November 24, 200916 yr I originally posted here... http://lime-technology.com/forum/index.php?topic=4631.0 However, after successfully swapping out what I thought was a drive that was going bad (disk 6 in the array), after a power outage, the UnRAID server has come back online and disk 6 has a red indicator next to it again. Now this drive was replaced with a large brand new Western Digital drive and, although it is possible that this drive has also failed, I think the odds are more in favour of something else going wrong. Either I've got a faulty SATA cable, the Promise Superswap 1100 drive cage has developed a fault, or the Promise 300 TX4 SATA controller has developed a fault. I tried swapping out the SATA cable with another I had spare but, after bringing the server back up, the disk is still showing the red indicator. Is anyone able to advise me how to narrow it down to what the fault is likely to be? I'm REALLY keen to get this sorted out!
November 24, 200916 yr The red ball will not go away even if you have resolved the problem. Read about the Try "trust my parity" procedure in the wiki.
November 24, 200916 yr I originally posted here... http://lime-technology.com/forum/index.php?topic=4631.0 However, after successfully swapping out what I thought was a drive that was going bad (disk 6 in the array), after a power outage, the UnRAID server has come back online and disk 6 has a red indicator next to it again. Now this drive was replaced with a large brand new Western Digital drive and, although it is possible that this drive has also failed, I think the odds are more in favour of something else going wrong. Either I've got a faulty SATA cable, the Promise Superswap 1100 drive cage has developed a fault, or the Promise 300 TX4 SATA controller has developed a fault. I tried swapping out the SATA cable with another I had spare but, after bringing the server back up, the disk is still showing the red indicator. Is anyone able to advise me how to narrow it down to what the fault is likely to be? I'm REALLY keen to get this sorted out! To determine if the disk is responding, you must run a smartctl command on it. Then you can swap the drive to a different port, use a different cable, etc until it does respond.
November 24, 200916 yr Author To determine if the disk is responding, you must run a smartctl command on it. Then you can swap the drive to a different port, use a different cable, etc until it does respond. Thank you. I will try this tonight.
November 25, 200916 yr Author OK, I ran the smartctl command on disk 6 and it looks fine. I then followed the "trust my array" procedure, and the array started up fine and is currently going through a parity check. The only small concern I have is that the Sync Errors count is showing as 80. As the parity check continues and I refresh the page, that number is not incrementing - which I guess is a good thing - but, do I need to be concerned about those sync errors? FWIW, the server was shut down in a power cut, but is that too many sync errors for that situation?
November 25, 200916 yr OK, I ran the smartctl command on disk 6 and it looks fine. I then followed the "trust my array" procedure, and the array started up fine and is currently going through a parity check. The only small concern I have is that the Sync Errors count is showing as 80. As the parity check continues and I refresh the page, that number is not incrementing - which I guess is a good thing - but, do I need to be concerned about those sync errors? FWIW, the server was shut down in a power cut, but is that too many sync errors for that situation? It is not unexpected... Those are probably in the housekeeping areas of the disk. (the data disk has a journal to replay its partially completed operations, but the parity disk does not. The un-expected power cut just got them out of sync. The parity check should fix things. You probably should run a second party check once this initial one completes. The second one should show no additional errors.
November 27, 200916 yr Author Just responding to say that 2nd parity check has now completed. All done with no sync errors. Thanks to all!
Archived
This topic is now archived and is closed to further replies.