| To: | Supratik Goswami <supratiksekhar@xxxxxxxxx> |
|---|---|
| Subject: | Re: File system remain unresponsive until the system is rebooted. |
| From: | Dave Chinner <david@xxxxxxxxxxxxx> |
| Date: | Wed, 1 Feb 2012 06:44:04 +1100 |
| Cc: | xfs@xxxxxxxxxxx |
| In-reply-to: | <CANs4eSBgmvJCR7vfFa1W5h8tUYFQi=LRPWDPQ1exB29D1o_RjA@xxxxxxxxxxxxxx> |
| References: | <CANs4eSBWLc4HxAbPZ8kOVOdJ7RKiA+-ai3Q2J+FAyuzHtUqfdg@xxxxxxxxxxxxxx> <20120131013124.GE9090@dastard> <CANs4eSBgmvJCR7vfFa1W5h8tUYFQi=LRPWDPQ1exB29D1o_RjA@xxxxxxxxxxxxxx> |
| User-agent: | Mutt/1.5.21 (2010-09-15) |
On Tue, Jan 31, 2012 at 10:34:10AM +0530, Supratik Goswami wrote: > We are using Amazon EC2 instances. > > ubuntu@ip-10-0-0-10:~$ uname -aLinux ip-10-0-0-10 2.6.32-318-ec2 > #38-Ubuntu SMP Thu Sep 1 18:09:30 UTC 2011 x86_64 GNU/Linux The growfs hang problem was fixed in 2.6.34. On earlier kernels, if you do a grow while the system is under allocation load it could deadlock. growing on a mostly idle filesystem was generally OK, but under heavy load problems could occur. This hang is what xfstests 104 exercises... Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH] Introduce SEEK_DATA/SEEK_HOLE support to XFS V6, Mark Tinguely |
|---|---|
| Next by Date: | Re: Performance problem - reads slower than writes, Dave Chinner |
| Previous by Thread: | Re: File system remain unresponsive until the system is rebooted., Stan Hoeppner |
| Next by Thread: | [PATCH] xfs_quota: remove calls to XFS_QSYNC, Christoph Hellwig |
| Indexes: | [Date] [Thread] [Top] [All Lists] |