Hi I'm a newbie, well, I've been using Unraid for a couple of years without issue.

 

Unraid v6.8.1 is crashing 0-3% into Parity check. I don't know what's causing the issue, can anyone help?

 

Diagnostics attached.

 

Thanks!

unraid-diagnostics-20200124-0241.zip

 

System info:

ASRock B365M Phantom Gaming 4 Version - s/n: M80-C1009500474

BIOS: American Megatrends Inc. Version P4.20. Dated: 05/15/2019

CPU: Intel® Core™ i7-8700 CPU @ 3.20GHz

HVM: Enabled

IOMMU: Enabled

Cache: 0 KiB, 0 KiB, 0 KiB

Memory: 32 GiB DDR4 (max. installable capacity 64 GiB)

Network: eth0: 1000 Mbps, full duplex, mtu 1500

Kernel: Linux 4.19.94-Unraid x86_64

Edited January 24, 20206 yr by circa1665
Added system info

Diags are just after rebooting, see here to try and catch the problem.

Apologies, please find logs attached.

 

syslog-192.168.232.133.log

 

I recently had to replace x2 drives as they failed. I have dual parity drives and the data rebuilt no problem.

There was a period a couple of days ago when the Ethernet switch my Unraid server is connected to stopped working (have since replaced it) - the server did not have network access for about 12 hours - don't know if that could be related... Now I'm noticing that both drives I replaced are throwing UDMA CRC errors, every time Parity check fails and it crashes the error count goes up (see screenshot). Both drives were pre-cleared without issue and had been working fine up until now.

3 disks (including disks 3 and 4) are connected using a SATA port multiplier (can't say if it's embedded on the Asmedia controller or external), those are not recommended and can cause several issues, first thing would be to try without it.

 

 

I have x6 SATA ports on my motherboard and a x4 ports on a PCI-E card (AC1539).

I don't have enough ports on the motherboard for all my drives, should I be using a different type of card? Am happy to to try - but like I said in my original post have had this current server for 2 years without issue and a the same drives in a previous self build for 3 (using the same PCI-E expansion card) <<< obviously not the two drives (3 and 4) which I recently replaced.

 

5 minutes ago, circa1665 said:

should I be using a different type of card?

Yes, that's a 2 port Amsedia controller with a port multiplier, if you need just 2 ports a regular 2 port Asmedia is fine, if you need 4/5 ports there's this one, or get an 8 port LSI, they can be bough used for $40/50.

Note that I'm not saying it's that controller crashing your server, since the log posted doesn't show the crash, but they are a known source of problems, possibly including the crashes.

Would this be suitable? https://www.ebay.co.uk/itm/LSI-Logic-MegaRaid-SAS-8708EM2-SAS-SATA-RAID-Controller/372924841101

Nope, that's a SAS1 model limited to 2TB disks, you want any LSI with a SAS2008/2308/3008/3408 chipset in IT mode, e.g., 9201-8i, 9211-8i, 9207-8i, 9300-8i, 9400-8i, etc and clones, like the Dell H200/H310 and IBM M1015, these latter ones need to be crossflashed.

 

 

Cool, thank you for your help 👍 will look into it, will replace with this: https://www.amazon.co.uk/gp/product/B07X21J35K/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1

Hi, so replaced the SATA card with the one above. Now Parity check doesn't cause Unraid to crash, but the Parity check aborted around the 7-8% mark. 

unraid-diagnostics-20200126-1237.zip

The diags posted don't have any parity check starting or aborting.

Hmmm, sorry about that. Running parity check again. Will post again when it either fails or completes.

Parity check completed successfully! So i guess that you were right - it must have been the SATA card, Thank you. I've attached the logs - hopefully all good now?

 

Am I fine to acknowledge the UDMA CRC errors on Drives 3 and 4?

 

Thanks again for you help, much appreciated 👍

syslog-192.168.232.133.log

12 hours ago, circa1665 said:

Am I fine to acknowledge the UDMA CRC errors on Drives 3 and 4?

Yep.