Josh this is ready for your final review and publishing . David FYI and https://github.com/ceph/ceph/pull/41958 On Tue, Jun 29, 2021 at 12:50 AM Ilya Dryomov <idryomov@gmail.com> wrote:
On Fri, Jun 25, 2021 at 9:19 PM Yuri Weinstein <yweinste@redhat.com> wrote:
While we are waiting for finial rbd approval pls add your commits and apporavls to the Release Notes PR
It appears that this bug exists in older nautilus releases (as far back as 14.2.12 at least) and is specific to journaling which is a rather narrow use case.
rbd approved.
Thanks,
Ilya
https://github.com/ceph/ceph/pull/41958
Thx
On Fri, Jun 25, 2021 at 8:46 AM Mykola Golub <to.my.trociny@gmail.com> wrote:
On Fri, Jun 25, 2021 at 11:19:25AM +0200, Ilya Dryomov wrote:
On Fri, Jun 25, 2021 at 7:00 AM Mykola Golub <to.my.trociny@gmail.com> wrote:
On Wed, Jun 23, 2021 at 05:56:00PM +0200, Ilya Dryomov wrote:
For rbd, there is a suspicious failure that persisted through the rerun:
http://qa-proxy.ceph.com/teuthology/yuriw-2021-06-21_15:45:39-rbd-nautilus-d... http://qa-proxy.ceph.com/teuthology/yuriw-2021-06-22_15:18:36-rbd-nautilus-d...
Just FYI, it is also reproducible on v14.2.21 (sha1: 5ef401921d7a88aea18ec7558f7f9374ebd8f5a6):
http://qa-proxy.ceph.com/teuthology/trociny-2021-06-24_16:42:49-rbd-wip-mgol...
So it does not look like a regression. I was able to reproduce it on centos only while on ubuntu it always passed for me.
Yuri's successful run which didn't include the last couple of rbd PRs was also on ubuntu. But I do see a green centos job among your runs so it doesn't appear to be OS specific:
http://qa-proxy.ceph.com/teuthology/trociny-2021-06-24_15:01:11-rbd-wip-mgol...
Yes it could succeed on centos. What I meant is that I had naver seen it to fail on ubuntu, though I also don't think it is OS specific. It also did not fail when I increased the rbd debug level to track the hang.
Filed https://tracker.ceph.com/issues/51362 with the (so far scarce) details.
Thanks,
Ilya
-- Mykola Golub