(Solved) Low Parity Sync Errors - Next Step Advice

[SuudoXR]

Hello all,

 

Been running Unraid for about 3 months now and been really happy with everything it offers. This morning my scheduled monthly non-corrective parity check came back with 23 errors. Since the previous month I have not had any unplanned power outages but I have had a clean shutdown during a storm to protect against one as I don't have a UPS currently. 

 

In terms of other events that may seem a little odd in the logs I do use a 'user-script' to stop and start my docker containers on a daily bases i.e. stop plex, radarr, sonarr and sabnzbd overnight when I'm not using them. I did this originally as I was spinning my drives down overnight but I no longer do this so probably will remove that script once this is sorted.

Reading the forums I get the feeling that when a small number of errors occur that I should run a correct parity check and then see what the following month brings. However after reading the horror stories of people making changes before seeking advice I wanted to air on the side of caution and get some feedback

 

I have attached the diagnostics and would appreciate any comments suggestions on my next steps. Apologies in advance if I have missed some information!

 

Cheers,

Danny

kronk-diagnostics-20200901-0824.zip

Edited September 2, 2020 by SuudoXR
Mark as solved

[JorgeB]

Run another non-correcting check without rebooting and post new diags.

[SuudoXR]

Thanks for the reply Johnnie, running now, will report back in 8 hours.

 

Danny

[SuudoXR]

@johnnie.black The second non-correcting check has completed, it reported the same 23 sectors as before, I have attached diags as requested.

 

Thanks again,

Danny

kronk-diagnostics-20200901-1707.zip

[JorgeB]

Since they are exactly the same errors on the same blocks it rules out for any transient memory error, are you absolutely sure there wasn't any unclean shutdown? It's what makes more sense, it could be a disk but those cases are extremely rare and the sectors are not all together so even more unlikely, in any case and IMHO you should run a correcting check and as long as there are no more errors in the future you're fine.

[SuudoXR]

OK, i guess since I haven't set up syslog server your correct that there is the chance I missed one, as you say the evidence fits!

 

I will run corrective and setup syslog just so I can review better in future.

 

Thanks for taking the time to help me.

 

Cheers,

Danny