Bug 797997 - stale files in indices/ directory
Summary: stale files in indices/ directory
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: GlusterFS
Classification: Community
Component: unclassified
Version: mainline
Hardware: Unspecified
OS: Unspecified
high
high
Target Milestone: ---
Assignee: Pranith Kumar K
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2012-02-27 18:36 UTC by Amar Tumballi
Modified: 2013-12-19 00:07 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2012-03-05 12:24:50 UTC
Regression: ---
Mount Type: fuse
Documentation: ---
CRM:
Verified Versions:


Attachments (Terms of Use)

Description Amar Tumballi 2012-02-27 18:36:03 UTC
Description of problem:
Ran Posix-compliance test on a newly created/mounted volume, deleted all the empty directories of posix tests, so that there are no files or directories on mount point. But on the backend saw some stale files remaining in hidden directory.


Version-Release number of selected component (if applicable):
master (git head : c2c82c360fda20533093c27c82ce0725cf40cab5)

How reproducible:
3/3

Steps to Reproduce:
1. create 2x2 dist-replica volume; enable quota;
2. start the volume; mount it;
3. run posix tests; remove all the entries from the mountpoint after tests..
4. ls -l /export/dir/.glusterfs/*/*/* shows stale entries.
  
Actual results:
stale files seen

Expected results:
no files should be in those directories

Comment 1 Pranith Kumar K 2012-03-05 12:24:50 UTC
  Index xlator does not delete the index file on unlink/rmdir.
   The stale entries will be there only till the next index-self-heal (every 10 minutes) is performed on it.
One thing to note is that, for the self-heal to happen atleast 2 subvolumes of the replica subvols should be up.

[root@pranithk-laptop repl2]# ls /gfs/repl2_?/.glusterfs/indices/xattrop | wc -l
380
[root@pranithk-laptop repl2]# sleep 600 && ls /gfs/repl2_?/.glusterfs/indices/xattrop | wc -l << one of the subvols is down.
380
[root@pranithk-laptop repl2]# gluster volume start repl2 force
Starting volume repl2 has been successful
[root@pranithk-laptop repl2]# sleep 600 && ls /gfs/repl2_?/.glusterfs/indices/xattrop | wc -l << both the subvols are up.
3
[root@pranithk-laptop repl2]# ls /gfs/repl2_?/.glusterfs/indices/xattrop
/gfs/repl2_0/.glusterfs/indices/xattrop:

/gfs/repl2_1/.glusterfs/indices/xattrop:
[root@pranithk-laptop repl2]# 

Pranith


Note You need to log in before you can comment on or make changes to this bug.