|To:||Ben Myers <bpm@xxxxxxx>, Alex Elder <elder@xxxxxxxxxx>, xfs@xxxxxxxxxxx|
|Subject:||global serialization points in XFS circa 2.6.27?|
|From:||Chris Friesen <chris.friesen@xxxxxxxxxxx>|
|Date:||Thu, 13 Sep 2012 17:23:33 -0600|
|User-agent:||Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:184.108.40.206) Gecko/20111108 Fedora/3.1.16-1.fc14 Lightning/1.0b3pre Thunderbird/3.1.16|
Hi,I've got an embedded system running a modified 2.6.27 (yes, we're trying to upgrade) and we're running into a scenario where something is happening that causes accesses to *all* XFS filesystems to hang or be delayed for some time.
I know it may be a bit of a stretch since the kernel is old, but I was wondering if you might have any suggestions as to what I might look at as far as locks/algorithms within XFS that would affect all filesystems.
In case it matters, the hardware is fairly beefy-- 32 threads of Sandy Bridge CPU, 64GB RAM, SAS drives, etc.
Chris -- Chris Friesen Software Designer 3500 Carling Avenue Ottawa, Ontario K2H 8E9 www.genband.com
|<Prev in Thread]||Current Thread||[Next in Thread>|
|Previous by Date:||Re: xfs: stop the sync worker before xfs_unmountfs, Ben Myers|
|Next by Date:||Re: [PATCH] xfs: stop the sync worker before xfs_unmountfs, Dave Chinner|
|Previous by Thread:||Re: [PATCH] libxcmd: link against readline, Mark Tinguely|
|Next by Thread:||[PATCH] xfstests: 275, Don't remove all the files in SCRATCH_MNT., Tao Ma|
|Indexes:||[Date] [Thread] [Top] [All Lists]|