splnut Posted March 7, 2016 Share Posted March 7, 2016 The other day I noticed that I was affected by the Marvell firmware issue (http://lime-technology.com/forum/index.php?topic=40683.0). When I accessed unRAID, there were 3 disks missing. I shutdown the array, checked the connections, upgraded the Marvell firmware and started it back up. All the drives are found, except 1 disk is showing unmountable. Following the unraid wiki, I started the array in maintenance mode and ran xfs_repair. Phase 1 - find and verify superblock... bad primary superblock - bad magic number !!! attempting to find secondary superblock... ...Sorry, could not find valid secondary superblock Exiting now. What are my next steps? Should I let unRAID format the drive and rebuild data? tower-diagnostics-20160307-1834.zip Quote Link to comment
JorgeB Posted March 8, 2016 Share Posted March 8, 2016 It’s late here and I’m going to bed, just some quick observations: -Don’t hit format, that won’t help in any circumstance. -Disk7 is disabled and unmountbale, rebuilding it will result in the same issue, as you're seeing the emulated disk, the end result of a rebuild. -Looks like disk3 dropped offline, it’s showing errors and no SMART info, is it still green? Post a screenshot of your mainpage. Quote Link to comment
splnut Posted March 8, 2016 Author Share Posted March 8, 2016 Here is a screenshot. I saw the disk3 errors as well, and when I restated the server I got a green notification about disk3. Although it's had a green ball the whole time. I think I will just not use the marvel controller to avoid further issues, but hopefully can prevent any data loss. Quote Link to comment
JorgeB Posted March 8, 2016 Share Posted March 8, 2016 The marvell virtualization issue usually manifests itself differently, with vt-d enable you can’t see any disks, they don’t usually drop out, but disable vt-d if you can just in case, or better yet and if you have one, use a different controller for now. If you rebooted since the first diagnostics post a new one to see if there’s a SMART report for disk3. Quote Link to comment
splnut Posted March 8, 2016 Author Share Posted March 8, 2016 Here is the disk 3 smart report and a new diagnostics after a reboot. tower-smart-20160308-0742.zip tower-diagnostics-20160308-0743.zip Quote Link to comment
JorgeB Posted March 8, 2016 Share Posted March 8, 2016 Since rebuilding disk7 won’t make any difference and the disk itself looks healthy I would try a new config with the actual disk, for now without parity. -Go to tools and click new config -Reassign all disks except parity (no disk in the parity slot) -Start array and check if disk7 mounts Don’t write or delete anything from the array or you'll invalidate parity. Quote Link to comment
splnut Posted March 8, 2016 Author Share Posted March 8, 2016 That appears to work. Should I now assign the parity disk and start the array? Quote Link to comment
JorgeB Posted March 8, 2016 Share Posted March 8, 2016 You can, but since your system is not very stable it may not complete. At least disable vt-d in your bios in case it has something to do with the marvell issue. If you have irreplaceable data that is not backup up on another place probably a good time to copy the most important stuff before doing the parity sync. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.