NLS

New problem is that there are things in /system/docker/docker/btrfs/subvolumes/ that I cannot move and mover also cannot move. My moves finished (with the skips mentioned above), mover "finished" also, so I started again to see if it will in the end move what MC couldn't. And although there are just like 2GB in cache left behind, it takes its time, not sure if it actually does anything. I'll try copy instead of move and we'll see. I am not sure how docker can recover from various possible errors. I'll probably also use backup to just backup those remaining things and see if I need to recover them in the end. --- I saw your new reply while writing this. I thought about using "previous apps" (I first need to keep my appdata or something, right?) but the problem is that system also holds some VM data IIRC... --- EDIT: Actually 4GB, 8000 folders and 149000 files remain in cache folder. (all inside subvolumes folder) Don't know what to expect...

Yeah I thought too late about that. You already had the solution. Mover was running and MC was also moving. And still is, but is close to the end. Few things remain, possibly things that somehow conflicted. We'll see. I guess I'll wait and see. I noticed a "supervisor.conf" file is reported many times as existing and is overwritten. I guess this is normally a link, not sure if it remains a link. Also some moves in system report "cannot overwrite directory /mnt/disk4/syst~g/systemd/user - invalid cross-device link (18)... which I skip... Actually do "skip all" so I don't know if it would say this for other files. Not sure what I will have in the end.

So, I am doing an upgrade to my server and among other things my SSD cache will be swapped for an M.2. So I did the "cache YES" thing and started mover. Mover though is very slow (maybe the safety checks it does), so I decided to "assist" it by using server's own local MC to make some moves myself, trying to not conflict with the current working folder of the mover. This more or less goes smooth, but I noticed a thing, even before I started the moves. The claimed space used by appdata/system etc. (not I use a docker folder NOT docker file), was much larger (double) the space actually used in my cache. This leads me to believe much of the work is done using links. By doing some moves using MC, do I "destroy" those links? I haven't finished to actally move things to data disk, let alone move back to cache, but I want to be ready in case something is not as it should. My VMs should be ok. My appdata, I don't know, but I suspect there is a process to actually "recreate" the whole docker set I have, so this can be "cleaned" up. What about "system" and the millions of files and folders there? Note that already MC reported some files as already in the destination (which I am 100% sure is not a matter of the mover as it is doing something else), which seem to be the same files so I overwrite them to be sure. Also a single time it reported me a link will not work over disks, which I skipped, so mover can (possibly) handle it. Note also that I don't have "follow links" ticked in move... and I feel I should?

Nobody can help?

I am trying to setup an UNRAID for a friend. I want to use XFS Encrypted FS for the data, so I changed the filesystem to it, formated the (still empty) array and is looking OK. Problem is that using a pass phrase means manual work for my friend: First login to GUI, then navigate to Main, go to the bottom of the page and supply the keyphrase. (btw, when array is encrypted, I think UNRAID should just pop-up a requester for the key immediately going to ANY page in the GUI, not having to search for this!) Anyway, I want to use a keyfile so that (I want to believe - please verify) the array will autostart if it finds the keyfile. I do NOT plan to use FTP or hide it too much or whatever (it won't be used for hyper-sensitive data, just as an extra pre-caution), I just want to put it somewhere in the USB boot stick, so that if the stick is removed, even if someone gets the rest of the server, it is useless to them. So here are my problems: 1) UNRAID own GUI (accessed local on server with the default browser), doesn't do anything if I press "Browse" to supply keyfile. It is not even a matter of pop-up blocking, as I allowed localhost. Still doesn't work. 2) I tried to use another computer to access the GUI, clicked "Browse", indeed allowed me to select the file (which was local to the SECOND computer and I would copy to server later - or I expected that by selecting it, UNRAID would just copy it to a default location) and this in turn started the array and encoded it using that keyfile. I then copied the keyfile named in some arbitrary name (same as the one I "browsed" above), containing the "key" in it (no CR/LF, just a single line) to the root of /boot (which is the USB stick, ain't it?)... Rebooted and... nothing. Asks for the key. I need to re-supply it somehow (from a second computer that allows me to click "Browse"). So... help!? How can I put the keyfile manually somewhere in USB stick AND set UNRAID to look for it exactly there, automatically, without using the problematic local browser that I cannot use "Browse" on it? (so I can never supply using the web, the proper path to the keyfile)

thanks I am back in most servers I remember being, not on this yet

Will actually need some help from someone... I got that dreadful pizza floodbot trojan recently and it made me get kicked from crafty discord. Invites don't work, I am probably banned. Since I have no way to reach crafty team, can SOMEONE page some mod in crafty discord to let me back in? (NLS #7917) (obviously I cleared the spambot)

I have an issue too. When webui (probably) auto-logs off (after some time that is not set by the admin anywhere), if there is a tab pointing to dashboard, the system does not handle this properly and doesn't produce a page. It should just redirect to login page. If someone doesn't know, they can think the server is actually down. Is this something to bring up to Crafty 4 discord myself, or will you @binhex push it there?

They seem to have released an important update.

Well it is "kind of RAID4"...

OK this may have been discussed before, but I can only find fragments of info, nothing really clear. I need to setup a small server for a friend that currently uses a Windows machine with VeraCrypt on an encrypted (400GB?) file that then shares (the unencrypted volume) over his network. When his Windows machine reboots, he needs to re-enter the password for the encrypted volume then the share works. It works more or less well, but since he will move to UNRAID, I want to provide him a way to again have a way to encrypt his data (not all his data, not even 400GB are needed, probably way less). I know UNRAID supports encrypting whole disks (although I didn't find a guide for this), but I have questions: 1) How to set this up? (found the guide in the docs, but doesn't give any info on how to use after the setup) 2) How will this work after server reboot? (having to go to web interface to enter a password or something is fine - if it doesn't do that, the problem is that someone that can take the whole machine and start it, the decryption is automatic, so like non-existent) 3) Can I do some kind of encryption only on a single folder or share? 4) Some alternative way? (possibly through a plugin or container) Encrypted "whatever" needs to be accessible using SMB and allow a way to unlock of course. Help?

I migrated my world fine to this. Thanks man.

I hope Craft Controller 4.0 docker is "sooner" rather than "later". There is already a Crafty 4 container, but I prefer binhex's quality (without having an issue with the other guy - except maybe that his Crafty 3 was not updated ever?).

,8 indeed SOLVES the issue!

How easy is this? mineos-node compared to this, feature-wise?

Same for me... And I get timeout joining (although seem to "run"). [01:14:53] [Network Listener - #0/WARN]: [io.netty.channel.epoll.EpollEventLoop] Unexpected exception in the selector loop. io.netty.channel.unix.Errors$NativeIoException: epoll_wait(..) failed: Function not implemented [01:14:53] [Network Listener - #1/WARN]: [io.netty.channel.epoll.EpollEventLoop] Unexpected exception in the selector loop. io.netty.channel.unix.Errors$NativeIoException: epoll_wait(..) failed: Function not implemented [01:14:53] [Netty Epoll Server IO #1/WARN]: [io.netty.channel.epoll.EpollEventLoop] Unexpected exception in the selector loop. io.netty.channel.unix.Errors$NativeIoException: epoll_wait(..) failed: Function not implemented Any ideas?

Please tell me you are going to upgrade this container to crafty controller 4.0... I think it goes out in 3 days...

Well since the same two PCIe cards will be used, I guess they will report them the same exact way.

My server will have completely new mobo/CPU/RAM, but my disks will stay the same and cache will switch from SSD to NVMe. Actual disks are on two PCIe SAS/SATA cards. I know what to do with the cache (mark all cache based shares to cache YES instead of PREFER, disable VMs and Docker, invoke mover, old cache should be empty). What I am not sure is what to do with the actual array when I replace the hardware. I suspect UNRAID on "new" server will see the disks in a different order maybe? I need to be careful to set the same disks (by ID and serial) in same order on "new" server? This will be enough to start array and "accept" it as healthy with no need for rebuilds? Or the order doesn't actually matter? (as parity calculation doesn't care about that normally) Help? (and yes then with new cache, set cache based shares back to PREFER, I know that part)

Why is this still in 4.4.2? Something wrong? Latest is 4.4.3.1.

No so I stopped trying to use this, long ago.

Just a note. With last ddns container update, nothing changed. Still "starts and stops" immediately.

RC3d, same issue (if it was related anyway)...

Are you on RC2? Maybe you could try RC3c? It could be some backbone update that conflicts with the container. In that case I am sure the devs would be interested to know.

Man thanks for your effort. You don't have to apologize!