Ceph » Linux kernel client

06/19/2021

04:33 PM Bug #51279: kclient hangs on umount (testing branch)

Maybe, I think this job is dead so nothing was collected. Could rerun and wait for the hang then ssh into the smithi ... Patrick Donnelly

10:39 AM Bug #51279: kclient hangs on umount (testing branch)

Definitely hung. The console log has this:... Jeff Layton

12:12 AM Bug #51279 (Resolved): kclient hangs on umount (testing branch)

... Patrick Donnelly

06/18/2021

06:14 PM Bug #51158: KASAN use after free when destroying inode at umount

I have a patch series to fix this in the testing branch now. It removes some locking that had an unclear purpose, so ... Jeff Layton

06/16/2021

08:39 AM Bug #51112 (Pending Backport): RHEL 8.4: libceph considers osdmap corrupt if numbering is discont...

Moving to Pending Backport, will update once the fix lands in CentOS/RHEL 8.4. Ilya Dryomov

06/15/2021

06:38 PM Bug #51112: RHEL 8.4: libceph considers osdmap corrupt if numbering is discontinuous

Same issue here.
We worked around by filling all the id gaps with "ceph osd new `uuid`", then purging those newly ... Dan van der Ster

03:24 PM Bug #51112: RHEL 8.4: libceph considers osdmap corrupt if numbering is discontinuous

https://bugzilla.redhat.com/show_bug.cgi?id=1972278 Ilya Dryomov

02:24 PM Bug #51112: RHEL 8.4: libceph considers osdmap corrupt if numbering is discontinuous

No need for additional output, I managed to reproduce with the attached osdmap:... Ilya Dryomov

12:13 PM Bug #51112 (Need More Info): RHEL 8.4: libceph considers osdmap corrupt if numbering is discontin...

Ilya Dryomov

12:03 PM Bug #51112: RHEL 8.4: libceph considers osdmap corrupt if numbering is discontinuous

The linked output is incomplete -- it starts with... Ilya Dryomov

11:49 AM Bug #45456 (Need More Info): The read and write operations of libceph are blocked waiting for IO...

It's not just the monitor, looks more like a general connectivity issue. Note that the monitor session was successfu... Ilya Dryomov

11:34 AM Bug #50234 (Closed): krbd failed to mount after map image with read-only option

Ilya Dryomov

11:33 AM Bug #50234: krbd failed to mount after map image with read-only option

... Ilya Dryomov

11:31 AM Bug #42757 (In Progress): deadlock on lock_rwsem: rbd_quiesce_lock() vs watch errors

Ilya Dryomov

11:28 AM Bug #51136: Random hanging issues with rbd after network issues

I'm pretty sure this is https://tracker.ceph.com/issues/42757.
I'm working on a fix. Ilya Dryomov

06/09/2021

01:47 PM Bug #51158: KASAN use after free when destroying inode at umount

There are some potential workarounds, but the issue is itself caused by the workaround that added ceph_async_iput:
... Jeff Layton

01:43 PM Bug #51158 (Resolved): KASAN use after free when destroying inode at umount

KASAN occasionally pops this warning when when unmounting:... Jeff Layton

08:25 AM Bug #48895: ceph: krbd client con get stuck in CON_SOCK_STATE_CONNECTING state

Ilya Dryomov wrote:
> Hi Fan,
>
> Sorry, not yet. I hit a locking issue which I didn't have time to investigate ... fan chen

08:07 AM Bug #48895: ceph: krbd client con get stuck in CON_SOCK_STATE_CONNECTING state

Hi Fan,
Sorry, not yet. I hit a locking issue which I didn't have time to investigate and rather than risking the... Ilya Dryomov

01:53 AM Bug #48895: ceph: krbd client con get stuck in CON_SOCK_STATE_CONNECTING state

hi Ilya, the v5.12.9 kernel has been released already. were those patches pushed?
fan chen wrote:
> fan chen wrot... fan chen

01:52 AM Bug #48895: ceph: krbd client con get stuck in CON_SOCK_STATE_CONNECTING state

hi Ilya,
the v5.12.9 kernel has been released already. were those patches pushed? fan chen

06/08/2021

03:18 PM Bug #51136 (Duplicate): Random hanging issues with rbd after network issues

Hello,
We are experimenting random hanging issues with our LXC VM using a mapped RBD image.
We're using version... Maximilien Cuony

06/07/2021

05:50 AM Bug #51112 (Pending Backport): RHEL 8.4: libceph considers osdmap corrupt if numbering is discont...

If I try to map rbd from a cluster where osd numbering is discontinuous it fails with message @[85115.711643] libce... Matti Saarinen

06/01/2021

03:29 PM Bug #46419: ceph: ceph_add_cap: couldn't find snap realm 110

The snaprealm locking is not as broken as I had originally feared. Some of the comments were definitely wrong, but th... Jeff Layton

12:57 PM Bug #46419: ceph: ceph_add_cap: couldn't find snap realm 110

I'm still trying to wrap my brain around the locking in this code, but it's pretty clear that it's badly broken (or a... Jeff Layton