November 8, 201114 yr It looks like I have had my first drive failure. The drive is missing from the array for no apparent reason. Is there a process that I should go through to ensure that the drive actually is dead?
November 8, 201114 yr perhaps posting a syslog and smart report we might be able to guess a little better. note any changes you might have made.
November 9, 201114 yr Author Attached is a system log and smart report. finding a drive these days might be a little expensive. smart.txt syslog.txt
November 9, 201114 yr Attached is a system log and smart report. finding a drive these days might be a little expensive. which drive is missing? I don't see evidence in the syslog.
November 9, 201114 yr Attached is a system log and smart report. finding a drive these days might be a little expensive. Wow, 2TB drives shot up in price. I've been under a rock for awhile. Why are these expensive now days?
November 9, 201114 yr Attached is a system log and smart report. finding a drive these days might be a little expensive. Wow, 2TB drives shot up in price. I've been under a rock for awhile. Why are these expensive now days? Thailand Flooding
November 9, 201114 yr If memory serves its /dev/sdc, disk 6. As far as I can see, it is there in the syslog: Nov 8 19:07:33 Unraid kernel: md: unRAID driver 1.1.1 installed Nov 8 19:07:33 Unraid kernel: md: import disk0: [8,48] (sdd) ST32000542AS 6XW048F6 size: 1953514552 Nov 8 19:07:33 Unraid kernel: md: import disk1: [8,64] (sde) WDC WD7500AACS-0 WD-WCASM0003415 size: 732574552 Nov 8 19:07:33 Unraid kernel: md: import disk2: [8,112] (sdh) ST3500630AS 6QG09VT9 size: 488385492 Nov 8 19:07:33 Unraid kernel: md: import disk3: [8,96] (sdg) ST31500541AS 5XW0AHXN size: 1465138552 Nov 8 19:07:33 Unraid kernel: md: import disk4: [8,80] (sdf) ST31500341AS 9VS1V882 size: 1465138552 Nov 8 19:07:33 Unraid kernel: md: import disk5: [8,16] (sdb) ST32000542AS 6XW054H9 size: 1953513492 Nov 8 19:07:33 Unraid kernel: md: import disk6: [8,32] (sdc) ST32000542AS 5XW120SK size: 1953513492 and I see no failures. Did you post the correct syslog?
November 17, 201114 yr Author So out of curiosity I moved the disk to a second hot swap bay/sata port. after a reboot drive 6 showed as missing and no other drives were waiting to be assigned. I shut down the system and moved the drive back to the original bay and now the suste thinks that it is a new drive. thoughts? Also I took another syslog before doing all this. syslog-2011-11-16.txt
November 22, 201114 yr Author So, I got frustrated with this and ha an extra disk do I went ahead an popped it in and the array rebuilt and everything seemed fine. Last night I go to the web interface and the same disk is missing again. This indicates to me that there must be some other hardware issue. Maybe the hot swap backplane or the cable so I'll start there. Also, I think I'll test my original drive and if it's good add it back to the system once I find the problem. Thoughts? Advice?
November 22, 201114 yr So, I got frustrated with this and ha an extra disk do I went ahead an popped it in and the array rebuilt and everything seemed fine. Last night I go to the web interface and the same disk is missing again. This indicates to me that there must be some other hardware issue. Maybe the hot swap backplane or the cable so I'll start there. Also, I think I'll test my original drive and if it's good add it back to the system once I find the problem. Thoughts? Advice? no, you have an intermittent hardware issue. You must isolate. Nothing else anyone can do for you. It could be a loose cable, or, it could be a power supply splitter, or a drive tray, or even a power supply inadequate to supply all your disks. Just remember, a disk that has been disabled was disabled because a write to it failed. (it is guaranteed to be incorrect) It will not go back to green without you un-assigning it, starting the array with it un-assigned, re-assigning it, then starting the array once more and letting it re-construct the drive with the correct data, as re-constructed from parity and all the other data disks.
November 22, 201114 yr Author If I move one of the good disks to a different data port will it still be recognized?
November 22, 201114 yr If I move one of the good disks to a different data port will it still be recognized? depends on the version of unRAID. In older releases, you'll probably need to use the devices assignment page to assign it to its correct slot. On the newest beta release, it will probably recognize it on its own by the serial number.
November 25, 201114 yr Author all SATA cables are secure and seem to be fine. I took out a power splitter that I had in there and rearranged the power cords to ensure that the hot swap black planes that were actually being used were connected directly to the power supply, its a 750W corsair. The disk still didn't show up so I assume that it is an issue with the specific slot on the SATA backplane. I moved the disk to another slot on a different backplane and the disk was recognized and parity is being rebuilt. While I have the machine on the workbench is there anything else that I should check? Also, is there a log or something that i can get from the boot up process to see if I need to update the bios for the motherboard or the SATA add on card?
November 26, 201114 yr Author Something is very wrong here. I got almost 6% through the array rebuild and the parity drive dropped out and is now missing as well.
November 27, 201114 yr Author the server consists of a super micro x7sla atom based motherboard, a supermicro aoc-saslp-mv8 data expansion card, 7 assorted hard drives and a 750 watt corsair power supply. This is all in a norco 4020 case.
Archived
This topic is now archived and is closed to further replies.