| To: | Laurence Oberman <loberman@xxxxxxxxxx> |
|---|---|
| Subject: | Re: Tasks blocked in xlog_grant_log_space |
| From: | Dave Chinner <david@xxxxxxxxxxxxx> |
| Date: | Mon, 17 Feb 2014 09:28:11 +1100 |
| Cc: | xfs@xxxxxxxxxxx, Leonardo Menezes Vaz <lvaz@xxxxxxxxxx> |
| Delivered-to: | xfs@xxxxxxxxxxx |
| In-reply-to: | <461400101.2501294.1392394773671.JavaMail.zimbra@xxxxxxxxxx> |
| References: | <532438291.2495843.1392394364516.JavaMail.zimbra@xxxxxxxxxx> <461400101.2501294.1392394773671.JavaMail.zimbra@xxxxxxxxxx> |
| User-agent: | Mutt/1.5.21 (2010-09-15) |
On Fri, Feb 14, 2014 at 11:19:33AM -0500, Laurence Oberman wrote: > Hello > > I have a customer seeing multiple tasks blocked in xlog_grant_log_space > This lands up hitting the hungtask timer for pdflush. > > I have seen on the list that this issue is still out there. > I also don't see outstanding I/O requests on the XFS file system that > experienced this block. > > The last hung_task timeout was on dm-9 > ffff81011c78c080 ffff81081a43bc00 xfs /dev/vgautocomp/lv_autocomp > /bb/autocomp > > crash> gendisk ffff810814d24400 | grep in_flight > in_flight = 0, > > So just wondering if I am hitting an existing bug or if its a new one. > > This is an older 2.6.18 kernel There's been lots of problems fixed in this area in the past seven years. Your best bet is to upgrade to a more recent 3.x based kernel where the known issues in this area have been resolved. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Volume fine on x86_64, corruption on ARM, Dave Chinner |
|---|---|
| Next by Date: | Re: [PATCH v2] xfstests: add regression test for btrfs incremental send, Dave Chinner |
| Previous by Thread: | Tasks blocked in xlog_grant_log_space, Laurence Oberman |
| Next by Thread: | [PATCH] xfstests: add regression test for btrfs incremental send, Filipe David Borba Manana |
| Indexes: | [Date] [Thread] [Top] [All Lists] |