Bug 762701 (GLUSTER-969) - rm -rf hangs the clients
Summary: rm -rf hangs the clients
Keywords:
Status: CLOSED WORKSFORME
Alias: GLUSTER-969
Product: GlusterFS
Classification: Community
Component: locks
Version: 3.0.4
Hardware: All
OS: Linux
low
high
Target Milestone: ---
Assignee: Pavan Vilas Sondur
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2010-05-31 09:07 UTC by Raghavendra Bhat
Modified: 2015-12-01 16:45 UTC (History)
3 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed:
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:

Attachments (Terms of Use)
Add an attachment (proposed patch, testcase, etc.)

Description Raghavendra Bhat 2010-05-31 06:27:33 UTC 
log files and statedumps of servers and clients are in /data/share/tickets/969
Comment 1 Raghavendra Bhat 2010-05-31 09:07:03 UTC 
Setup:

glusterfs-3.0.5rc3-> distributed replicate.
4- server processes
2 clients (with all the performance xlators)

server-1 -> error-gen loaded with ENOMEM for all fops above posix

server-2 -> error-gen loaded with ENOMEM for all fops above locks

server-3 -> error-gen loaded with ENOENT for all fops above io-threads

server-4 -> no error-gen


mount-point 1 -> sanity script started

mount-point 2 -> untarring of linux kernel


doing rm -rf on any of the mountpoints hangs for a long time and then comes back. Doing rm -rf again will repeat the hang. Seems to be an issue with locks as per the glusterstatedump information. The logfiles and the statedump files are attached.
Comment 2 Amar Tumballi 2010-10-05 07:19:45 UTC 
Is it happening now? even though its targeted to be fixed in 3.2, it would be nice to check this behavior with post 3.1qa40
Comment 3 Vijay Bellur 2010-10-25 08:07:48 UTC 
Doesn't happen any more. Please re-open if found.
Note You need to log in before you can comment on or make changes to this bug.