markc99 Posted January 23, 2023 Share Posted January 23, 2023 I'm having errors on my parity drive, and it keeps going disabled. I've tried resetting it, and running through another parity check, and its not helping... I have at least two data drives which aren't being used yet, and I thought I would be able to shrink the array to pull a data drive out to be able to replace the parity drive, but I can't figure out how to do it. I've googled around, and not able to find the right answer. Could someone point me in the right direction? I'm running Unraid 6.11.5. Mark Quote Link to comment
markc99 Posted January 24, 2023 Author Share Posted January 24, 2023 So, I was able to build a new config, remove a data drive, and replace the erroring Parity disk with it. It got probably 8 hours through a rebuild, and now the NEW parity drive is doing the same thing. The new data drive was already pre-cleared, and only has a month or two of service so far. I'm perplexed as to what to do next. Quote Link to comment
DanW Posted January 24, 2023 Share Posted January 24, 2023 Have you changed the SATA data cable? Could be a bad sata cable or something wrong with the SATA port on the motherboard. Especially if a new drive dropped in its place does the same thing, it's unlikely you have two bad drives in a row. Quote Link to comment
JorgeB Posted January 24, 2023 Share Posted January 24, 2023 Please post the diagnostics. Quote Link to comment
markc99 Posted January 24, 2023 Author Share Posted January 24, 2023 (edited) 6 hours ago, DanW said: Have you changed the SATA data cable? Could be a bad sata cable or something wrong with the SATA port on the motherboard. Especially if a new drive dropped in its place does the same thing, it's unlikely you have two bad drives in a row. I thought the same thing, but when I did the disk swap in the GUI, I didn't open the case. The original "bad" disk is still connected to the same cable it was before, and the new "bad" disk is on the cable which wasn't having problems before... The server currently has my parity disk disabled and is running a "Read-Check". Based on the percentage done, it's probably going to take another day or two. Edited January 24, 2023 by markc99 Added more info. Quote Link to comment
markc99 Posted January 24, 2023 Author Share Posted January 24, 2023 Diag uploaded. tower-diagnostics-20230124-0849.zip Quote Link to comment
JorgeB Posted January 24, 2023 Share Posted January 24, 2023 Parity disk dropped offline, check/replace cables and post new diags. Quote Link to comment
markc99 Posted January 24, 2023 Author Share Posted January 24, 2023 1 hour ago, JorgeB said: Parity disk dropped offline, check/replace cables and post new diags. New diag attached. This is after I shutdown the server, reseated all the cables and powered it back on. Parity drive is still offline. tower-diagnostics-20230124-1044.zip Quote Link to comment
trurl Posted January 24, 2023 Share Posted January 24, 2023 3 minutes ago, markc99 said: Parity drive is still offline. Not offline, disabled, and will be until rebuilt. Start rebuild and if it continues to have errors you still need to fix the connections. Quote Link to comment
JorgeB Posted January 24, 2023 Share Posted January 24, 2023 It's still disabled as expected, but it's online now, SMART looks OK, you can try again, suggest replacing or swapping cables with another disk to rule that out. Quote Link to comment
markc99 Posted January 24, 2023 Author Share Posted January 24, 2023 Running a "read-check" and now I'm chasing errors on another disk. I guess the next step is switching cables around, but it's so odd that it's been multiple disks in the system, and as of now, I haven't changed any of the cables, just reseated them. Quote Link to comment
trurl Posted January 24, 2023 Share Posted January 24, 2023 1 hour ago, markc99 said: Running a "read-check" No point in letting that continue. Fix the problems so you can rebuild parity Quote Link to comment
markc99 Posted January 24, 2023 Author Share Posted January 24, 2023 22 minutes ago, trurl said: No point in letting that continue. Fix the problems so you can rebuild parity Understood. I stopped everything, powered the server off, checked the cables again, and the parity is rebuilding. Finger crossed... Only 1 day and 19 hours to get through without issue. Quote Link to comment
trurl Posted January 24, 2023 Share Posted January 24, 2023 Just now, markc99 said: Finger crossed... Are you seeing Errors on any disk now? Quote Link to comment
markc99 Posted January 24, 2023 Author Share Posted January 24, 2023 2 minutes ago, trurl said: Are you seeing Errors on any disk now? So far, so good. Quote Link to comment
markc99 Posted January 26, 2023 Author Share Posted January 26, 2023 (edited) Well, I thought I was in the clear, but I have errors on two disks again. One is brand new, and the other is a bit older. I'm thinking the first step may be to replace the SATA breakout cable. I haven't used any SATA connections on my motherboard however, so maybe I switch them to there to make sure it's the cable? Thoughts? Edited January 26, 2023 by markc99 Added more info. Quote Link to comment
trurl Posted January 26, 2023 Share Posted January 26, 2023 post new diagnostics Quote Link to comment
markc99 Posted January 26, 2023 Author Share Posted January 26, 2023 Attached. tower-diagnostics-20230126-1436.zip Quote Link to comment
trurl Posted January 26, 2023 Share Posted January 26, 2023 Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [crit] 4264#4264: ngx_slab_alloc() failed: no memory Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [error] 4264#4264: shpool alloc failed Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [error] 4264#4264: nchan: Out of shared memory while allocating message of size 10638. Increase nchan_max_reserved_memory. Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [error] 4264#4264: *333417 nchan: error publishing message (HTTP status code 500), client: unix:, server: , request: "POST /pub/devices?buffer_length=1 HTTP/1.1", host: "localhost" Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [error] 4264#4264: MEMSTORE:00: can't create shared message for channel /devices Don't know what these are all about but they are filling your syslog Quote Link to comment
markc99 Posted January 26, 2023 Author Share Posted January 26, 2023 21 minutes ago, trurl said: Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [crit] 4264#4264: ngx_slab_alloc() failed: no memory Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [error] 4264#4264: shpool alloc failed Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [error] 4264#4264: nchan: Out of shared memory while allocating message of size 10638. Increase nchan_max_reserved_memory. Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [error] 4264#4264: *333417 nchan: error publishing message (HTTP status code 500), client: unix:, server: , request: "POST /pub/devices?buffer_length=1 HTTP/1.1", host: "localhost" Jan 24 22:54:21 Tower nginx: 2023/01/24 22:54:21 [error] 4264#4264: MEMSTORE:00: can't create shared message for channel /devices Don't know what these are all about but they are filling your syslog I'm not sure what they are either... I did power off the server and put the two disks experiencing errors on SATA cables attached to the mobo, instead of the SATA card, and it's running a parity check. The interesting thing is it's running WAY faster. It's been running a half an hour, and it's already 10% complete, with 5 or so hours left... The last one that ended up erroring out, was estimated to take almost 2 days. My gut is thinking one of my breakout cables may be going bad. Quote Link to comment
trurl Posted January 26, 2023 Share Posted January 26, 2023 1 hour ago, markc99 said: The last one that ended up erroring out, was estimated to take almost 2 days 4TB should only take 8-12 hours if things are working right. 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.