You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently we only replace one failed Pod when we enable the feature. I think it makes sense to allow the user to specify how many Pods should/can be replaced.
The text was updated successfully, but these errors were encountered:
johscheuer
added
the
reliability
Protections against unsafe behavior, blockers go reconciliation, or other reliability concerns
label
Apr 9, 2021
Think about that a little bit we would implement that as some kind of a max inflight replacements which would be the upper limit of replacements for the automatic replacements. If the user would specify 5 that means the controller will evaluate how many process groups are currently marked for removal and if the limit is not reached the controller can replace as many process groups until the limit is reached. A concrete example a user sets the limit to 5 and the operator has 3 ongoing replacements, that means the operator will replace at most 2 process groups.
Currently we only replace one failed Pod when we enable the feature. I think it makes sense to allow the user to specify how many Pods should/can be replaced.
The text was updated successfully, but these errors were encountered: