Hi Stefan, Yeah, the same happened to us on around half of our clusters. I didn't ever find the pattern what triggered it or not on the different clusters: https://tracker.ceph.com/issues/43596 Cheers, Dan On Fri, Oct 30, 2020 at 1:23 AM Stefan Kooman <stefan@bit.nl> wrote:

Hi List,

After a successful upgrade from Mimic 13.2.8 to Nautilus 14.2.12 we enabled msgr2. Soon after that both of the MDS servers (active / active-standby) restarted.

We did not hit any ASSERTS this time, so that's good :>.

However, I have not seen this happening on four different test clusters (while running a slightly older Nautilus release), so I certainly did not expect that.

Most of the connections switched over to 3300 (apart from the cephfs kernel clients) and that all kept on working.

Anybody else has seen this behavior before?

Gr. Stefan _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io