February 23, 200818 yr Nothing has changed on my Array... no new disks, case hasn't been opened... haven't touched the box... I posted Screenshots in another thread this AM and everything was fine. All of a sudden one of the disk shows up with a Blue indicator on my status page....ZERO errors reported during any of the file copy operations that I've been doing over the past week or so... I have rebooted and shutdown and restarted... 1) Why would this have happened, how can I prevent from happening agin 2) How do I recover the data that may have been on that disk and get my array going again?
February 23, 200818 yr Author Syslog Attached.... Since I rebooted already might not be helpful, not really sure...
February 24, 200818 yr Your syslog does not show what happened earlier, but does show problems with Disk 5 (sdc). I researched 'PHY RDY changed' and it seems to be involved with an improper hotplugging, similar to in Windows yanking a removable drive too quick, without doing the proper procedure, such as 'Safely Remove Hardware'. It looks possibly like a brief disconnection. I wonder if this drive is in a removable cage or carrier, and is a little loose, perhaps vibrating loose? Or perhaps the power or data cable to it is loose? Other ideas, you might want to try replacing the cable to it; also try moving it to a different connector. If everything is fine, and this same drive keeps giving these errors, you may have to suspect it as defective. (I should add that I am not a Linux expert, so my analysis may be flawed.) Your syslog looks great apart from that. You didn't start the array, so I can't tell how mounting the file systems will go, but I only expect to possibly see a few transactions replayed on md5, because it may have been unable to shutdown properly. After trying the various hardware suggestions above and below, you will be up and running faster if you select the Start option, and rebuild the 400GB Disk 5, than choosing the Restore option, which will begin a 750GB parity sync. You should know pretty quickly on starting the array if it will be successful. If it runs fine for the first few minutes, then you probably have no data loss at all, and that is what I expect. One other hardware thing you may want to check: at least 3 of the drives attached to the motherboard are being linked at 1.5 Gbps, instead of 3.0 Gbps. That either means they are an older 1.5 Gbps drive, or they still have their SATA limiter jumper on the back. I can't tell about the ones connected to the 8 SATA port card, because they are in some kind of IDE mode. I would check all of the hard drives for those jumpers, and remove them. Those jumpers are only required if you discover a compatibility problem between the drive and an older controller.
February 24, 200818 yr Author Thanks for the reply Rob...The drive is in a 5in3 IcyDock in a Tray... I did shut down the system and pull and re-seat the tray already just in case. Once the rebuild is done (almost there), I will shut it down and swap the cable on the array as well. I will also check the jumpers... I think the 400GB Drives in the array are some older 7200.8 Drives which if I remember did not support 3.0, but will double check anyhow... I did a "Start Array" as you suggested and it is almost done rebuilding just that one drive.
February 25, 200818 yr Author ... and new Syslog before a Reboot or anything... Changed the SATA Cabled, Removed, Reseated the Tray, rebuilt the data on that drive... WITH some errors reported now...
February 25, 200818 yr Yeah, that same drive just cannot seem to stay connected. Thankfully, there is NO evidence of data problems, your data on it is probably intact, but that is not much use if you can't maintain communication with the drive. It keeps cutting out, until unRAID gives up on the drive and disables it. You might try removing it completely from any trays, and connecting it directly to the motherboard, on a different port and cable. The 750GB is now showing a link speed of 3.0 Gbps, it was 1.5 before.
February 25, 200818 yr Author Yeah, that same drive just cannot seem to stay connected. Thankfully, there is NO evidence of data problems, your data on it is probably intact, but that is not much use if you can't maintain communication with the drive. It keeps cutting out, until unRAID gives up on the drive and disables it. You might try removing it completely from any trays, and connecting it directly to the motherboard, on a different port and cable. The 750GB is now showing a link speed of 3.0 Gbps, it was 1.5 before. Yah, my 750 gb Parity and 750gb Disk 1 both had the jumper On it... I think I have another 300 or 400 GB @ work, so I might just go ahead and replace that disk 5 and see what happens....
February 25, 200818 yr Author CRAP... now I work up to see this... Drive 4 is 5th drive in the first 5-in-3 box and Drive 5 is the first drive in the 2nd Backplane...
February 25, 200818 yr From hard experience I have learned, that after I have been working inside a computer case, to snug every cable connection (especially non-locking SATA cables), before closing it up. It's possible that with the work you have been doing inside, that something has come loose.
February 26, 200818 yr This is the kind of thing I saw when using the round IDE cables. Replacing them with the flat ones fixed it. You said these were SATA drives, but thought I'd mention it anyway.
February 26, 200818 yr Lots of companies sell too-long SATA cables, so make sure they are in spec (18"). Bill
February 26, 200818 yr Author OK, I pulled Disk 4 Out and plugged into a new SATA Cable and Power... and it appears to be online and "green" again. Disk 5 is still "RED" and needs to be replaced, I assume... So question...if the "failed" drive is a 400 GB, I can't install a "smaller" drive in its place??? The could be a problem... I don't have a bigger drive available right now...I have 360GB Seagate drive... "You must replace a failed disk with a disk which is as big or bigger than the original and not bigger than the parity disk. If the replacement disk is larger than your parity disk, then the system permits a special configuration change called swap-disable."
February 26, 200818 yr Author Ok, I need some help! I replaced the Failed 400 GB Drive #5 with a new 360GB Drive... I know that the instructions say that It has to be the same or larger, but I assume that is if I want or need to save the data that was on the failed drive. If I don't care (and still have a few files that were on that drive saved somehwhere else, which I do), then can I, and should I just use the "Restore" feature without losing everything on all my other drives, correct? Please help! Thanks SW2
February 26, 200818 yr If I don't care (and still have a few files that were on that drive saved somehwhere else, which I do), then can I, and should I just use the "Restore" feature without losing everything on all my other drives, correct? Correct.
February 26, 200818 yr Author If I don't care (and still have a few files that were on that drive saved somehwhere else, which I do), then can I, and should I just use the "Restore" feature without losing everything on all my other drives, correct? Correct. Thank you... powering back up and clicking restore in a couple mins <fingers crossed> SW2
Archived
This topic is now archived and is closed to further replies.