OCPBUGS-48479: Adding MHC exception to Pausing MHC cluster update … #91781
Conversation
openshift-ci-robot
added
jira/valid-reference
|
@obrown1205: This pull request references Jira Issue OCPBUGS-48479, which is invalid:
Comment The bug has been updated to refer to the pull request using the external bug tracker. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository. |
obrown1205
changed the title
obrown1205
changed the title
openshift-ci
bot
added
the
size/XS
|
🤖 Tue Apr 08 14:32:42 - Prow CI generated the docs preview: https://91781--ocpdocs-pr.netlify.app/openshift-enterprise/latest/disconnected/updating/disconnected-update.html |
|
|
||
| [NOTE] | ||
| ==== | ||
| As an exception, you do not have to pause the `machine-api-termination-handler` resource, as it does not deploy a new node if the node has been flagged as `notReady`. |
There was a problem hiding this comment.
instead of naming a particular MachineHealthCheck, maybe call out the property of the MHC that means you don't need to pause it? In this case, I expect it's using the fatal Terminating as its unhealthyConditions, so something like:
As an exception, you do not have to pause MachineHealthChecks where the
unhealthyConditionsis a fatal condition likeTerminatingbeingTrue. Those Machines are toast anyway, so no use waiting before reaping.
or whatever the public-docs analog of that is?
There was a problem hiding this comment.
Basically LGTM, but I'd like to see how to address @wking 's question.
There was a problem hiding this comment.
Makes sense :) TY!
There was a problem hiding this comment.
@wking how about:
"Some MachineHealthChecks might not need to be paused. If your MachineHealthCheck (MHC) resource has a fatal condition met, new nodes cannot be deployed, and pausing that MHC is unnecessary."
obrown1205
force-pushed
the
OCPBUGS-48479
branch
from
ee39a44 to
26ea783
Compare
obrown1205
force-pushed
the
OCPBUGS-48479
branch
from
26ea783 to
ed84141
Compare
|
|
||
| [NOTE] | ||
| ==== | ||
| Some MachineHealthChecks might not need to be paused. If your MachineHealthCheck (MHC) resource has a fatal condition met, new nodes cannot be deployed, and pausing that MHC is unnecessary. |
There was a problem hiding this comment.
🤖 [error] RedHat.TermsErrors: Use 'unrecoverable' rather than 'fatal'. For more information, see RedHat.TermsErrors.
There was a problem hiding this comment.
Nit: maybe just drop "new nodes cannot be deployed", because it's not about new Nodes watched by the MHC, it's about the Node that the MHC is sad about being terminal. Maybe something like:
If your MachineHealthCheck (MHC) resource relies on unrecoverable conditions, pausing that MHC is unnecessary.
|
@obrown1205: all tests passed! Full PR test history. Your PR dashboard. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here. |
…step
Updating to add an exception note for one of the MachineHealthCheck resources that does not need to be paused during this step.
Version(s):
4.14+
Issue:
OCPBUGS-48479
Link to docs preview:
https://91781--ocpdocs-pr.netlify.app/openshift-enterprise/latest/disconnected/updating/disconnected-update.html
https://91781--ocpdocs-pr.netlify.app/openshift-enterprise/latest/updating/updating_a_cluster/updating-cluster-cli.html
QE review:
Additional information: