-
Notifications
You must be signed in to change notification settings - Fork 14.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Document impact of DRA on scheduling #43907
Conversation
👷 Deploy Preview for kubernetes-io-vnext-staging processing.
|
/lgtm |
LGTM label has been added. Git tree hash: 7315572cb02f166d75e4c522bafa581c47c64999
|
/sig node |
/retitle Document impact of DRA on scheduling |
{{< warning >}} | ||
|
||
Scheduling of pods which use ResourceClaims is going to be slower because of | ||
the additional communication that is required. Beware that this may also impact | ||
pods that don't use ResourceClaims because only one pod at a time gets | ||
scheduled, blocking API calls are made while handling a pod with | ||
ResourceClaims, and thus scheduling the next pod gets delayed. | ||
|
||
{{< /warning >}} |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Try this:
{{< warning >}} | |
Scheduling of pods which use ResourceClaims is going to be slower because of | |
the additional communication that is required. Beware that this may also impact | |
pods that don't use ResourceClaims because only one pod at a time gets | |
scheduled, blocking API calls are made while handling a pod with | |
ResourceClaims, and thus scheduling the next pod gets delayed. | |
{{< /warning >}} | |
{{< note >}} | |
Scheduling of pods that use ResourceClaims is going to be slower because of | |
the additional communication that is required. Beware that using ResourceClaims | |
in your cluster may also impact scheduling of pods that don't use ResourceClaims. | |
This impact happens because the kube-scheduler schedules one pod at a time; | |
blocking API calls are made while handling a pod with ResourceClaims, and thus | |
scheduling for other Pods in the scheduling queue is also delayed. | |
{{< /note >}} |
Don't use a warning
; it's not the best fit for this level of advice.
Once DRA is beta, we might want to add a one-line mention in https://kubernetes.io/docs/concepts/workloads/pods/pod-lifecycle/ about potential sources of scheduling delays, and link to this page for more detail of how DRA can contribute there.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Changed to note
and created an issue to track updating pod-lifecycle
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
(the suggestion also proposed a wording tweak, though it's not immediately obvious from the diff)
That pods with ResourceClaims get scheduled more slowly, but that this also affects other pods may be surprising and is worth calling out.
252dc35
to
725f68f
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
/lgtm
/approve
We'd prefer to have this text updated for beta to match the docs style guide.
LGTM label has been added. Git tree hash: baa251f2148d1908cdb3b3137b917d2823833de4
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: sftim The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
That pods with ResourceClaims get scheduled more slowly, but that this also affects other pods may be surprising and is worth calling out.