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:

(In reply to Baptiste PELLEGRIN from comment #22)
> I patched the current Debian stable kernels :
> 
> Debian Linux 6.1.0-29-amd64 (= 6.1.123-1) Based on Linux 6.1.120
> Debian Linux 6.1.0-30-amd64 (= 6.1.124-1) Based on Linux 6.1.124
> 
> with there two commits :
> 
> 8626664c87ee NFSD: Replace dprintks in nfsd4_cb_sequence_done()
> 961b4b5e86bf NFSD: Reset cb_seq_status after NFS4ERR_DELAY
> 
> Now my two NFS servers are running these patched kernel. I need to wait one
> week to see If the crash still occur.

It's probable that a hang might still occur, but the symptoms might be closer to the hangs that happen on v6.12. If so, I count that as step forward: one less bug.


> Chuck, does this patch may remove the "unrecognized reply" messages from my
> syslog ?

It doesn't.


> I see you have improved the backchannel session logging. Did I need to
> change my trace-record command ?

The trace-cmd command line I provided above uses a glob expression that should also enable the trace points added in 8626664c87ee ("NFSD: Replace dprintks in nfsd4_cb_sequence_done()").

View: https://bugzilla.kernel.org/show_bug.cgi?id=219710#c23
You can reply to this message to join the discussion.
-- 
Deet-doot-dot, I am a bot.
Kernel.org Bugzilla (bugspray 0.1-dev)