|To:||"IKARASHI, Seiichi" <ikarashi@xxxxxxxxxxxxxxxx>|
|Subject:||Re: synchronization of XFS|
|From:||Chris Wedgwood <cw@xxxxxxxx>|
|Date:||Thu, 25 Mar 2004 04:47:51 -0800|
|References:||<4060F7FC.8090602@xxxxxxxxxxxxxxxx> <20040325063902.GA9697@xxxxxxxxxxxxxxxxxxxxxxx> <4062C97A.6030702@xxxxxxxxxxxxxxxx> <20040325124152.GA12078@xxxxxxxxxxxxxxxxxxxxxxx>|
On Thu, Mar 25, 2004 at 04:41:52AM -0800, Chris Wedgwood wrote: > That sounds like something else. sync() ensures dirty buffers are > on-disk. Even if you don't sync() you can still read-back using the > buffer-cache/page-cache. OK, I understand now. Calling sync() multuple times is still baindead. How about umount/mount and see if that forces the data to appear? If so, please try a newer kernel as the sync paths were rewritten and might fix your problem.
|<Prev in Thread]||Current Thread||[Next in Thread>|