-
Posts
12 -
Joined
-
Last visited
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by SkippyAlpha
-
-
Whoops! Sorry about that, I've attached it now. Thank you for the tip on the LSI cards firmware, I'll definitely look at getting it updated asap. Also good call, parity disk 2 and those three drives all share a backplane... I'll be making sure all cables are tight once everything completes. The udma crc errors are all old I believe, at least the majority. Thanks again
-
Hello guys! Looking for maybe a bit of clarity on why this may have happened. Monthly Parity check kicked off last night @2:00 April 1st. When I checked this morning I noticed that 3 consecutive disks had equal numbered read errors, as well as sync errors, all at 128. I've looked through the syslog a bit but not 100% sure what I might be looking for. One thing that stood out was that my weekly appdata backup kicked off at the same time and happened to finish at almost the exact moment the read errors stopped. I've adjusted the scheduling for that now in case it caused a problem. Other than that, nothing glaring stood out. Also the 3 disks in question looked healthy in smart data. The 3 disks were also not dropped from the array, so it seems Unraid was able to recover.
Any thoughts on this? From memory, I don't recall ever having any sync/read errors before, so I was a bit surprised. Might some random file be corrupted because of this? I've attached a copy of the syslog, as well as the smart data for the 3 disks and a screenshot of my dashboard. The machine has been up for 29 days so there should be lots of historical data in there. Any help would be appreciated, thanks in advance! Parity check is still running currently btw, at 18% now.
elmserver-smart-20190401-0746.zip
elmserver-smart-20190401-0747 (1).zip
-
One more thing, the second parity drive is not just a copy of the first parity drive, it actually goes through another algorithm (that I believe is actually sort of cpu heavy) to calculate the second disk. This means you can lose any 2 drives, not just 1 data + 1 parity drive at once. It could be for example, 2 data drives dying, and you would be fine.
From the way you worded your question, just wanted to clear that up
Edit: I do agree with them though, for your number of disks I think just 1 parity is appropriate. I personally moved to 2 parities around the 12 disk mark I believe.
-
If you want to just rebuild a disk, simply stop the array>unassign the trouble disk>start array>stop array>reassign disk>start array and start rebuilding.
-
4 hours ago, johnnie.black said:
You can rebuild to a different disk.
Doh! I didn't even think of that! Oh well, I'll remember it for the future. I guess this post can be closed now
-
Aha! Pretty sure I've lost it all lol. I guess thats one of the beauties of Unraid, I ONLY lost one disks worth of media, and not all of it, movies and tv shows that I can get back. Oh well, live and learn I suppose.
EDIT: Plus Plex has a trashcan icon over all of the media that it cant find, so I know what I lost. Not the end of the world
-
Also through googling I'm seeing mentions of a couple commands "init-csum-tree" and "init-extent-tree" they are talked about like they're really dangerous. Any experience with these?
-
Hmm alright I may try changing the metadata to dup. I really only used btrfs because it seemed like the modern thing to use, but it just seems too fragile and finicky. Do you know if there is an easy way in unraid to make a copy of the disk so that I can continue trying things on the disk without worry? Like a copy of the disk at the bit level?
-
@itimpi Gotcha, do you think I should go ahead with trying "btrfs check --repair /dev/sdX1" on the disk? I still have it out of the array. Or is there anything else I should try first?
-
Here is this syslog. Also another question, why am I unable to rebuild the disk from parity? Is it because this is already the state that the parity remembers everything in? That would make sense as to why it still shows the error even when the disk is emulated I suppose
-
This is my first time posting here so please let me know if I've posted correctly!
On Saturday 10/20 I was informed by a friend that he was unable to access my Plex server. I went to check it out and realized that the server seemed locked up, couldn't access webui nor any of the dockers running on it (plex, radarr, sonarr, ombi etc.) So I (dumbly) did an unclean shutdown of the server. After it restarted everything came back up as normal and everything was working with the exception of disk 7 in my array... which said it was "Unmountable: No file system" I tried rebooting the server again but it was still like this. The array drives are all formatted as btrfs (will probably be switching them to xfs after this). I stopped the array, removed disk 7, and restarted the array, however it still says the same message even with no device in that slot. Also I cannot mount the disk in UD. I've tried all of the recovery options found in the post I linked below by johnnie.black, with the exception of the very last option (destructive recovery). I wanted to wait until I posted here before I tried that option. Also I found this on an arch linux forum https://bbs.archlinux.org/viewtopic.php?id=233724 and was wondering if this might be able to help me, also haven't tried this yet. I'm at work right now but once I can get home I can try to provide any necessary logs. I've also attached a screenshot of my array. Thanks!
SysLog Messages - RPCBIND
in General Support
Posted
Would like to add that I too am getting a massive amount of these messages after mounting an nfs unraid share from another of my servers