Bug 2006407

Summary:	pvc from azuredisk taking time to attach to the node
Product:	OpenShift Container Platform	Reporter:	Rajnikant <rkant>
Component:	Storage	Assignee:	aos-storage-staff <aos-storage-staff>
Storage sub component:	Storage	QA Contact:	Wei Duan <wduan>
Status:	CLOSED DUPLICATE	Docs Contact:
Severity:	high
Priority:	high	CC:	adeshpan, aos-bugs, fbertina, jsafrane
Version:	4.7	Keywords:	Reopened
Target Milestone:	---
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2021-11-30 15:23:03 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Rajnikant 2021-09-21 16:42:54 UTC

Description of problem:
pvc from azuredisk taking time to attach to the node. 

Version-Release number of selected component (if applicable):
OpenShift Container Platform Version 4.7

How reproducible:

It is happening for newly created pvc volume. 

From kubelet logs saw following repetitave logs for ~4 min
---
operationExecutor.VerifyControllerAttachedVolume started for volume "pvc-xxxxx"

Error: "Volume has not been added to the list of VolumesInUse in the node's volume status for volume \"pvc-xxxxx\"
---

Saw once volume attach to the node it mounted immediately.

---
Controller attach succeeded for volume pvc-xxxx
operationExecutor.MountVolume started for volume pvc-xxx
MountVolume.WaitForAttach succeeded for volume 


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

~                                                                                                                                                                                                                  
~                                                                                                                                                                                                                  
~

Comment 9 Jan Safranek 2021-11-30 15:20:49 UTC

Idea: in VolumeManager reconciler, add an informer for Node object and check the informer Node object volumesAttached (every 100 ms) *before* trying the first VerifyControllerAttachedVolume. This way, we start the exp. backoff when there is a good chance the volume is already attached.

Comment 10 Jan Safranek 2021-11-30 15:23:03 UTC

This is already tracked as 2023500, let's continue there.

*** This bug has been marked as a duplicate of bug 2023500 ***