behlendorf1@llnl.gov
2006-Dec-21 12:59 UTC
[Lustre-devel] [Bug 11471] Many 160-stripe files cause memory fragmentation and failures
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=11471 I don''t have access to bug 10091, could that bug be made visable or the partial proposed patch added to this bug.
pbojanic@clusterfs.com
2007-Jan-09 11:52 UTC
[Lustre-devel] [Bug 11471] Many 160-stripe files cause memory fragmentation and failures
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=11471 Andreas, can you please advise what testing should be undertaken to thoroughly qualify the patch from bug 10091. We should involve Cliff White and/or Peter Belau to undertake recommended testing on ALC test.
Fergal.McCarthy@HP.com
2007-Jan-10 04:39 UTC
[Lustre-devel] [Bug 11471] Many 160-stripe files cause memory fragmentation and failures
Please don''t reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=11471 (In reply to comment #5)> Andreas, can you please advise what testing should be undertaken to thoroughly > qualify the patch from bug 10091. We should involve Cliff White and/or Peter > Belau to undertake recommended testing on ALC test.Peter, Just for clarification, are you referring to the combined patch (attachment 7693) that I developed for Lustre 1.4.6.2 from the original patches for Lustre 1.4.1.10? If so I can say that it stands up under basic operation (simple IOR workload) but has not been rigorously tested. If you are referring to some new patch I''d appreciate a peek at it as I''m hoping to get some time again soon for verifying the fix under our Lustre 1.4.6.4 code base. Fergal.