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:

v6.1 symptomology:

The following stack trace shows that the active callback workqueue item is stuck in rpc_shutdown_client() waiting for RPC operations to complete. Because the callback workqueue is single-threaded (max_active = 1) and this work item never completes, it prevents subsequent flush_workqueue() calls on the callback work queue from completing.

> [7257352.347503] task:kworker/u32:6   state:D stack:0     pid:173983 ppid:2  
>    flags:0x00004000
> [7257352.347511] Workqueue: nfsd4_callbacks nfsd4_run_cb_work [nfsd]
> [7257352.347568] Call Trace:
> [7257352.347571]  <TASK>
> [7257352.347577]  __schedule+0x34d/0x9e0
> [7257352.347589]  schedule+0x5a/0xd0
> [7257352.347597]  schedule_timeout+0x94/0x150
> [7257352.347606]  ? __bpf_trace_tick_stop+0x10/0x10
> [7257352.347616]  rpc_shutdown_client+0xf2/0x150 [sunrpc]
> [7257352.347683]  ? cpuusage_read+0x10/0x10
> [7257352.347694]  nfsd4_process_cb_update+0x4c/0x270 [nfsd]
> [7257352.347763]  nfsd4_run_cb_work+0x9f/0x150 [nfsd]
> [7257352.347812]  process_one_work+0x1c7/0x380
> [7257352.347824]  worker_thread+0x4d/0x380
> [7257352.347835]  ? rescuer_thread+0x3a0/0x3a0
> [7257352.347843]  kthread+0xda/0x100
> [7257352.347849]  ? kthread_complete_and_exit+0x20/0x20
> [7257352.347859]  ret_from_fork+0x22/0x30
> [7257352.347875]  </TASK>

Based on the full stack trace output, there do not appear to be any outstanding RPC operations.

View: https://bugzilla.kernel.org/show_bug.cgi?id=219710#c3
You can reply to this message to join the discussion.
-- 
Deet-doot-dot, I am a bot.
Kernel.org Bugzilla (bugspray 0.1-dev)