Version: 6.8.3

 

Could not access my server this morning, requiring a hard reset. I was able to capture some details on the attached syslog.txt. Its appears to be multiple rcu_sched detected stalls on CPUs/tasks. I've also attached the diagnostics after the reboot. Any help is appreciated.syslog.txtdeathstar-diagnostics-20200813-1032.zip

same here today ... did anything ever become of this?

 

my server needed a hard reset after these errors flooded the syslog file.

 

diagnostics only available from syslog server backend attached

towerdiagnostics130521.txt

Edited May 13, 20215 yr by superloopy1

And again today ... anyone able to take a look at logs?

17 hours ago, superloopy1 said:

And again today

Since the traces are mentioning xfs run a filesystem check (without -n) on all your xfs filesystems and see if it helps.

Thanks JB, will do and let you know ...

HI again ... xfs checked out just fine.

Not always clear to see if anything was fixed with xfs_repair, unless you check the exit code, see if it makes any difference.