I am running latest stable. I did a parity check (30 January) and 696 errors showed up. Then I did next day another one and the same 696 errors were found.

Is there a way I can check what errors did parity check find?

Anything I should to look into?

Rgds.

Edited February 3, 20224 yr by luca2
SOLVED

Repeated errors like that mean that you haven't as of yet run a correcting parity check to actually fix them.

23 minutes ago, Squid said:

Repeated errors like that mean that you haven't as of yet run a correcting parity check to actually fix them.

I do have the check box marked to perform corrections to parity. How can I check for this to perform the needed corrections?

 

Diagnostics will show if you didn't reboot yet.

On 1/31/2022 at 1:40 PM, JorgeB said:

Diagnostics will show if you didn't reboot yet.

OK here we go. Atttached are my diagnostics. Pls, let me know what could be happening.

 

Edited February 3, 20224 yr by luca2

Jan 29 10:39:42 shogun emhttpd: unclean shutdown detected
...
Jan 29 10:40:03 shogun kernel: mdcmd (36): check nocorrect
...
Jan 30 05:45:16 shogun kernel: md: sync done. time=68713sec
Jan 30 05:45:16 shogun kernel: md: recovery thread: exit status: 0

Unclean shutdown automatic parity checks are non-correcting.

 

Then you manually ran a correcting parity check which corrected the errors the non-correcting parity check found.

Jan 30 12:44:39 shogun kernel: mdcmd (37): check 
Jan 30 12:44:39 shogun kernel: md: recovery thread: check P ...
Jan 30 12:44:39 shogun kernel: md: recovery thread: P corrected, sector=0
...
Jan 30 20:07:40 shogun kernel: md: recovery thread: P corrected, sector=5438978488
Jan 30 20:07:40 shogun kernel: md: recovery thread: stopped logging
...
Jan 31 08:00:55 shogun kernel: md: sync done. time=69376sec
Jan 31 08:00:55 shogun kernel: md: recovery thread: exit status: 0

Run another non-correcting parity check to verify.

 

Do you know why you had unclean shutdown?

Jan 29 10:40:03 shogun kernel: mdcmd (36): check nocorrect

Looks like check run on 29th was non-correcting.

 

Jan 30 12:44:39 shogun kernel: mdcmd (37): check 

 

whereas the one run 0n 30th was a correcting check.   That would explain why they gave the same number of errors found.

 

if you now run a non-correcting check it should find 0 errors as they are (hopefully) now all corrected.

 

On 1/31/2022 at 3:02 PM, trurl said:

Jan 29 10:39:42 shogun emhttpd: unclean shutdown detected
...
Jan 29 10:40:03 shogun kernel: mdcmd (36): check nocorrect
...
Jan 30 05:45:16 shogun kernel: md: sync done. time=68713sec
Jan 30 05:45:16 shogun kernel: md: recovery thread: exit status: 0

Unclean shutdown automatic parity checks are non-correcting.

 

Then you manually ran a correcting parity check which corrected the errors the non-correcting parity check found.

Jan 30 12:44:39 shogun kernel: mdcmd (37): check 
Jan 30 12:44:39 shogun kernel: md: recovery thread: check P ...
Jan 30 12:44:39 shogun kernel: md: recovery thread: P corrected, sector=0
...
Jan 30 20:07:40 shogun kernel: md: recovery thread: P corrected, sector=5438978488
Jan 30 20:07:40 shogun kernel: md: recovery thread: stopped logging
...
Jan 31 08:00:55 shogun kernel: md: sync done. time=69376sec
Jan 31 08:00:55 shogun kernel: md: recovery thread: exit status: 0

Run another non-correcting parity check to verify.

 

Do you know why you had unclean shutdown?

Hi, sorry for delay, but could not test it till yesterday. You are right, this is exactly what happened. I ran a third parity check and it showed no errors.

 

I don´t know why I had an unclean shutdown. Sometimes the server just cannot shutdown, so I have to do a hard reset. Usually it has to when I have been messing up editing the xml of the vm´s.

 

Thx another time for support! Have a nice day.

1 hour ago, luca2 said:

why I had an unclean shutdown

 

On 2/3/2022 at 1:53 PM, trurl said:

Thx for the tip to this thread Trurl!. Had no idea about this. Really usefull. I will check if my unclean shutdows now dissapear.

:)