Re: NFSD threads hang when destroying a session or client ID

[Chuck Lever via Bugspray Bot <bugbot@kernel.org>]

Chuck Lever writes via Kernel.org Bugzilla:

During server operation, you can capture trace events that might lead up to the hang with this command:

# trace-cmd record -e nfsd:nfsd_cb_\* -e sunrpc:svc_xprt_detach -p function -l nfsd4_destroy_session

These events are infrequent and tracing is lightweight, so this capture won't be a burden.

When a server hang is observed, typically, "INFO: task has been blocked for 120 seconds" appears in the server's system journal. Basic collection instructions are:

# rpcdebug -m rpc -c && echo t > /proc/sysrq-trigger

Then stop the trace and package trace.dat and the system journal and attach them to this bug.

View: https://bugzilla.kernel.org/show_bug.cgi?id=219710#c2
You can reply to this message to join the discussion.
-- 
Deet-doot-dot, I am a bot.
Kernel.org Bugzilla (bugspray 0.1-dev)