Skip to content

Commit

Permalink
chore: move terminationGracePeriod from NodePool.spec.disruption to N…
Browse files Browse the repository at this point in the history 
…odeClaim.spec

Signed-off-by: wmgroot <wmgroot@gmail.com>
  • Loading branch information
@wmgroot
wmgroot committed Jul 11, 2024
1 parent 164cfe3 commit 8a58b20
Show file tree
Hide file tree
Showing 22 changed files with 470 additions and 193 deletions.
40 changes: 40 additions & 0 deletions kwok/charts/crds/karpenter.sh_nodeclaims.yaml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -248,6 +248,26 @@ spec:
- key
type: object
type: array
terminationGracePeriod:
description: |-
TerminationGracePeriod is the maximum duration the controller will wait before forcefully deleting the pods on a node, measured from when deletion is first initiated.
Warning: this feature takes precedence over a Pod's terminationGracePeriodSeconds value, and bypasses any blocked PDBs or the karpenter.sh/do-not-disrupt annotation.
This field is intended to be used by cluster administrators to enforce that nodes can be cycled within a given time period.
When set, drifted nodes will begin draining even if there are pods blocking eviction. Draining will respect PDBs and the do-not-disrupt annotation until the TGP is reached.
Karpenter will preemptively delete pods so their terminationGracePeriodSeconds align with the node's terminationGracePeriod.
If a pod would be terminated without being granted its full terminationGracePeriodSeconds prior to the node timeout,
that pod will be deleted at T = node timeout - pod terminationGracePeriodSeconds.
The feature can also be used to allow maximum time limits for long-running jobs which can delay node termination with preStop hooks.
If left undefined, the controller will wait indefinitely for pods to be drained.
type: string
required:
- nodeClassRef
- requirements
Expand Down Expand Up @@ -692,6 +712,26 @@ spec:
- key
type: object
type: array
terminationGracePeriod:
description: |-
TerminationGracePeriod is the maximum duration the controller will wait before forcefully deleting the pods on a node, measured from when deletion is first initiated.
Warning: this feature takes precedence over a Pod's terminationGracePeriodSeconds value, and bypasses any blocked PDBs or the karpenter.sh/do-not-disrupt annotation.
This field is intended to be used by cluster administrators to enforce that nodes can be cycled within a given time period.
When set, drifted nodes will begin draining even if there are pods blocking eviction. Draining will respect PDBs and the do-not-disrupt annotation until the TGP is reached.
Karpenter will preemptively delete pods so their terminationGracePeriodSeconds align with the node's terminationGracePeriod.
If a pod would be terminated without being granted its full terminationGracePeriodSeconds prior to the node timeout,
that pod will be deleted at T = node timeout - pod terminationGracePeriodSeconds.
The feature can also be used to allow maximum time limits for long-running jobs which can delay node termination with preStop hooks.
If left undefined, the controller will wait indefinitely for pods to be drained.
type: string
required:
- nodeClassRef
- requirements
Expand Down
82 changes: 40 additions & 42 deletions kwok/charts/crds/karpenter.sh_nodepools.yaml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -165,27 +165,6 @@ spec:
memory leak protection, and disruption testing.
pattern: ^(([0-9]+(s|m|h))+)|(Never)$
type: string
terminationGracePeriod:
description: |-
TerminationGracePeriod is the duration the controller will wait before forcefully terminating a node, measured from when deletion is first initiated.
Once the GracePeriod has expired, all pods on the node will be shutdown using the official non-graceful shutdown taint.
Warning: this feature takes precedence over a Pod's terminationGracePeriodSeconds value, and bypasses any blocked PDBs or the karpenter.sh/do-not-disrupt annotation.
This field is intended to be used by cluster administrators to enforce that nodes can be cycled within a given time period.
When set, drifted nodes will begin draining even if there are pods blocking eviction. Draining will respect PDBs and do-not-disrupt until the TGP is reached.
Karpenter will preemptively delete pods so their terminationGracePeriodSeconds align with the node's terminationGracePeriod.
If a pod would be terminated without being granted its full terminationGracePeriodSeconds prior to the node timeout,
that pod will be deleted at T = node timeout - pod terminationGracePeriodSeconds.
The feature can also be used to allow maximum time limits for long-running jobs which can delay node termination with preStop hooks.
If left undefined, the controller will wait indefinitely for pods to be drained.
type: string
type: object
x-kubernetes-validations:
- message: consolidateAfter cannot be combined with consolidationPolicy=WhenUnderutilized
Expand Down Expand Up @@ -409,6 +388,26 @@ spec:
- key
type: object
type: array
terminationGracePeriod:
description: |-
TerminationGracePeriod is the maximum duration the controller will wait before forcefully deleting the pods on a node, measured from when deletion is first initiated.
Warning: this feature takes precedence over a Pod's terminationGracePeriodSeconds value, and bypasses any blocked PDBs or the karpenter.sh/do-not-disrupt annotation.
This field is intended to be used by cluster administrators to enforce that nodes can be cycled within a given time period.
When set, drifted nodes will begin draining even if there are pods blocking eviction. Draining will respect PDBs and the do-not-disrupt annotation until the TGP is reached.
Karpenter will preemptively delete pods so their terminationGracePeriodSeconds align with the node's terminationGracePeriod.
If a pod would be terminated without being granted its full terminationGracePeriodSeconds prior to the node timeout,
that pod will be deleted at T = node timeout - pod terminationGracePeriodSeconds.
The feature can also be used to allow maximum time limits for long-running jobs which can delay node termination with preStop hooks.
If left undefined, the controller will wait indefinitely for pods to be drained.
type: string
required:
- nodeClassRef
- requirements
Expand Down Expand Up @@ -642,27 +641,6 @@ spec:
memory leak protection, and disruption testing.
pattern: ^(([0-9]+(s|m|h))+)|(Never)$
type: string
terminationGracePeriod:
description: |-
TerminationGracePeriod is the duration the controller will wait before forcefully terminating a node, measured from when deletion is first initiated.
Once the GracePeriod has expired, all pods on the node will be shutdown using the official non-graceful shutdown taint.
Warning: this feature takes precedence over a Pod's terminationGracePeriodSeconds value, and bypasses any blocked PDBs or the karpenter.sh/do-not-disrupt annotation.
This field is intended to be used by cluster administrators to enforce that nodes can be cycled within a given time period.
When set, drifted nodes will begin draining even if there are pods blocking eviction. Draining will respect PDBs and do-not-disrupt until the TGP is reached.
Karpenter will preemptively delete pods so their terminationGracePeriodSeconds align with the node's terminationGracePeriod.
If a pod would be terminated without being granted its full terminationGracePeriodSeconds prior to the node timeout,
that pod will be deleted at T = node timeout - pod terminationGracePeriodSeconds.
The feature can also be used to allow maximum time limits for long-running jobs which can delay node termination with preStop hooks.
If left undefined, the controller will wait indefinitely for pods to be drained.
type: string
type: object
x-kubernetes-validations:
- message: consolidateAfter cannot be combined with consolidationPolicy=WhenUnderutilized
Expand Down Expand Up @@ -1012,6 +990,26 @@ spec:
- key
type: object
type: array
terminationGracePeriod:
description: |-
TerminationGracePeriod is the maximum duration the controller will wait before forcefully deleting the pods on a node, measured from when deletion is first initiated.
Warning: this feature takes precedence over a Pod's terminationGracePeriodSeconds value, and bypasses any blocked PDBs or the karpenter.sh/do-not-disrupt annotation.
This field is intended to be used by cluster administrators to enforce that nodes can be cycled within a given time period.
When set, drifted nodes will begin draining even if there are pods blocking eviction. Draining will respect PDBs and the do-not-disrupt annotation until the TGP is reached.
Karpenter will preemptively delete pods so their terminationGracePeriodSeconds align with the node's terminationGracePeriod.
If a pod would be terminated without being granted its full terminationGracePeriodSeconds prior to the node timeout,
that pod will be deleted at T = node timeout - pod terminationGracePeriodSeconds.
The feature can also be used to allow maximum time limits for long-running jobs which can delay node termination with preStop hooks.
If left undefined, the controller will wait indefinitely for pods to be drained.
type: string
required:
- nodeClassRef
- requirements
Expand Down
40 changes: 40 additions & 0 deletions pkg/apis/crds/karpenter.sh_nodeclaims.yaml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -246,6 +246,26 @@ spec:
- key
type: object
type: array
terminationGracePeriod:
description: |-
TerminationGracePeriod is the maximum duration the controller will wait before forcefully deleting the pods on a node, measured from when deletion is first initiated.
Warning: this feature takes precedence over a Pod's terminationGracePeriodSeconds value, and bypasses any blocked PDBs or the karpenter.sh/do-not-disrupt annotation.
This field is intended to be used by cluster administrators to enforce that nodes can be cycled within a given time period.
When set, drifted nodes will begin draining even if there are pods blocking eviction. Draining will respect PDBs and the do-not-disrupt annotation until the TGP is reached.
Karpenter will preemptively delete pods so their terminationGracePeriodSeconds align with the node's terminationGracePeriod.
If a pod would be terminated without being granted its full terminationGracePeriodSeconds prior to the node timeout,
that pod will be deleted at T = node timeout - pod terminationGracePeriodSeconds.
The feature can also be used to allow maximum time limits for long-running jobs which can delay node termination with preStop hooks.
If left undefined, the controller will wait indefinitely for pods to be drained.
type: string
required:
- nodeClassRef
- requirements
Expand Down Expand Up @@ -688,6 +708,26 @@ spec:
- key
type: object
type: array
terminationGracePeriod:
description: |-
TerminationGracePeriod is the maximum duration the controller will wait before forcefully deleting the pods on a node, measured from when deletion is first initiated.
Warning: this feature takes precedence over a Pod's terminationGracePeriodSeconds value, and bypasses any blocked PDBs or the karpenter.sh/do-not-disrupt annotation.
This field is intended to be used by cluster administrators to enforce that nodes can be cycled within a given time period.
When set, drifted nodes will begin draining even if there are pods blocking eviction. Draining will respect PDBs and the do-not-disrupt annotation until the TGP is reached.
Karpenter will preemptively delete pods so their terminationGracePeriodSeconds align with the node's terminationGracePeriod.
If a pod would be terminated without being granted its full terminationGracePeriodSeconds prior to the node timeout,
that pod will be deleted at T = node timeout - pod terminationGracePeriodSeconds.
The feature can also be used to allow maximum time limits for long-running jobs which can delay node termination with preStop hooks.
If left undefined, the controller will wait indefinitely for pods to be drained.
type: string
required:
- nodeClassRef
- requirements
Expand Down

0 comments on commit 8a58b20

Please sign in to comment.