January 30, 20188 yr Dear All, I really hope you can help. Last week I was unable to connect to my server, and so restarted it - no luck! So I then took the logical approach, checked and double-checked my cabling. I then pull all my drives except for my parity drive and fired up my server. Connected well enough, or so I thought. My issues are: 1. I now have 2 parity drive entries instead of 1. One entry has a disk assigned (to be expected), the other entry does not 2. Unable to perform a parity check after re-inserting my other drives, any attempts to result in the server just freezing (or at least it looks that way) Currently at work, so unable to provide anything useful like screenshots or logs etc., but has anyone else experienced the same thing or something similar? If so, how did you resolve? Really do not want to start from scratch, as I have quite a bit of data on my server. Thanks in advance.
January 30, 20188 yr When you get home I'd run a Tools->Diagnostics before it freezes and post the file here. For #1 it just sounds like the array isn't started...
January 30, 20188 yr Author It freezes almost instantly, so have copied up the diagnostic without attempting the parity check (dumb thing to do?). Also copied up a screenshot of my drives. tower-diagnostics-20180130-2041.zip
January 30, 20188 yr Author activated maintenance mode, then ran the diagnostic tower-diagnostics-20180130-2048.zip
January 30, 20188 yr Community Expert unRAID v6 allows 2 parity disks. You only have one assigned. When the array is started and you don't have parity2 assigned, it isn't listed, but when the array is stopped it is shown with a dropdown to allow you to assign a disk to that slot. There is no need to be in maintenance mode. The basic problem is it is missing disk3. Why is disk3 missing?
January 30, 20188 yr Community Expert 2 hours ago, troymeredith said: It freezes almost instantly Most likely filesystem corruption, maybe the emulated disk, you should still be able to get the diagnostics using the console/SSH by typing diagnostics But like trurl said 2 hours ago, trurl said: Why is disk3 missing?
January 31, 20188 yr Author 7 hours ago, johnnie.black said: Most likely filesystem corruption, maybe the emulated disk, you should still be able to get the diagnostics using the console/SSH by typing diagnostics But like trurl said disk 3 missing because i haven't inserted it, as i found the risk of freezing greatly reduced without it inserted. will re-insert and try to run the diagnostic again
February 4, 20188 yr Author ok, after plenty of faffing, managed to start the array with the parity and 3 other drives. daren't restart the server, in case things freeze again! no docker as the cache isn't being picked picked - connection issue? diagnostics report attached tower-diagnostics-20180204-0951.zip
February 4, 20188 yr Community Expert The server isn't fully starting, no disks are mounted, I also see ATA timeouts on several disks, something weird going on there, you may try to redo your flashdrive, back it up first, then use the latest release and restore the config folder from current flash, then boot and upload new diags.
February 4, 20188 yr Author 6 hours ago, johnnie.black said: The server isn't fully starting, no disks are mounted, I also see ATA timeouts on several disks, something weird going on there, you may try to redo your flashdrive, back it up first, then use the latest release and restore the config folder from current flash, then boot and upload new diags. updated flash drive with latest release, fired up server, disk light on server is solid, disks being mounted for 45mins now. diag attached tower-diagnostics-20180204-1718.zip
February 4, 20188 yr Community Expert Parity failed: Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369288 Feb 4 16:47:38 Tower kernel: ata1: EH complete Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369296 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369304 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369312 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369320 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369328 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369336 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369344 197 Current_Pending_Sector -O--C- 100 100 000 - 8 198 Offline_Uncorrectable ----C- 100 100 000 - 8
February 4, 20188 yr Author 30 minutes ago, johnnie.black said: Parity failed: Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369288 Feb 4 16:47:38 Tower kernel: ata1: EH complete Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369296 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369304 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369312 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369320 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369328 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369336 Feb 4 16:47:38 Tower kernel: md: disk0 read error, sector=2930369344 197 Current_Pending_Sector -O--C- 100 100 000 - 8 198 Offline_Uncorrectable ----C- 100 100 000 - 8 ok, complete novice at this, so how do i proceed?
February 4, 20188 yr Community Expert You need to replace it, but since you currently have disk3 disable you need to figure out what to do about it, IIRC you disabled it on purpose, assuming nothing was written to the emulated disk you could do a new config without parity, see if everything is working correctly and if yes add a new parity disk later and then resync parity.
February 4, 20188 yr Author 2 minutes ago, johnnie.black said: You need to replace it, but since you currently have disk3 disable you need to figure out what to do about it, IIRC you disabled it on purpose, assuming nothing was written to the emulated disk you could do a new config without parity, see if everything is working correctly and if yes add a new parity disk later and then resync parity. ok, I can pop in a replacement for the parity now, disk3 has been re-inserted when i updated the flash - no idea if anything has been written to it. so should i still replace the parity?
February 4, 20188 yr Community Expert 7 minutes ago, troymeredith said: disk3 has been re-inserted when i updated the flash But it's still disabled, once a disk is disabled it's needs to be rebuilt or reset with a new config. 8 minutes ago, troymeredith said: no idea if anything has been written to it. so should i still replace the parity? Not much you can do about it now, since parity is failing it can no longer correctly emulate disk3.
February 4, 20188 yr Author 3 minutes ago, johnnie.black said: But it's still disabled, once a disk is disabled it's needs to be rebuilt or reset with a new config. Not much you can do about it now, since parity is failing it can no longer correctly emulate disk3. swap out disk 3 and parity, or is that too much?
February 4, 20188 yr Community Expert Just now, troymeredith said: swap out disk 3 and parity, or is that too much? Don't understand the question. I would do a new config (tools -> new config) and then assign only the 3 data disks, start the array and if all works correctly add a new parity disk when you get a spare, if not post new diags, but note that like I said you'll lose any data written to disk3 since you disabled it.
February 4, 20188 yr Author 3 minutes ago, johnnie.black said: Don't understand the question. I would do a new config (tools -> new config) and then assign only the 3 data disks, start the array and if all works correctly add a new parity disk when you get a spare, if not post new diags, but note that like I said you'll lose any data written to disk3 since you disabled it. i've got no problem losing any data from disk3, nothing i can't get back anyway - will create a new config...
Archived
This topic is now archived and is closed to further replies.