Bug 1538358

Summary:	EC users are hitting bugs WRT healing that look to be fixed in upstream.
Product:	[Red Hat Storage] Red Hat Gluster Storage	Reporter:	Ben Turner <bturner>
Component:	disperse	Assignee:	Ashish Pandey <aspandey>
Status:	CLOSED NOTABUG	QA Contact:	Nag Pavan Chilakam <nchilaka>
Severity:	urgent	Docs Contact:
Priority:	unspecified
Version:	rhgs-3.3	CC:	aspandey, bturner, jahernan, rhs-bugs, storage-qa-internal
Target Milestone:	---
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2018-01-31 15:54:13 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 Ben Turner 2018-01-24 22:04:19 UTC

Description of problem:

We think we are hitting the following issues that have been fixed in upstream gluster but haven't made it to RHGS yet:

http://lists.gluster.org/pipermail/gluster-users/2017-August/032225.html
https://review.gluster.org/#/c/16468/

https://bugzilla.redhat.com/show_bug.cgi?id=1475789
https://review.gluster.org/#/c/16772/9/xlators/cluster/ec/src/ec-heal.c

https://bugzilla.redhat.com/show_bug.cgi?id=1427159

We have several customers hitting combinations of some recently fixed EC bugs that look to be fixed in the upstream.  I saw there have been some upstream EC fixes that haven't made it to downstream, I was thinking we could just backport any upstream EC fix that could be relevant.  I am pretty confident we are hitting some combination of these bugs in several different environments.

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

RHGS 3.3.1

How reproducible:

Frequently.

Steps to Reproduce:
1.  Induce healing on an EC volume through disconnects
2.  Stat files / writes new data / append / read 
3.  EC healing doesn't complete due to various issues.

Actual results:

Bricks are unable to fully heal.

Expected results:

Normal operation.


Additional info:

Comment 3 Ashish Pandey 2018-01-25 04:24:12 UTC

*** Bug 1535326 has been marked as a duplicate of this bug. ***