At 04:12 PM 7/8/2002, Tom Lane wrote:
>This is beginning to look like a kernel problem.
Sigh. The hardest ones to fix. It's very nice however that the fsync=off
fixes the immediate problem, inasmuch as I can run with that for the moment
until such a time as we figure out the "real" problem.
>That should be FSYNC not FSYNCA, I believe.
OK I'll try that.
> > Of note, in the fdatasync() man page: Currently (Linux 2.2)
> > fdatasync is equivalent to fsync.
>
>Are you really using a 2.2 kernel? I thought that that point had been
>fixed in 2.4 kernels. (I see the comment is still there in the man page
>on my RH7.2 box, though.)
No, I'm using 2.4.18 on Debian/Woody, but as you say, the man page still
refers to 2.2.
>Now that we've eliminated MoveOfflineLogs as the time-consuming part of
>checkpoint, would you make the same check for mdsync()? That's the
>routine that actually issues the sync() calls. Again, it would be
>useful to know how long it takes, and whether the other backends appear
>to be blocked at the times mdsync() is entered and exited.
Will do.
Thanks,
Doug