statecowboy

Here you go. someflix-unraid-diagnostics-20240411-1004.zip

Had a fairly new (I think it's maybe a year or so old) drive throw an error. Before I go about replacing it, I was hoping someone with more experience in drive errors could lend some advice. Thanks someflix-unraid-smart-20240411-0930.zip

Well, this is embarassing, but you guys are right. I totally forgot I recently replaced (upsized) my parity drive. This drive was the old parity drive, and I must just not have reformatted it. I just ticked the "format" button down below the main drive page. Sorry about that.

Was warned by fix common problems that I had a disk with errors. The main page shows "Unmountable: Unsupported or no file system". Reading the suggested fix, because the disk had data on it, it recommended to seek assistance here, so here I am. I do have a replacement disk on standby should that be the case, but figured I would seek guidance here before tossing this drive. someflix-unraid-diagnostics-20231215-0832.zip

I am not sure what caused it, but I no longer have this issue. I suspect that it was some sort of bug in the Ubiquiti software that was triggering something. I took my 48 port switch and my pi-hole machine out of my network stack trying to eliminate possible causes, and one of those seems to have resolved it.

Follow up post. It's happened again. My server has become unresponsive. I had to reboot it to get it back, but it came back up. I've attached my diagnostics. I noticed on the startup screen I get the following error: modprobe: ERROR: could not insert 'nct6775': No such device So, this does appear to be either a system or hardware error, not a plex thing. Any ideas? someflix-unraid-diagnostics-20220628-2143.zip

I'm going to apologize in advanced for this being an obscure request for help. I'm really not sure where to start. A couple of weeks ago my plex server was unresponsive. After logging into unraid's management interface and pulling up the docker logs I got an error message that said something along the lines of "preferences.xml" could not be read. When I opened my appdata folder to check that file it was completely blank. I did have a backup from appdata backup/restore and managed to replace the blank file. Plex started back up just fine and I went on my merry way, for about a day until this happened again a couple of more times. During these instances I also noticed that my unraid machine was not reachable at times, almost like there was a network problem. A reboot usually corrects this issue. I've tried a few things. At first, I thought somehow my pihole instance (running separately on an ubuntu server) was screwed up and I was having DNS issues. So, I did a fresh install of pihole. Things seemed to be working great until last night when my plex server was unresponsive again. I also could not log into my unraid machine. One thing I tried differently last night was I bypassed one of my network switches thinking possibly it could be going bad (I run a unifi network and my server was going through a USW-48 switch). We will see if that somehow corrects the issue. I guess my main question here is, does anyone have any idea what could be causing my "preferences.xml" to wipe itself like this? It seems somehow connected to network issues either with my server or with my netowrk hardware or software. I suppose these all could be symptoms of failing hardware, but I'm unsure of what steps I can take to diagnose that.

I have been getting the following alerts out of the blue for the last month+, seemingly every half hour: "Communications restored with UPS unraid". I have seen this before, but very sporadically. Now, it seems as if something has changed and it wont stop. That said, I've read some other forum posts and the consensus was to replace the batteries. So, that's what I just did. I have a Cyberpower 1500AVRLCD. However, about 30 minutes after replacing the batteries I just got the same alert again. Any ideas? I recently updated to 6.9.0-beta35 for what it's worth.

If you don't mind me asking what do you use for the setup? I have, for better or worse, invested in unifi. I ran pfsense for a while and it was great, but I didn't feel like diving in deep when wanting to do simple things. Unifi is just a bit easier to manage for me. So, I currently have the UDM-Pro, the US-24-250W, and the item I am considering is the US-16-XG. I have my server and network eqpt in my basement, so the 16 port XG would go down there. I have an outdoor rated CAT6 running up to my office where I would need the other end of the 10gb connection, so would probably need some sort of switch there as well (I feed an AP off of that line in addition to my office workstation).

Hi all. I was curious to hear from those of you who have a 10gb connection on your network for use with unraid. I understand ultimately that any reads from the array are limited by disk speed (unless of course you’re reading from an unassigned SSD or have all SSDs in your array). It looks like writes to the array can benefit to a certain extent on an SSD cache drive. That said, Ive talked myself out of investing in 10gb gear a few times now. I don’t think there’s a huge gain in performance to be had for my use case (would love to increase speeds between my server and my office pc for editing photos and general file swaps). All of that said, for those of you who have jumped in to 10gb, have you noticed a significant difference in things? Any real world examples of what you’ve noticed that you wouldn’t mind sharing? I of course don’t NEED to go 10gb, but part of me wants to tinker/play. I also don’t want to invest in gear and have a lot of remorse because it didn’t make much of a difference.

Attached. And yes, my dockers live on an unassigned device set to R/W slave mode.

Hi everyone. Was wondering if someone could do a sanity check on my config. I've had some weird issues with my docker image filling up, and I believe this only happens when I leave krusader running for a long time. Probably worth noting, every time I start krusader I get a notification that I can't save bookmarks to the home/nobody location. I am wondering if this is resulting in tmp files being written elsewhere within the docker image.

Circling back to this. I suspect my krusader docker is the culprit, although I'm not sure what I may have set up incorrectly. I usually don't leave it running unless I need it, but the few times I've had docker image size warning (I think) have come when it's been left on. Every time I start the docker I have to go through set up because it tells me I can't save bookmarks to the hom/nobody profile. I've always ignored that, but think it may have something to do with things. If it can't write to that location I suppose it's possible it's creating a temp folder somewhere in my docker image? I've pasted my config below in case anyone can spot something I've done wrong. I may post over on the binhex krusader help thread as well.

Thanks for responding. Booting into safe mode works. I at least can access the gui. Should I now try to edit the disk.cfg you mentioned above? Or, since I can access the UI now should I disable array autostart and update plugins? (I did have some US set to start on array start) EDIT - I turned off autostart on the array (settings->disk settings->enable auto start->"no"). I succesfully booted! I went and updated the US plugin as well. Do you think I'm good now? Or is there something else I need to check out? EDIT 2 - I think I found the culprit. The rclone mount script was hung for some reason (showing as still running). I disabled those scripts for now so I can debug. Pretty sure that was the issue. That script must have been hung and wouldnt allow the gui to start. Thanks so much, Squid!

Hi guys, I think I've exhausted everything I can think of at this point. I even tried putting an older version of my network.cfg back into the boot folder and rebooting. Same symptoms. I can SSH into the machine just fine, but the OS GUI is unreachable and my dockers and shares dont appear to be initializing.

I've left the server on even though I can't access it. My array doesn't seem to have started either. That said, I just got this error sent to my inbox over night: error: stat of /var/log/nginx/*.log failed: No such file or directory EDIT - I should probably add I deleted all files out of my nextcloud directory using krusader. I am confident I did not delete anything out of the wrong directory or something like that, but if there were links to files on my boot drive from nextcloud somehow those would have been wiped. I dont suspect that's the case, but just trying to provide as much info as possible. In addition, I did recently change how I was integrating pihole. I switched from a hardware pi to docker pihole. That said, the server survived the reboot into gui mode yesterday so I don't know if that's what's going on. EDIT 2 - As I'm scouring the web looking for similar issues it may be worth mentioning I was using my boards built in remote console when booting to gui mode (something I have not done in the past - I just added the remote module to my board not too long ago). I'm curious if this could have caused the server to boot into legacy vs UEFI (I dont know much about this topic), but after seeing some issues with this in the past I'm going to pull that string. EDIT 3 - I have tried renaming the "EFI-" folder in boot to "EFI" but this did not work. I have also deleted the "network.cfg" file from boot, but this also did not work (and I also noticed that after reboot unraid did not regenerate that file, which I found odd). I have also rebooted my router (UDM-Pro) because the software on it is, well, not the stablest. That also did not do anything. Edit 4 - Tried deleting contents of SSH and SSL config folders thinking maybe it there was an issue there. No go.

Problem - Can't access webUI, either from my LAN or even booting in GUI mode. I can SSH into the machine. Everything had been running fine previous to this. I was making some rclone changes tonight, so booted to gui mode to grab certs for google drive, and when I rebooted back to headless I could not access the web interface. I also removed nextcloud and deleted my nextcloud share off my array. Those are the only things I did. It looks like unraid is grabbing an ip from my dhcp server, but I simply can't access the web interface. I did notice this line during bootup "RTNETLINK answers: Network is down". Any ideas? someflix-unraid-diagnostics-20200227-0046.zip

So, increased from 30 GB to 55 GB. Got another notification today that docker image was 75+% full. Obviously something is going on. I'll dig back into my dockers and make sure there's nothing temporarily writing to the image.

"docker rmi" requires at least 1 argument. See 'docker rmi --help'. Usage: docker rmi [OPTIONS] IMAGE [IMAGE...] Remove one or more images

I did as you suggested above. Things have been warning/error-free for a couple of weeks now, but last night they started popping up. Think I should increase the size of docker.img? Screenshot of current state of docker containers below and updated diag. attached. someflix-unraid-diagnostics-20200222-0918.zip

Here you go. Turns out I did have geolite2 installed to the right path (config which is mapped to my unassigned device storing dockers), so that's definitely not the issue.

I'm not sure what's going on, but I just started getting docker image size warnings/errors last night. I've checked the container sizes on my docker home page and nothing seems to be out of the ordinary. I am going through all my dockers to make sure I did not pass an invalid parameter that's filling up my docker image. As a side, I suspect the issue may be with tautulli. They changed the way the geolite2 api works and I had to sign up for an account with the developer and reinstall the geoip app (I did this yesterday). I suspect that this may be causing my issues and is what initially filled up my docker image. That said, I uninstalled geolite and the tautulli container size is not very large, so I'm not sure if that's it or not. Diagnostics attached. EDIT - It has returned back to normal a couple of times now, but keeps fluctuating and throwing warnings. Thanks. someflix-unraid-diagnostics-20200209-1045.zip

For the OP, if you didn't already look it up, CPUBenchmark has score of 11,990 for your i7-7700K. So, given the "rough guides/recommendations/etc." (and opinions of those) as stated above you should just try it. Your GPU can't transcode H265, so you'd be left only with your CPU. It would seem on the surface that you'd really be pushing things with that config and 4K transcodes. Especially when you factor in the load for whatever else you'd want to run on this box. If I had that hardware laying around, would it be worth it for me to build out an unraid box? Absolutely. You can do lots with that. Would I build it out as a 4K transcoding machine? No.

Fair enough. Here's a good read on the subject (for the OP ). https://forums.plex.tv/t/plex-4k-transcoding-and-you-aka-the-rules-of-4k-a-faq/378203

I think you will find a lot, if not most, folks will tell you not to transcode 4K. You will want to direct stream. Just make sure your client (or tv) plex app plays well with your plex server.