Bug 1578353

Summary:	Persist the max_retries xfs settings
Product:	[Red Hat Storage] Red Hat Gluster Storage	Reporter:	SATHEESARAN <sasundar>
Component:	rhhi	Assignee:	Sahina Bose <sabose>
Status:	CLOSED WONTFIX	QA Contact:	SATHEESARAN <sasundar>
Severity:	high	Docs Contact:
Priority:	high
Version:	rhhiv-1.5	CC:	amukherj, bgurney, dkeefe, dkota, rcyriac, rhs-bugs, rreddy, sabose, sankarshan, sasundar, smohan, storage-qa-internal, surs
Target Milestone:	---
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:	1523562
Clones:	1578354 (view as bug list)		Environment:
Last Closed:	2018-05-29 14:26:31 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:
Bug Depends On:	1523562, 1578354
Bug Blocks:

Description SATHEESARAN 2018-05-15 11:30:35 UTC

Description of problem:
-----------------------
According to Dennis Keefe,

 Use these xfs settings for XFS in RHHI, so that XFS handle conditions when VDO runs out of physical space.

echo 4 | tee /sys/fs/xfs/dm-13/error/metadata/ENOSPC/max_retries
echo 4 | tee /sys/fs/xfs/dm-13/error/metadata/EIO/max_retries

Gdeploy bug ( BZ 1523562 ) was fixed to solve this request, but it turned out to be that these values are not persisted in the config files - max_retries

This bug is required to make these changes persistent.

Comment 1 SATHEESARAN 2018-05-16 11:22:51 UTC

Sahina,

As explained towards the end of comment0, the original issue is solved, where the max_retries under ENOSPC and EIO were updated to 4. But these values are not persisted. So opened this bug, but by mistake, did the cloning and missed to change the subject and context

Comment 2 Sahina Bose 2018-05-17 05:58:39 UTC

We need a change in RHEL to persist this setting. Sas, have you opened a corresponding bug for this?

Comment 3 SATHEESARAN 2018-05-17 07:47:14 UTC

(In reply to Sahina Bose from comment #2)
> We need a change in RHEL to persist this setting. Sas, have you opened a
> corresponding bug for this?

Sahina,

Yes, there is already a RHEL XFS bug opened, but connected as 'depends on' on this bug. I have made the required changes now to reflect that on this bug

Here is the bug - https://bugzilla.redhat.com/show_bug.cgi?id=1578354

Comment 4 Bryan Gurney 2018-05-17 14:53:35 UTC

I did some testing in RHEL 7.5, and found that the default settings of XFS no longer retry forever, which means that the "error/metadata/ENOSPC" and "error/metadata/EIO" settings no longer need to be made on each mount.

After an "out of physical space" condition, all that needs to be done is umount, add space to the backing device of VDO (i.e.: resize partition, lvextend, etc.), run "vdo --growPhysical", then mount the filesystem on the VDO volume, and run xfs_growfs.

Therefore, there's no need to persist the setting in RHEL 7.5.

Comment 5 Sahina Bose 2018-05-29 14:26:31 UTC

Closing as this is no longer required