mau471

I am also having the issue with with the docker.img getting maxed out on size with the linuxserver plex container. I haven't changed anything other than just doing the standard update container option in the unraid gui. The problem for me started around the time Plex enabled the new skip intro feature. The tmp folder is what is eating up all the space, in about 2-3 days it ballooned to 27 GB. Docker run command is as follows although nothing has changed on my side in months in regards to config but the problem started about 2 weeks ago. "docker run -d --name='Plex' --net='bond0' --ip='10.0.1.230' -e TZ="America/New_York" -e HOST_OS="Unraid" -e 'NVIDIA_DRIVER_CAPABILITIES'='all' -e 'NVIDIA_VISIBLE_DEVICES'='GPU-163441c5-a503-1b9f-b353-1b36622ff622' -e 'PUID'='99' -e 'PGID'='100' -e 'CHANGE_CONFIG_DIR_OWNERSHIP'='true' -e 'VERSION'='latest' -v '/mnt/user/Containers/Plex/Data/':'/config':'rw' -v '/dev/shm/':'/transcode':'rw' -v '/mnt/user/Plex/':'/data':'rw' --runtime=nvidia --log-opt max-size=50m --log-opt max-file=1 'linuxserver/plex:latest'"

I love the overall simple to use interface One (well two) things I would to see added is the ability for VMs to preserve XML edits if you go back into gui mode and a gui way to clone all or parts of an array to another unraid machine.

Just finally had a chance to update and both issues are resolved.

I am running the default 2x scaling for macOS and on that particular monitor an effective resolution of 2048x1152. Your slightly smaller and smaller again screen shots are essentially what I am getting, significant blank space where an additional column of docker icons could go on the right. I assume the reason I am not getting the same full screen results as you is that I am running the Gray theme where the navigation icons are down the lefthand side instead of across the top so it squeezes the view a bit side to side.

Just to update, if I load the dashboard directly on my laptop the page displays fine, but as soon as I connect to the external display is when the issue occurs. Desktop monitor is an LG Ultrafine 4k display. I am able to duplicate this on both Safari and Chrome.

If I go full screen the columns don't adjust in size, it just adds more dead space to the outside edges of the page. The columns used to dynamically resize (I am 99% sure) with the window but at least for me on 6.8 the only change that happens is the number of columns, not any changes to column width or layout within a particular section.

Did you ever figure anything out with this? I am having the same issue with some (but not all) docker images and can't figure it out.

I went back and forth on a ton of options and originally settled on the Node 804 but that ended up not working out as I couldn't get proper cooling with my particular setup. After agonizing a lot more because I needed a case I could fit between rack rails and < 8u height lay on its side but didn't have the rack depth for a proper rackmount case. I ended up with the Define R6, I was originally going to go with the R5 with a 2x5.25 to 3x3.5 cage since I wanted to be able to hold 10 3.5" drives but the R6 nets the same number of hard drives (technically 2 more 2.5 SSDs) and still gives you a single 5.25 bay to do something (or nothing) with. The extra drive trays are available on Amazon for $15 for a two pack. If the hot swap capabilities aren't a big deal and it's within your budget I would recommend the R6 over the R5 because of the greater flexibility.

I just went through a migration myself, I don't think its worth the hassle to use green drives as cache, unless you are creating a number of temp files that would go away before mover runs it's not really saving a ton of wear and tear on the array drives and certainly won't be any faster than just writing to the array. As for moving files later, there is a function in the plugin "Fix common problems" to move the files I did the same thing as I did not initially setup the cache when I spun up the system.

So at least for now I got around it by deleting the icons off of the flash drive and rebooting the system. It gets the job done as I don't need to change often and mostly just needed to fix me copying the wrong url initially but it seems a little extreme to have to reboot the the entire system to fix a docker icon issue.

I am having an issue trying to update the icon that shows for the docker container. This isn't specific to a particular container, it is happening with everything I try. The issue crops up if I set an icon and then try and change it. It seems to stick with whatever was set first and won't update from there. It will update if I don't assign an icon but then go back and do so, but any further updates don't work. I am on 6.7.0-rc7 but the issue was happening on 6.7.0-rc6 as well. Not sure if its specific to the beta versions as I started fresh with 6.7.0-rc5. Steps to reproduce: Create docker container Assign an icon url Deploy container Edit container and change icon url Any help in being able to change the icons would be much appreciated.

Ah perfect, I didn't realize the cache pool was raid 1 by default with multiple drives, I missed that. Some of the documentation references the pool being raid 0 so thats where I got confused. All 4 drives are the same capacity/model/etc. so I think the standard cache pool with all 4 drives would be sufficient for what I am trying to accomplish as long as I set the specific share to be cache only. Thanks for all of the responses.

I am currently testing out unraid on a spare machine to see if it would be a good fit and I think I am mostly ready to pull the trigger but I have a question around the best way to handle a few SSDs I would want to migrate to the build. I can't fully test the scenarios I am looking for since the machine I am testing on only has 3 drives total. I have 4x samsung pro nvme ssd's on a bifurcated PCI express card, I want to take 2 and use them as cache drives and the other two and use them as a separate mirrored volume to keep things I always want to live on SSDs for fast access there and the drives mirrored. At least as I understand it the cache pool is striped and not mirrored so theres a chance of data loss for the persistent ssd data if one of the drives were to fail so I am looking for something a little more stable locally in the machine that can allow me to recover pretty quickly (in addition to my offsite backup which is definitely more of a cold storage type situation.) Am I thinking about this right or is there a better solution that will accomplish what I want? I don't necessarily have a problem with the data permanently in the cache pool and saw the option to store the share only on cache but that doesn't give me the mirroring I am looking for. I don't think unassigned devices would necessarily be the right fit either as that doesn't seem to allow mirroring of multiple drives together.