avoid hot-tier SLA to be tied to warm-tier SLA (#18581)

it is okay if the warm-tier cannot keep up, we should continue
to take I/O at hot-tier, only fail hot-tier or block it when
we are disk full.

Bonus: add metrics counter for these missed tasks, we will
know for sure if one of the node is lagging behind or is
losing too many tasks during transitioning.
This commit is contained in:
Harshavardhana
2023-12-02 13:02:12 -08:00
committed by GitHub
parent f2d063e7b9
commit e98172d72d
4 changed files with 226 additions and 214 deletions

View File

@@ -1190,7 +1190,7 @@ func applyTransitionRule(event lifecycle.Event, src lcEventSrc, obj ObjectInfo)
if obj.DeleteMarker {
return false
}
globalTransitionState.queueTransitionTask(obj, event, src, false)
globalTransitionState.queueTransitionTask(obj, event, src)
return true
}