trurl

Post your docker run command for plex as explained at this very first link in the Docker FAQ: https://forums.unraid.net/topic/57181-docker-faq/?do=findComment&comment=564345 Not necessarily, and unless you are running the latest beta you can't have multiple pools anyway. If you installed cache before enabling dockers or VMs then things may be already setup by default. The easiest way to get us the information we need to look at this is to Go to Tools - Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread.

This is what I did when I did my conversion, long before there was any wiki for this, maybe even before this thread was started. Lots of ways to move files and maybe simpler is better or at least less prone to mistakes. Midnight Commander (mc from the command line) is easy to use and understand. Google it.

This is an old thread but nobody else is using it so you can have it now. There are several settings for each user share that control how Unraid chooses which disk to write. Sounds like you must have not used the defaults, which would have (eventually) began using other disks long before filling any. Go to Tools - Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread.

Probably almost everyone is running CA on lots of Unraid versions and not having these problems, so it must be something specific to your setup.

Missed that. @BenjaminMetzler don't install a lot of stuff you aren't using and may not even know what it is.

Do you have other browsers or tabs or mobile devices open to your server?

Yes

Cache is a good place for the shares used by dockers/VMs, (appdata, domains, system). If these are on the array performance will be impacted by slower parity updates, and array disks will be kept spinning since there will be open files. If anything, only use cache for this purpose. But really you have plenty of cache. You just need to figure out how to use it. We can help you work through all that.

This is wrong Any idea how it got that way? Should be drwxrwxrwx 1 nobody users 16 Sep 4 04:40 user/ I know there was similar thread in the not too distant past but having a little trouble searching for it. Are you trying to do this?

What size are these disks if not 3.5 or 2.5?

What do you get from the command line with this? ls -lah /mnt

Wrong, it's builtin since 6.8. You have to install a plugin to configure it.

@Danuel Sort of grasping at straws here, but are you sure there isn't something going on with your network, or possibly the browser you are using to access the webUI? Have you tried booting in GUI mode and using the webUI directly on the server?

I and many others are using Wireguard and have no problems with Community Applications. Really no idea what your problem might be since you replied to this old thread and didn't supply any information. Not even clear what "this issue" means to you. Maybe if you You can go directly to the correct support thread for any of your plugins by clicking on its Support link on the Plugins page.

You have a reply on that thread awaiting your reply.

Before the 6.9 betas people usually handled this with Unassigned Devices plugin, and that will still work for now though in some ways it is simpler and more flexible with multiple pools. I have a 2x500GB cache pool which I use for caching user share writes, and 1x256GB NVMe "fast" pool for my dockers/VMs (appdata, domains, system shares). Here is what those shares are normally used for: appdata - the "working storage" for each docker container, for example, the plex library (database). domains - the vdisks for VMs. You can put some of this on the array if you want, or you can just put the OS vdisk in domains and use the virtual network to access Unraid storage for other purposes. system - docker.img and libvirt.img, docker.img is the executable code for the containers, and libvirt.img is the XML for your VMs. dockers and VMs will also typically use other user shares on Unraid which may be on the array or cache (or cached then moved to the array). For example, your media for plex or downloads from whatever app would be on one or more user shares. Yes, upgrade just replaces the Unraid OS archives on flash. Those archives are unpacked fresh from flash into RAM at each boot and the OS runs completely in RAM (think of it as firmware except easier to work with). Your configuration (config folder on flash) is independent of that upgrade. Sometimes new features might be a reason to change some things about your configuration though, such as multiple pools.

Do you have any command line open anywhere? It was also having problems unmounting cache. In fact, your diagnostics showed cache and docker.img still mounted. I also noticed FTP running in syslog. Is there anything connected there?

Sounds like you've got a plan that should work. One thing you might reconsider Since these are already known to be good might as well just go ahead and put them in with the New Config. If you put them in during New Config no need for them to be clear since parity is being rebuilt anyway, and Unraid will let you format them after you start the array and begin parity sync. You can even format them during parity sync if you want.

Each data disk in the parity array is an independent filesystem, so no individual file can span disks. Unraid user shares allow folders to span disks, but files cannot. Why are you using such small disks anyway?

There is also this in syslog when you try to assign the new disk: Sep 4 16:24:30 NAS kernel: mdcmd (6): import 5 sde 64 7814026532 0 HGST_HUS728T8TALE6L4_VDKY234M Sep 4 16:24:30 NAS kernel: md: import disk5: lock_bdev error: -13 Sep 4 16:24:30 NAS kernel: md: import_slot: 5 missing I don't recall seeing that lock_bdev before but maybe I have just never had a reason to look for it

Looks like something is keeping docker from shutting down. Can you get to the Docker page?

Looks like you mean 2 SSD cache drives. You're not going to get any performance out of an SSD in the parity array because writes to the array are always going to be slower than the HDD parity disk. If you really want to treat that disk separately, you should take advantage of the multiple pools feature in the latest beta. appdata, domains, system shares are all on cache and set to stay on cache. That is they way you want them so your dockers / VMs don't have performance impacted by slower parity, and so dockers / VMs won't keep array disks spinning. And, none of your other shares have any files on cache. So, I would say you actually have all that just like it needs to be.

Some GPFs in syslog. Have you done memtest?

Done

Done. Here is the thread: Changed Status to Closed, and Locked. Can be reopened if it does appear to be a bug, of course. Changed Priority to Other