Project

General

Profile

Bug #47679

kceph: kernel does not open session with MDS importing subtree

Added by Patrick Donnelly 4 months ago. Updated 6 days ago.

Status:
New
Priority:
High
Assignee:
-
Category:
-
Target version:
% Done:

0%

Source:
Q/A
Tags:
Backport:
pacific,octopus,nautilus
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
MDS, kceph
Labels (FS):
multimds, qa-failure
Pull request ID:
Crash signature:

Description

2020-09-27T19:29:40.984+0000 7f1863bcb700 10 mds.1.server finish_force_open_sessions on 1 clients, initial v 7
2020-09-27T19:29:40.984+0000 7f1863bcb700 10 mds.1.server force_open_sessions opened client.4529 v1:172.21.15.23:0/1640007701
2020-09-27T19:29:40.984+0000 7f1863bcb700 10 mds.1.sessionmap touch_session s=0x55d71d224400 name=client.4529
2020-09-27T19:29:40.984+0000 7f1863bcb700 10 add_session: mds.metrics: session=0x55d71d224400, client=client.4529 v1:172.21.15.23:0/1640007701
2020-09-27T19:29:40.984+0000 7f1863bcb700 20 add_session: mds.metrics: metrics=[update_type=0, metrics={cap_hit_metric={hits=0, misses=0}, read_latency={latency=0.000000}, write_latency={latency=0.000000}, metadata_latency={latency=0.000000}}]
2020-09-27T19:29:40.984+0000 7f1863bcb700 10 mds.1.9 send_message_client client.4529 v1:172.21.15.23:0/1640007701 client_session(open) v4
2020-09-27T19:29:40.984+0000 7f1863bcb700 20 mds.1.sessionmap mark_dirty s=0x55d71d224400 name=client.4529 v=7
2020-09-27T19:29:40.984+0000 7f1863bcb700 10 mds.1.server finish_force_open_sessions: final v 8
...
2020-09-27T19:29:50.972+0000 7f18683d4700 20 mds.1.server laggiest active session is client.4529 v1:172.21.15.23:0/1640007701 and renewed caps recently (9.988s ago)
...
2020-09-27T19:30:45.972+0000 7f18683d4700 10 mds.1.server new stale session client.4529 v1:172.21.15.23:0/1640007701 last renewed caps 64.988s ago
2020-09-27T19:30:45.972+0000 7f18683d4700 20 mds.1.server laggiest active session is client.4609 v1:172.21.15.23:0/775260981 and renewed caps recently (18.092s ago)
2020-09-27T19:30:45.972+0000 7f18683d4700 10 mds.1.locker revoke_stale_caps for client.4529
2020-09-27T19:30:45.972+0000 7f18683d4700 20 mds.1.server oldest stale session is client.4529 v1:172.21.15.23:0/1640007701 and recently renewed caps 64.988s ago
2020-09-27T19:30:45.972+0000 7f18683d4700  0 log_channel(cluster) log [WRN] : evicting unresponsive client  (4529), after 64.988 seconds

From: /ceph/teuthology-archive/pdonnell-2020-09-26_05:51:09-multimds-wip-pdonnell-testing-20200926.000836-distro-basic-smithi/5471731/remote/smithi086/log/ceph-mds.a.log.gz

So far this is only with the testing kernel. see also this job, same error:

/ceph/teuthology-archive/pdonnell-2020-09-26_05:51:09-multimds-wip-pdonnell-testing-20200926.000836-distro-basic-smithi/5472185/remote/smithi040/log/ceph-mds.c.log.gz

History

#1 Updated by Patrick Donnelly 3 months ago

  • Priority changed from Normal to High
  • Labels (FS) multimds, qa-failure added

#2 Updated by Patrick Donnelly 6 days ago

  • Target version changed from v16.0.0 to v17.0.0
  • Backport set to pacific,octopus,nautilus

Also available in: Atom PDF