[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Freezing GFS (Re: [Linux-cluster] GFS and CLVM-snapshot)

I'm sorry for a long interval.

AJ Lewis wrote:
> On Thu, May 12, 2005 at 11:28:37AM +0900, Kenji Wakamiya wrote:
>>By the way, sometimes I get warnings from gfs_fsck for snapshot-LV,
>>even though I froze GFS before doing lvcreate -s.
>>For example, the followings are.  Is this usual thing?
>>  # gfs_fsck -y /dev/vg0/lv0ss0
>>  Initializing fsck
>>  Starting pass1
>>  Pass1 complete
>>  Starting pass1b
>>  Pass1b complete
>>  Starting pass1c
>>  Pass1c complete
>>  Starting pass2
>>  Pass2 complete
>>  Starting pass3
>>  Pass3 complete
>>  Starting pass4
>>  Found unlinked inode at 1929233
>>	  Adjusting freemeta block count (59 -> 60).
>>	  Adjusting used dinode block count (10 -> 9).
>>  l+f directory at 29
>>  Added inode #1929233 to l+f dir
>>  Found unlinked inode at 1800635
>>  Added inode #1800635 to l+f dir
>>  Link count inconsistent for inode 25 - 5 6
>>  Link count updated for inode 25
>>  Pass4 complete
>>  Starting pass5
>>  ondisk and fsck bitmaps differ at block 29
>>  Succeeded.
>>  used inode count inconsistent: is 9 should be 10
>>  free meta count inconsistent: is 60 should be 59
>>  Pass5 complete
>>  #
> Interesting.  Have you noticed what inode gets stuck in l+f?

I'm using IlohaMail-0.9-20050415, a PHP based POP/IMAP web mailer
- http://blog.ilohamail.org/ on 3-nodes GFS cluster with a server
load balancer.  The lost file is produced in this way:

1. Reload all three HTML frames or only folder-tree frame of IMAP
   client mode of IlohaMail, and wait.
2. Just after reload is complete, execute "gfs_tool freeze" on
   any node of GFS cluster (I tried also on all nodes).
3. Create a block/volume level snapshot on file server.  I tried
   GNBD/LVM/LinuxBox and iSCSI(sfnet-initiator-
4. Execute gfs_tool unfreeze.
5. Import the snapshot LV/LUN on any Linux machine, and execute

By this procedure, the lost file is always an IlohaMail's cache
file for IMAP folder list and each property.
The inconsistency is sometimes not caused.

I also tried to sync before/after freezing GFS, but nothing seemed
to make difference.

Freezing GFS suspends filesystem's activity, but doesn't it
ensure filesystem level consisntency?

Right now I'm using GFS on FC4-test3 with the newest kernel RPMs
and the newest Cluster/GFS RPMs in:


-- Kenji

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]