Will do. Becky -- Becky Ligon PVFS Developer Clemson University 864-650-4065
> I am using a single RHEL4 box with a single server and one RHEL5 client > accessing it. That was the simplest way I could reproduce, so I have stuck > with it. There is nothing spectacular about either machine. If you track > down the issues with running multiple servers on the same box, I would > love > to get the details. > > Bart. > > > > On Wed, Oct 20, 2010 at 9:39 AM, Becky Ligon <[email protected]> wrote: > >> Bart: >> >> What is the topology of your test system? >> >> At Clemson, when running tests on a SUN box with 8 Quad-Core AMD Opteron >> Processor 2356 and the latest cvs version of pvfs, I was getting a lot >> of >> timing issues when running 4 servers on the one box, justing using the >> cp >> command. I am still trying to run down the problems. >> >> Becky >> -- >> Becky Ligon >> PVFS Developer >> Clemson University >> 864-650-4065 >> >> > I can reproduce it fairly reliably with my test case; I average >> somewhere >> > around a 50% chance of generating one or more of these files per test. >> My >> > preliminary testing with db 4.8.30 is encouraging. I will update this >> > thread >> > with any conclusions. >> > >> > Bart. >> > >> > >> > >> > On Tue, Oct 19, 2010 at 3:41 PM, Phil Carns <[email protected]> wrote: >> > >> >> >> >> Removing the metadata object for a file does indeed produce the same >> >>> symptoms we are seeing. It produces a similar effect on 2.8 as well. >> I >> >>> believe I was working with Sam and possibly you on this a few weeks >> ago >> >>> but >> >>> had to drop it for something more urgent. Our conversation can be >> found >> >>> here: >> >>> >> >>> >> >>> >> http://www.beowulf-underground.org/pipermail/pvfs2-developers/2010-June/004605.html >> >>> >> >>> In 2.8, pvfs2-rm does not currently remove the file entries. Sam had >> >>> the >> >>> same ENOTENT fix for 2.8. After creating a file and removing the >> >>> metadata >> >>> object, it allows the file entry to be removed. >> >>> >> >> >> >> Ah! Believe it or not I had forgotten about that thread, but Sam's >> fix >> >> must have stuck in my head subconsciously because I figured out where >> to >> >> add >> >> that check in 2.6 pretty quickly :) >> >> >> >> We need to get this fix into cvs if it isn't there already... >> >> >> >> >> >> I believe Jim Kusznir may have been experiencing similar issues when >> he >> >>> posted to the Users list here: >> >>> >> >>> >> >>> >> http://www.beowulf-underground.org/pipermail/pvfs2-users/2010-September/003186.html >> >>> >> >>> It appears that something is still triggering this issue in 2.8. >> Based >> >>> on >> >>> my own anecdotal evidence with 2.6 and 2.8, it looks like it can >> happen >> >>> when >> >>> a file system is getting hammered with creates and/or deletes. My >> test >> >>> case >> >>> to reproduce on 2.8 involved several threads executing a script that >> >>> creates, opens, writes, and then deletes a file many thousands of >> >>> times. The >> >>> problems we have had on 2.6 also correlate to heavy loads of file >> >>> creations >> >>> and deletions. >> >>> >> >>> Anyone have thoughts on why some files are getting left without >> their >> >>> metadata object? >> >>> >> >>> Bart. >> >>> >> >> >> >> I think Becky had some good ideas. How hard is this to reproduce? >> Can >> >> you >> >> do it pretty reliably with your test case? >> >> >> >> -Phil >> >> >> > _______________________________________________ >> > Pvfs2-developers mailing list >> > [email protected] >> > http://www.beowulf-underground.org/mailman/listinfo/pvfs2-developers >> > >> >> > _______________________________________________ Pvfs2-developers mailing list [email protected] http://www.beowulf-underground.org/mailman/listinfo/pvfs2-developers
