tech-kern archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: 10/amd64 lockup with lots of rpcbinds in fstchg
The system is waiting for a file system suspension to complete.
Brian Buhrow has a similiar problem.
Running "call fstrans_dump(1)" from DDB will give more information.
Do you have ptyfs mounted?
Do you have a kernel crash dump or backtraces of all threads at the time of hang?
Any thread running some kind of revoke()?
--
J. Hannken-Illjes - hannken%mailbox.org@localhost
> On 9. Oct 2025, at 16:46, Edgar Fuß <ef%math.uni-bonn.de@localhost> wrote:
>
> For the third time, one of our identical login servers (the same one each
> time) became unresponsive. Network is unresponsive, switching consoles with
> Ctl-Alt-Fn works, but they are unresponsive, entering DDB works. Last kernel
> message on screen is "in_chksum: out of data".
> DDB reveals hundreds (thousands?) of rpcbind processes in fstchg state
> (whatever that is), as are several ssh processes, some cron and one ipmon
> one and ioflush. syslogd, ntpd and one icinga2 process is in tstile state.
> swapper is in uvm. The four active proceses seem to be four idle.
>
> We never experienced that under 8; the other identical login sever has also
> been unaffected so far.
>
> Does that ring a bell to someone?
Home |
Main Index |
Thread Index |
Old Index