Bug 1399050

Summary: [GSS] High CPU usage for glusterfsd process when thin LV is 100% full
Product: Red Hat Gluster Storage Reporter: Bipin Kunal <bkunal>
Component: glusterfsAssignee: Bug Updates Notification Mailing List <rhs-bugs>
Status: CLOSED NOTABUG QA Contact: Byreddy <bsrirama>
Severity: medium Docs Contact:
Priority: medium    
Version: rhgs-3.1CC: amukherj, bkunal, rhs-bugs, storage-qa-internal, vbellur
Target Milestone: ---   
Target Release: RHGS 3.2.0   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-12-06 10:11:48 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:

Description Bipin Kunal 2016-11-28 07:35:44 UTC 
Description of problem:
High CPU usage for glusterfsd process when thin LV is 100% full. This can lead to crash and should be handled.


Version-Release number of selected component (if applicable):
RHGS-3.1
glusterfs-server-3.7.9-12.el7rhgs.x86_64

Reproducible:

Not very easy to reproduce as we need 100% LV full with FS space available.
Theoretically we should be able to reproduce this issue. As LV in question here is thin LV, on deleting(running rm) a file, filesystem usage will be reclaimed but LV usage won't decrease. Thus we can end up in a situation where LV is full but FS still has ample to space.

Steps to Reproduce:
1. Create a volume
2. Fill it with data such that brick LV is 100% full but FS still has space left 


Actual results:
Brick process CPU consumption goes high when LV reaches 100% full.

Expected results:
100% LV usage should be intercepted for XFS and ultimately by gluster and appropriate action should be performed.

Additional info:

I am not sure if this is a expected behavior or this is a bug. But it is worth looking into issue once