Bug 2207567

Summary:	Filesystem: Improve stopping for large filesystems (RHEL9)
Product:	Red Hat Enterprise Linux 9	Reporter:	Oyvind Albrigtsen <oalbrigt>
Component:	resource-agents	Assignee:	Oyvind Albrigtsen <oalbrigt>
Status:	VERIFIED ---	QA Contact:	cluster-qe <cluster-qe>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	9.2	CC:	agk, cluster-maint, fdinitto, mjuricek, oalbrigt, phagara, pzimek, sbradley
Target Milestone:	rc	Keywords:	Triaged
Target Release:	9.3
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	resource-agents-4.10.0-43.el9	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:	2189243	Environment:
Last Closed:		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:
Bug Depends On:	2189242, 2189243
Bug Blocks:

Description Oyvind Albrigtsen 2023-05-16 10:06:41 UTC

+++ This bug was initially created as a clone of Bug #2189243 +++

+++ This bug was initially created as a clone of Bug #2189242 +++

Description of problem:

On high-end production workload systems with huge amount of (write-cache) RAM and big XFS file systems >= 8 TiB the unmount operation itself may take longer then 10 minutes on each attempt (even if it fails as processes are still utilizing it). In case login shells of users sit on the Filesystem resource then these do no respond  to SIGTERM, just to SIGHUP so when resource is stopping it deliberately fails to unmount and causes stop operation to fail/timeout.


Version-Release number of selected component (if applicable):

resource-agents-4.1.1-61.el7_9.15.x86_64

How reproducible:
repeatedly


Steps to Reproduce:
1. Create large filesystem resource with potentially long dirty unmount cycles (+- 30 minutes) with login shells on it
2. re-login during the long stop operation (login shells on any HA FS RA managed file system does not fail all standard FS RA stop operation)


Actual results:
unmount fails resulting stop operation to fail

Expected results:
unmount succeeds

Additional info:
https://github.com/ClusterLabs/resource-agents/pull/1868

Comment 2 Oyvind Albrigtsen 2023-07-12 10:43:33 UTC

New PR based on suggestions by QE: https://github.com/ClusterLabs/resource-agents/pull/1878