General

Profile

Jeff Layton

Issues

Projects

  • Ceph (Developer, Reporter, 06/23/2016)
  • Linux kernel client (Developer, Reporter, 06/23/2016)
  • phprados (Developer, Reporter, 06/23/2016)
  • devops (Developer, Reporter, 06/23/2016)
  • rbd (Developer, Reporter, 06/23/2016)
  • rgw (Developer, Reporter, 06/23/2016)
  • sepia (Developer, Reporter, Backporter, 06/23/2016)
  • fs (Developer, Reporter, 06/23/2016)
  • teuthology (Developer, Reporter, 06/23/2016)
  • rados-java (Developer, Reporter, 06/23/2016)
  • Calamari (Developer, Reporter, 06/23/2016)
  • Ceph-deploy (Developer, Reporter, 06/23/2016)
  • ceph-qa-suite (Developer, Reporter, 06/23/2016)
  • Tools (Developer, Reporter, Backporter, 06/23/2016)
  • Infrastructure (Developer, Reporter, Backporter, 06/23/2016)
  • ceph-cm-ansible (Developer, Reporter, Backporter, 06/23/2016)
  • paddles (Developer, Reporter, Backporter, 06/23/2016)
  • pulpito (Developer, Reporter, Backporter, 06/23/2016)
  • downburst (Developer, Reporter, Backporter, 06/23/2016)
  • ovh (Developer, Reporter, Backporter, 06/23/2016)
  • www.ceph.com (Developer, Reporter, Backporter, 08/22/2016)
  • CI (Reporter, 01/10/2017)
  • rgw-testing (Developer, Reporter, 11/01/2016)

Activity

04/17/2017

02:08 PM fs Bug #19426: knfs blogbench hang
Sorry I didn't see this sooner. Is this still cropping up?
So what might be helpful the next time this happens loc...
11:37 AM ceph-qa-suite Bug #19550: re-enable ENOSPC tests in ceph-qa-suite for kclient
John Spray wrote:
> Interesting thread.
>
> I think the "whether to sync on close" aspect is a bit of a red herri...
10:58 AM ceph-qa-suite Bug #19550: re-enable ENOSPC tests in ceph-qa-suite for kclient
John Spray wrote:
> As you say, exposing writeback errors in close() is not required. I like doing it from a pragma...

04/14/2017

02:17 PM ceph-qa-suite Bug #19550: re-enable ENOSPC tests in ceph-qa-suite for kclient
Ok, with this suite repo/branch and the code currently in testing, the failing test now passes:
https://github.com...
12:25 PM Linux kernel client Feature #17204: Implement new-style ENOSPC handling in kclient
I've changed the code in testing to set the barrier first, and added a patch to return writeback errors on close, whi...

04/13/2017

07:22 PM ceph-qa-suite Bug #19550: re-enable ENOSPC tests in ceph-qa-suite for kclient
Ok, I added a flush operation to kcephfs which does a vfs_fsync when the file is open for write, and returns the erro...
05:10 PM ceph-qa-suite Bug #19550: re-enable ENOSPC tests in ceph-qa-suite for kclient
John Spray wrote:
> What happens to dirty buffers when a file is closed in the kernel client? They just stick around...
04:33 PM ceph-qa-suite Bug #19550: re-enable ENOSPC tests in ceph-qa-suite for kclient
Ok, with the latest kernel testing branch, and John's ceph wip-19550 branch for the testsuite, it gets a bit farther....
02:28 PM Linux kernel client Feature #17204: Implement new-style ENOSPC handling in kclient
Ok, I see the problem. I added a down_read for the osdc->epoch_barrier fetch to send_cap_msg. That's dumb -- I'll jus...
02:08 PM Linux kernel client Feature #17204: Implement new-style ENOSPC handling in kclient
Ok, hit a deadlock in testing today. Yay for lockdep. I'll have to look at how to fix this:...

Also available in: Atom