Hi folks,

Yesterday my beloved dog stepped on the switch of the power strip of my server.

 

After scratching his ears and giving him a treat to show him who's boss, I restarted the server, and launched a parity check.

I was relived when I got this message :
 

Event: Unraid Parity-Check
Subject: Notice [SAUCISSE] - Parity-Check finished (0 errors)
Description: Duration: 1 day, 6 hours, 46 minutes, 53 seconds. Average speed: 144.4 MB/s
Importance: warning

 

To my surprise I got this other message 5 minutes latter :
 

Event: Parity Check Tuning
Subject: [SAUCISSE] Automatic Correcting Parity-Check finished (2414 errors)
Description: Elapsed Time 2 day, 3 hr, 56 min, 1 sec, Runtime 2 day, 3 hr, 56 min, 1 sec, Increments 1, Average Speed 85.6 MB/s
Importance: alert

 

I fail to understand the difference in error number. Could you help me with that ?

Bonus question : how can I know where/in which files are these 2414 errors ?

 

 

saucisse-diagnostics-20240503-2010.zip

Parity check related notifications have been known to sometimes be wrong, they show the previous results, according to the previous-syslog there were sync error, but they should now be corrected.

 

You can always look at the "History" stats on main, the stats for the last check will be correct there.

13 hours ago, Majyk Oyster said:

I fail to understand the difference in error number. Could you help me with that ?

I suspect the Parity Check Tuning plugin is correct as it tracks the check using a different mechanism to the built-in Unraid method, and Unraid has been known to at time incorrectly display the result of the previous check.  You might also know what the true elapsed time of the check was as the two reports give very different values.?

 

13 hours ago, Majyk Oyster said:

Bonus question : how can I know where/in which files are these 2414 errors ?

 

Unfortunately you cannot do this in practice.  The sectors that have errors are listed in the syslog but there is no way to know which drive generated them, or even if you knew that easily relate it to any specific file.   The only way to find out is to have checksums for your file or to be able to compare them to your backups.   If using XFS in the array then the File Integrity plugin can be used to generate these, or if using BTRFS or XFS they have check-summing built in at the file level.   The algorithm used for writing drives is such is that barring errors (which the drives should report back to Unraid) the parity drive is the one most likely to be in error.

 

If you have had any unclean shutdowns then a small number of errors are to be expected.  If not then doing some hardware checks such as extended SMART tests on your drives or a RAM check are a good idea.

Thanks for the replies ! 🍻

 

  

14 hours ago, JorgeB said:

Parity check related notifications have been known to sometimes be wrong, they show the previous results, according to the previous-syslog there were sync error, but they should now be corrected.

 

You can always look at the "History" stats on main, the stats for the last check will be correct there.

Here's what the history looks like :

 

I'm running another parity check right now, we'll see.

 

14 hours ago, itimpi said:

You might also know what the true elapsed time of the check was as the two reports give very different values.?

I have no idea why the elapsed time differ, looks like they run at the same time or something. Can't tell which would be right.

 

 

14 hours ago, itimpi said:

Unfortunately you cannot do this in practice.  The sectors that have errors are listed in the syslog but there is no way to know which drive generated them, or even if you knew that easily relate it to any specific file.   The only way to find out is to have checksums for your file or to be able to compare them to your backups.   If using XFS in the array then the File Integrity plugin can be used to generate these, or if using BTRFS or XFS they have check-summing built in at the file level.   The algorithm used for writing drives is such is that barring errors (which the drives should report back to Unraid) the parity drive is the one most likely to be in error.

I only have backups of a small portion off the array, the critical stuff. I'm going to check this, and hope for the best for the media library.

I'm indeed using XFS, I'm looking into the File Integrity plugin right now.