec -it ${DRAINO_POD} -- apk add curl
$ kubectl -n kube-system exec -it ${DRAINO_POD} -- curl http://localhost:10002/metrics
# HELP draino_drained_nodes_total Number of nodes drained.
# TYPE draino_drained_nodes_total counter
draino_drained_nodes_total{result="succeeded"} 1
draino_drained_nodes_total{result="failed"} 1
Draino is generating event for every relevant step of the eviction process.
When a drain is scheduled, on top of the event, a condition is added to the status of the node. This condition will hold information about the beginning and the end of the drain procedure. This is something that you can see by describing the node resource:
> kubectl describe node {node-name}
......
Unschedulable: true
Conditions:
Type Status LastHeartbeatTime LastTransitionTime Reason Message
---- ------ ----------------- ------------------ ------ -------
OutOfDisk False Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:01:59 +0100 KubeletHasSufficientDisk kubelet has sufficient disk space available
MemoryPressure False Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:01:59 +0100 KubeletHasSufficientMemory kubelet has sufficient memory available
DiskPressure False Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:01:59 +0100 KubeletHasNoDiskPressure kubelet has no disk pressure
PIDPressure False Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:01:59 +0100 KubeletHasSufficientPID kubelet has sufficient PID available
Ready True Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:02:09 +0100 KubeletReady kubelet is posting ready status. AppArmor enabled
ec2-host-retirement True Fri, 20 Mar 2020 15:23:26 +0100 Fri, 20 Mar 2020 15:23:26 +0100 NodeProblemDetector Condition added with tooling
DrainScheduled True Fri, 20 Mar 2020 15:50:50 +0100 Fri, 20 Mar 2020 15:23:26 +0100 Draino Drain activity scheduled 2020-03-20T15:50:34+01:00
Later when the drain activity will be completed the condition will be amended letting you know if it succeeded of failed:
> kubectl describe node {node-name}
......
Unschedulable: true
Conditions:
Type Status LastHeartbeatTime LastTransitionTime Reason Message
---- ------ ----------------- ------------------ ------ -------
OutOfDisk False Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:01:59 +0100 KubeletHasSufficientDisk kubelet has sufficient disk space available
MemoryPressure False Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:01:59 +0100 KubeletHasSufficientMemory kubelet has sufficient memory available
DiskPressure False Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:01:59 +0100 KubeletHasNoDiskPressure kubelet has no disk pressure
PIDPressure False Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:01:59 +0100 KubeletHasSufficientPID kubelet has sufficient PID available
Ready True Fri, 20 Mar 2020 15:52:41 +0100 Fri, 20 Mar 2020 14:02:09 +0100 KubeletReady kubelet is posting ready status. AppArmor enabled
ec2-host-retirement True Fri, 20 Mar 2020 15:23:26 +0100 Fri, 20 Mar 2020 15:23:26 +0100 NodeProblemDetector Condition added with tooling
DrainScheduled True Fri, 20 Mar 2020 15:50:50 +0100 Fri, 20 Mar 2020 15:23:26 +0100 Draino Drain activity scheduled 2020-03-20T15:50:34+01:00 | Completed: 2020-03-20T15:50:50+01:00
If the drain had failed the condition line would look like:
DrainScheduled True Fri, 20 Mar 2020 15:50:50 +0100 Fri, 20 Mar 2020 15:23:26 +0100 Draino Drain activity scheduled 2020-03-20T15:50:34+01:00| Failed:2020-03-20T15:55:50+01:00
In some cases the drain activity may failed because of restrictive Pod Disruption Budget or any other reason external to Draino. The node remains cordon
and the drain condition
is marked as Failed
. If you want to reschedule a drain tentative on that node, add the annotation: draino/drain-retry: true
. A new drain schedule will be created. Note that the annotation is not modified and will trigger retries in loop in case the drain fails again.
kubectl annotate node {node-name} draino/drain-retry=true
A node replacement is automatically requested by draino
if a node is marked with drain completed for a duration longer than --duration-before-replacement=1h0m0s
. This behavior allows us to unlock situation where the CA cannot collect the drained node due to minSize=1 on the Nodegroup. This node replacement feature is throttle thanks to parameter --max-node-replacement-per-hour=2
The user can proactively ask for pre-provisioning a replacement node before drain is actually started by putting the following annotation on the node: node-lifecycle.datadoghq.com/provision-new-node-before-drain=true
Draino can be run in dry run mode using the --dry-run
flag.
Draino can take care of deleting the PVs/PVCs associated with the evicted pod. This is interesting especially for pods using the local-storage
storage class. Since the old nodes are not eligible for scheduling, the PVCs/PVs of the pod must be recycled to ensure that the pods can be scheduled on another node.
The list of eligible storage classes must be given to draino at start-up: --storage-class-allows-pv-deletion=local-data
. This flag can be repeated if multiple classes are eligible.
Then each pods has to explicitly opt-in for that data deletion using an annotation: draino/delete-pvc-and-pv=true