Indi

I just installed this app but not seeing any usage data I have 2023.11.09 installed Unraid 6.12.4 1660 Super GPU Tried clearing browser cache Unchecked, applied, rechecked, applied the GPU

I am a little confused though. I am cleaning up a few other shares and when "Destructive" mode is set to "No" in the plugin I can't delete a share that is in the ZFS cache. When I set this to "Yes" in the plugin, I can delete the share. I am deleting the share via Unraid, by clicking the share and checking Delete, not through the plugin. I just replicated this again so something has to be wrong if the plugin doesn't prevent a dataset from being deleted. I assume that the plugin is somehow holding onto that share so that Unraid sees it as being in use. But changing that setting fixes it, for whatever reason.

When you mentioned this I immediately thought of the ZFS Master plugin, which was the culprit. I went into that plugin settings and "Destructive Mode" was set to "No". I changed this to "Yes" and then went back into the share in unraid, and was able to delete it. It seems the plugin prevents deletion even if the share is empty.

I gave this a try but it's still saying it's busy

Here is the output:

foxdie-diagnostics-20231024-1523.zip

I moved around some shares and /anime is now empty. I checked for hidden folders, nothing is inside it. I checked the plugins file activity and open files and nothing is pointing to that share. No docker running is pointing to that share, however I think it's an issue with ZFS. It looks like ZFS is still using the share and therefore, I cannot delete it. How do I fix this? I have other shares I need to remove as well, I assume i will have the same problem. Oct 24 14:13:07 FoxDie shfs: /usr/sbin/zfs unmount 'cache_system/anime' Oct 24 14:13:07 FoxDie shfs: /usr/sbin/zfs destroy 'cache_system/anime' Oct 24 14:13:07 FoxDie root: cannot destroy 'cache_system/anime': dataset is busy Oct 24 14:13:07 FoxDie shfs: retval: 1 attempting 'destroy' Oct 24 14:13:07 FoxDie shfs: /usr/sbin/zfs mount 'cache_system/anime'

I am looking at restructuring my server to support hardlinks and allow for permaseeding, but it's going to be a long, long process to do so and wanted to clarify a few things to ensure I don't do this and not achieve what I thought I understood lol. I've seen and read through the trash guide but I still want a more technical answer, which is why I am posting here. Currently in order to seed, two copies of the same file must exist on my system. For example: TorrentA - 10GB gets downloaded to /mnt/user/downloads/torrents Once complete, Sonarr/Radarr imports/copies that into /mnt/user/anime for example. Now, there are two files, for a total of 20GB. The one in /downloads/torrents will continue to seed, while the /mnt/user/anime is now in my media library. If I move to a single share of /data/media, the above would instead be: TorrentA - 10GB /mnt/user/data/media/downloads/torrents /mnt/user/data/media/anime For a total file size of 10GB because of hardlinks? The more technical answer I am looking for is, how is this achieved and where is the "master" file actually stored? Will using a cache and mover break hardlinks? My current setup is below, I plan on keeping mover tuner the same but just structure everything into a single share. Though, I just don't understand how unraid will see a single file, with two different names. For example when you download TorrentA, that might be the name but then Radarr/Sonarr importing it will reename it to TorrentA - 2023 - 1080p - DD - H264 etc. After 14 days, it will move the file from cache to array, the paths will stay the same but instead it will seed from the array? Though, which is the master file and how is that determined? If anyone has a good video that explains this or anything or the sort, I've watched ibracorps, spaceinvader and a ton of others but none really go into it on a technical level, which is what I am looking for. Current setup does not use hardlinks and is the following: 100+ TB HDD Array cache_system - 3 2TB Samsung 980 Pro in ZFS pool (total of 4GB usable space) spare_cache - 4TB 870 Evo cache_system is cache=yes for all shares below, as well as appdata and system shares. Mover tuner is set to move files from the below shares that are older than 14 days to the array. Shares: /movies /music /anime /animemovies /tv /cartoons /downloads The reason for the above is I wanted to have each media share separate and not housed under a central directory. This allowed me for more control, such as keeping specific media items to specific drives etc.

I think this might be a dead project. I have this one installed currently and it works, but I think smartphonelover abandoned maintaining it or something I am not sure

Gotcha, thank you! I will see if I can find more resources on this. I thought I had an understanding of how it worked based off YT videos I've seen and other forum posts, where if they share a root path, hardlinks will work (which is where my idea came in). I appreciate the response, seems like I have a lot of research to do

So the *arr's viewpoint, they are two root folders. For example Sonarr Root Folders /tv /anime Downloads has no remote path mapping configured Then SabNZBD Completed Downloads Folder - /downloads I notice now that /downloads for Sonarr is /mnt/cache/downloads while Sab is /mnt/user/downloads, could this be my issue? Appreciate the reply!

Thank you! I guess I was misunderstanding the dashboard, in that during an import operation from Radarr or Sonarr etc, the cache shows Reads of 450Mbps and Write 450Mbps as if it was reading from /mnt/cache/downloads and then writing back to itself to /mnt/cache/TV If that isn't the case, I might be okay. My SSD has a 2400TBW warranty, so my only thought was to help prevent unnecessary writes, if that were indeed happening

Please don't scald me, but I am looking to see if anyone has advice if I may be thinking unclear about solving some issues. Quick recap - I had my media all under data/media/movies etc, but after my library grew larger I had issues with: 1. Managing split level was a pain with anime/movies/tv as they all needed a different split level. 2. I wanted all of the same shows, the same anime to be on a single drive. I didn't use split level at all before, so I had .nfo files and random episodes, all split across 3-4 drives and it is inefficient. 3. I like having my array organized. Disks 1-4 are for TV, Disk 4 is for Anime, 5-7 are for Movies. Now, I have 3 shares: anime/animefolder/seasonfolder/episodes tv/showfolder/seasonfolder/episodes movies/moviefolder/movies Everything is working fine, but I am now noticing the benefits of hardlinks and atomic moves. I have a 4TB cache drive, where mover tuner is set to move files to the array after 15 days so all new files are accessed from the cache. However, when something is downloaded, it's incredibly inefficient. The process is now: Download (Write Process to downloads share) Unpack Import (Read Process from downloads share) Import (Write Process to tv share) Mover after 15 days (Read from cache, write to array) So for a download, it will read/write to itself multiple times and a 50GB file for example, can write 150+GB to the same drive. Now that I've sorta explained my setup, to solve this I am thinking of doing the following. Change shares to be something like anime/media/animefolders/animeseasons/episodes Then under "media" create another directory for downloads. Next, setup 2 instances of Sonarr, 1 instance of Radarr. Sonarr1 for TV, Sonarr2 for Anime, Radarr for Movies. Since SabZNBD looks at a parent download directory, categories wouldn't work I don't think. But maybe I can map each share in Sab to point to the correct media, if not, I will setup 3 Sab containers to point to each arr instance. With the above, I should be able to restore hardlinks for each appropriate share, right? Unless I am missing something. If anyone might have a different way of doing this, I'd love to know. I understand now the performance benefits of a single share, but my OCD will no longer allow that and I am much happier being able to properly manage media across disks and maintain split level for each media type.

I haven't rebooted yet, but I've attached my diagnostics and I'd like to figure out what went wrong. All my shared are gone. I hope they come back after a reboot...using 6.11.5 foxdie-diagnostics-20230208-2346.zip

It seems the general topic for this is looking for ways to reduce power consumption but I am looking at doing the opposite. Does anyone know how to force a 1660 Super to stay in P0? Here is some background: When I use Tdarr to transcode files to save space, it will run 4-5 transcodes at a time but stay in P2. When it decides to go into P0, it is WORLDS faster, my understanding is that there is faster memory clock speeds during this compute task. Second, in Emby if I enable "transcode throttling", it seems to have a difficult time switching back and forth for some reason. It will sometimes initiate P0 and transcode in seconds, but sometimes it decides "I'm just gonna use P2" and doesn't effectively transcode fast enough. To avoid this, I just want to force it into P0, if at all possible. I just don't know where, or how I can make this change.

Did you ever find a solution to this? I want to just force my GPU into P0 at all times if possible, doesn't really matter to me.

It just wasn't clear what @SSD said, in that "if the split level is set correctly". I don't want to assume what he meant by that, in terms of using it alongside "most free". IVe read that linked guide so many times but I'm just an idiot. So I thought I'd ask

I know this reply is 5 years old, but does this still apply in the current version of unraid? I am looking to optimize my server, to minimize disk spin ups. So I am creating a tv share (even though I will lose hard-linking I guess), then the share will be tv/series/season/ If I use a split level of 1, with Most Free, this will allow new shows to be put on the disk with the most space, but keep continuing series on the same disk?

I can't get this running, I've followed the advice of this entire thread but now I've gotten stuck with all 3 containers running, but the ES container is erroring with: Been at this for hours now, appreciate any suggestions.

Update: So the SATA cables from the motherboard worked, I guess I need to figure out why the SAS card won't see SSDs? I put in two SATA SSDs that are 1TB, to add to my cache pool but for some reason Unraid won't see them? I am not sure what the issue is, because I take those same drives and plug them into my Windows 10 machine and I can see/format/use them just fine. I have a SAS card I've been using, which works great. I took the cables that I had plugged into the two additional SSDs and swapped them for my system cache drive and it worked, so it can't be the sata cables I am using. What else could stop the system from seeing the drives? Right now I am going to try using one of the SATA cables coming direct from the motherboard, to see if maybe my SAS card doesn't like reading SSDs I've attached the diagnostics, if it helps (not sure it will if the drives aren't being seen) diagnostics-20221014-1141.zip

So that might be the issue...the container port shows 8080, but I can't edit this. Do you happen to know how I can modify that value?

This is what I have it set to currently, maybe I am missing something? The only ports I changed are where 8080 was referenced. I didn't touch anything else, that's all default. Also on Key 9: I have 8383 set for WEBUI_PORT

@wgstarks I am, my PC is 192.168.1.10, the server is 192.168.1.170 and I have the LAN_NETWORK the default as 192.168.1.0/24

I have been trying to get the binhex-qbittorrent docker running for the last few days but have failed. The logs show no errors, if I run curl ifconfig.io it's showing an IP from my VPN provider. The last message in the log is that qbittorrent process is listening on port 8383 (8080 is reserved for sabnzbd). I have changed this in the advanced settings, verified that it's set in the config file, navigated to it manually, it's showing correctly in Unraid for the port mappings, etc. No matter what I try, the WebUI will not pull up. I've tried in different browsers, incognito mode, etc. Any ideas?

New unRAID user here, testing out a trial trying to see if it's a good fit to replace my Windows Server. Issue is that when I run sabNZBD, Radarr and Sonarr (the only dockers I have installed from binhex), and Sab is downloading a file, once it gets to a repair or unpack, the entire array goes offline, I lose connection to the servers web gui for about 1-2 minutes where the connection is restored and I have to start the array back up. I've run into nothing but constant issues using unRAID and feeling a bit defeated giving the countless hours I feel like I've wasted just trying to get simple functionality to work. I tried using Krusader to search for any logs, which is another issue that I can only access that dockers Web GUI if I copy the URL and paste it into a incognito window. Rather than list all my issues, trying to tackle the ones that are keeping me from converting from a trial. I can't spend 8+ hours a day trying to get things to work in this OS, though I hope once things are "working" that I won't have to babysit this OS.