Unscheduled/Random parity checks

[reggienaz]

Hello. Over the past few days/weeks I've noticed despite the scheduler being set to weekly for parity check my server is performing parity checks on a daily basis. I tried changing the settings but its still doing the same thing. I'm not writing a ton of data to the array and im not sure where to start in order to diagnose this. I just changed the schedule to monthly because I'm pretty sure that weekly is overkill and probably shortening drive life for no reason given the amount of data written to the array on a weekly or even monthly basis. I've attached diagnostics and the syslog files. Any and all help is greatly appreciated.

infinity-diagnostics-20210430-1239.zip syslog

[codefaux]

Not seeing any immediate concerns in the diagnostic, but its runtime only covers Apr 29 to Apr 30. Is this a time window where an unscheduled check seemed to occur? If not, it may be best to submit a log from that period. (I'm a touch new at interpreting those diagnostics, so I may have missed a cue in the log that it was running. If so, I apologize.)

 

One thing that occurs to me -- have you tracked your uptime when this happens? If you have your array set to Auto-Start and it's rebooting unexpectedly, it will reset the uptime (visible in the top-right corner of the webUI) and of course do a Parity sync due to the unclean shutdown.

 

Ideally this at least adds clarity. Let us know.

[trurl]

10 hours ago, reggienaz said:

I've attached diagnostics and the syslog files

For future reference, syslog since last reboot is already included in diagnostics. The only time it might be useful to have a separate syslog is if you have Syslog Server setup to save syslogs so you can get us syslog after a crash.

 

Apr 30 11:00:01 Infinity kernel: mdcmd (36): check 
Apr 30 11:00:01 Infinity kernel: md: recovery thread: check P ...
Apr 30 11:00:14 Infinity emhttpd: read SMART /dev/sdb
Apr 30 11:00:14 Infinity emhttpd: read SMART /dev/sdc
Apr 30 11:00:20 Infinity flash_backup: adding task: php /usr/local/emhttp/plugins/dynamix.unraid.net/include/UpdateFlashBackup.php update
Apr 30 11:00:35 Infinity webGUI: Successful login user root from 192.168.86.49
Apr 30 11:07:50 Infinity kernel: mdcmd (37): nocheck Cancel
Apr 30 11:07:50 Infinity kernel: md: recovery thread: exit status: -4

That is so close to on the hour that I wonder if it isn't a scheduled check. And then you cancel it several minutes later.

 

What do you get from the command line with this?

crontab -l

and this?

cat /etc/cron.d/root