Remove resync period for disruption controller

What type of PR is this?
/kind cleanup

What this PR does / why we need it:
The disruption controller is resyncing all ssets every 30 seconds, this is not necessary, and make the depth of disruption workqueue longer and can cause delays processing actual updates when large amounts of disruptions exist.

Special notes for your reviewer:

Does this PR introduce a user-facing change?
Disruption controllers no longer force a resync every 30 seconds when nothing has changed.
This commit is contained in:
yuanhuaiwang 2020-07-06 20:31:28 +08:00 committed by GitHub
parent 865cbf0bdf
commit 8e1e1e9be2
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -140,13 +140,12 @@ func NewDisruptionController(
dc.podLister = podInformer.Lister()
dc.podListerSynced = podInformer.Informer().HasSynced
pdbInformer.Informer().AddEventHandlerWithResyncPeriod(
pdbInformer.Informer().AddEventHandler(
cache.ResourceEventHandlerFuncs{
AddFunc: dc.addDb,
UpdateFunc: dc.updateDb,
DeleteFunc: dc.removeDb,
},
30*time.Second,
)
dc.pdbLister = pdbInformer.Lister()
dc.pdbListerSynced = pdbInformer.Informer().HasSynced