From: Cool F. <fin...@ya...> - 2005-09-29 13:39:56
|
--- Franco Broi <fr...@fu...> wrote: > On Wed, 2005-09-28 at 10:51 -0700, Cool Frood wrote: > > Just my 2 cents on this. I have also been testing > NFS > > with FUSE, and for testing theoretical performance > > limits, I have been discarding write operations in > > userspace and/or kernel space. I'm using my own > > user-level code and getting lockups, so this > problem > > probably exists somewhere in the kernel code. The > > lockup only happens with two client machines, not > one. > > > > > I've been hammering this to death over the last few > weeks but I can't > get a handle where the problem is. I've tested > different combinations of > FUSE and kernel versions, none of which seem to make > any difference. > > The lockup is totally reproducible and only takes a > matter of minutes to > fail, so if anyone has any bright ideas on how to > diagnose the problem, > please let me know and I'll try it. > > I have one suggestion. You could try reproducing the problem with just one NFSD thread. Maybe it is related to some kind of concurrency issues. -Akshat > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Power Architecture Resource Center: Free content, > downloads, discussions, > and more. > http://solutions.newsforge.com/ibmarch.tmpl > _______________________________________________ > fuse-devel mailing list > fus...@li... > https://lists.sourceforge.net/lists/listinfo/fuse-devel > __________________________________ Yahoo! Mail - PC Magazine Editors' Choice 2005 http://mail.yahoo.com |