I've just finished replacing a 1 TB drive with a new 2 TB drive. The drive is a new Hitachi Deskstar 5K3000 2 TB drive. I've followed what I believe are the recommended procedures, which include:

 

1. Stopped the array.

2. Shutdown the server.

3. Connected new drive to system (have extra unused SATA port, so the existing array is still fully connected)

4. Started server

5. Ran preclear with screen. The Hitachi Deskstar 5K3000 drive is not an advanced format drive, so I didn't use the -A option. Preclear ran fine, but took just over 24 hours.

6. Stopped the array.

7. Shutdown server.

8. Swapped the existing 1 TB drive with the new 2 TB drive.

9. Started server

10. Started the array - it rebuilt the data in the new drive. Everything went fine, no errors were displayed after everything was finished approximately 12 hours later.

 

I just woke up (it's the middle of the night) and I decided to see if the rebuilding was finished. It was, so I decided to poke around the unRAID unMenu and I clicked on Syslog near the top of the screen. I noticed several lines highlighted in bright red, which indicates errors. These errors start at the lines starting with Sep 23 14:16:55. I'm running unRAID Server Plus version 4.7.

 

I've attached a copy of my syslog and was wondering if these errors are anything to worry about?

 

Thanks in advance for any help/advice,

Gregg

syslog-2011-09-24.txt

BadCRC means bad SATA cables. Check the wiki: http://lime-technology.com/wiki/index.php?title=The_Analysis_of_Drive_Issues

 

The syslog identifies ata7. Is it referring to the SATA 7 port?

 

It's interesting that I would get this error, which indicates a bad connection, but not have any drive issues or errors.

The syslog identifies ata7. Is it referring to the SATA 7 port?

 

It's interesting that I would get this error, which indicates a bad connection, but not have any drive issues or errors.

No, it is the 7th port that initialized in your hardware.... unfortunately, I do not know a way to identify which disk it is other than to see where in the syslog it appears and which disks are adjacent in the log.

 

It does NOT indicate a bad connection, but instead a bad checksum across the connection.  That is VERY different..  The cable might be picking up induced noise if it is tightly cabled with other cables...  (is it zip-tied to other SATA cables? Or other power cables?)  Is it a poor quality cable, with conductors that are not shielded perhaps? or not twisted internally to reduce noise pickup?

 

It could easily indicate a noisy power supply causing the poor communications... are you using multiple splitters for power? They too can result in poor power supply regulation.

 

Joe L.

I did my best to clean up the cabling. I double checked all the connections between the motherboard and hard drives - SATA cables and power. I made sure that any cable from the power supply only fed at the most 2 hard drives. I also made sure that the cables were only supplying power to the hard drives. I had one that had a couple 5V or 3.3V wires feeding fans, so I used a power supply cable that wasn't feeding anything else. I did my best to untangle some of the cabling.

 

This seems to have helped. After rebooting, it looks like all these errors have disappeared from my syslog.

 

Thanks all of you for your help.

 

-Gregg

syslog-2011-09-25.txt