Few days ago started getting errors in parity drive.

 

I have 2 drives that are 2TB each and the Parity that is 3TB.

 

I replaced the parity with a new 3 TB but that is giving me errors too.

 

syslog.zip

Few days ago started getting errors in parity drive.

 

I have 2 drives that are 2TB each and the Parity that is 3TB.

 

I replaced the parity with a new 3 TB but that is giving me errors too.

Looking at the syslog I would think that there may well be cabling issue to that drive.  You might want to ensure that the cables are firmly seated, and if you have another SATA cable that could be worth trying.  How is the drive connected - is it directly to the motherboard/controller or via a hot-swap cage?

The drive is connected directly from the drive to the motherboard.  I will try another cable.

 

My case is full of dust I am going to clean that as well.

I cleaned the case and reseated the cable and parity is fine now.

 

Thank you

Few days ago started getting errors in parity drive.

 

I have 2 drives that are 2TB each and the Parity that is 3TB.

 

I replaced the parity with a new 3 TB but that is giving me errors too.

Looking at the syslog I would think that there may well be cabling issue to that drive.  You might want to ensure that the cables are firmly seated, and if you have another SATA cable that could be worth trying.  How is the drive connected - is it directly to the motherboard/controller or via a hot-swap cage?

 

What in the logs indicated cabling issue? I am ignorant to diagnosing such issues via these logs, looking to learn what I can.

What in the logs indicated cabling issue? I am ignorant to diagnosing such issues via these logs, looking to learn what I can.

 

I wrote something that might be helpful ->  The Analysis of Drive Issues

 

It was written quite awhile ago, so may be dated, and I'm sure more could be added.  The primary thing that indicates packet corruption across a cable is CRC errors, the BadCRC or ICRC flags in syslog exception handler messages, or an increase in the UDMA CRC count in SMART reports.  This particular case has no CRC errors at all, so I'm not sure the problem was a bad cable.  There clearly were serious handshake problems, bad enough to call them 'interface fatal errors'.  I would normally associate that with possibly bad power issues, heat issues (involving the motherboard, controller, or bridge chipsets), bad controller problem, bad SATA port problem, or possibly incompatibilities between the old board and new drive.  I don't know.  Hard to know how dust would be involved.  If it happens again, I would switch to a different SATA port.

What in the logs indicated cabling issue? I am ignorant to diagnosing such issues via these logs, looking to learn what I can.

 

I wrote something that might be helpful ->  The Analysis of Drive Issues

 

It was written quite awhile ago, so may be dated, and I'm sure more could be added.  The primary thing that indicates packet corruption across a cable is CRC errors, the BadCRC or ICRC flags in syslog exception handler messages, or an increase in the UDMA CRC count in SMART reports.  This particular case has no CRC errors at all, so I'm not sure the problem was a bad cable.  There clearly were serious handshake problems, bad enough to call them 'interface fatal errors'.  I would normally associate that with possibly bad power issues, heat issues (involving the motherboard, controller, or bridge chipsets), bad controller problem, bad SATA port problem, or possibly incompatibilities between the old board and new drive.  I don't know.  Hard to know how dust would be involved.  If it happens again, I would switch to a different SATA port.

 

 

Thanks for the link, and the explanation. If excessively dusty, the chipset or controller could have been overheating I suppose... /shrug That is why I put an intake fan over a furnace filter housing on my server room, haha. Keep that dust out.