January 18, 201412 yr What would cause ? Jan 17 22:45:53 Tower kernel: md: disk1 read error, sector=848314944 Jan 17 22:45:54 Tower shfs/user: shfs_readdir: readdir_r: /mnt/disk1/storage (5) Input/output error Jan 17 22:45:54 Tower last message repeated 2 times Jan 17 22:45:54 Tower kernel: md: disk2 read error, sector=848314944 Jan 17 22:45:54 Tower kernel: md: disk3 read error, sector=848314944
January 18, 201412 yr Author Here is a syslog file. Before I went to bed I set it to run long smart test on each drive. By the morning it had errors and I log into web console and 3 drives were "missing". When I rebooted they all showed up fine. I had this happen about 2 weeks ago and after about 3 times of it one drive finally died. I do not understand why it would cause issues with other drives cause I can never tell which one is going bad till it is completely bad. Could it be the sas controller. I got a new one about a month ago (SUPERMICRO AOC-SAS2LP-MV8) before I was using the SUPERMICRO AOC-SASLP-MV8. With this log file these rear errors start at line 1799. syslog-20140118-114139.zip
January 18, 201412 yr Author New firmware just came out a few days ago for the sas controller, thinking about upgrading.
January 19, 201412 yr Author I think it is my disk 3 going bad. I pulled all the drives and tested them in windows with the linux reader program and all of them opened and I could see the ReiserFS, but disk 3 had other volumes with all chinese/japanese characters. All other disks just had the reiserfs. I started unraid without disk 3 and running smart long tests on 1,2,and 4 and so far they are going through without crashing like they do when disk 3 is in. I am going to replace 3 with a spare. Anyone ever see these weird volumes?
January 19, 201412 yr Zounds I must have blinked and missed some major leaps in drive technology! Your Disk 3 has a volume of size greater than 6,000,000,000 Gb. Is that 6 EXA Bytes? or is it 6 ZETA Bytes? Back in the old days (when mullets and cassette tapes were cool), drives could be reinitialized and there was software to (maybe) fix and (maybe) recover from disk partition errors. I'm not even sure if that's possible with current hard drive technology as its all moved onto the drive itself. If it were me, I'd see if any files are still readable and perhaps copy them to another drive. I'm not sure I'd *Ever* trust that drive again, NOR would I assume the files copied from the drive were accurate. I'd plop a replacement drive 3 into unRAID and let it rebuild...while keeping my fingers crossed.
January 19, 201412 yr Author Yup, I took it out and put another 2tb in its place. Unraid did its thing to recover. I still do not know why when one of my drives goes bad it tries to affect a few of the other drives on the same sas connector.
Archived
This topic is now archived and is closed to further replies.