salvdordalisdad

OK I give up. I'll have a go with unmanic or fileflows instead. Shame.

Hi All, Successfully installed tdarr & nvidia cards in 2 servers, thanks largely entirely to spaceinvaderone, top help. There are both server and node dockers on the same host. They are working fine on local files, but I want them to work on a shared library. host1 = server + node running as UID =99 & GID = 100 and with target files & folders locally in /mnt/user/share node2 is running as UID =99 & GID = 100 Files owner are largely set to nobody:users host2 = node only with SMB or NFS access using mapped drive to /mnt/remotes/n1 Files in this location also appear to be owned by nobody:users Node1 works OK, finds files, transcodes, puts the files back as expected. Node2 fails. Finds files, transcodes in local cache, then fails to put them back with "Copy Failed" I've attached an output, but it's not very helpful (as far as I can interpret anyway !) Oddly, if I open a console on node2 I can manually copy the files across AOK. So it looks like permissions, but it really doesn't - eek! Anyone with any suggestions please? thanks in advance. job-report-copy-failed.txt

Thanks very much - that's a perfect answer. Much appreciated. Specifically --runtime=nvidia inthe "extra parameters" section as you say. However - the double dash before it is vital - it fails otherwise. Yup that worked - thanks for the nudge! I knew i was doing sommat wrong, it's just not that clear sometimes where these things go... ;-)

Hi All, Huge Unraid fan - have been for many years. I apologise if I've put this post in the wrong bin, I wasn't sure where to put it! Aim of this mini-plan = get some hardware transcoding going for Emby & Tdarr. So I have a NVIDIA Quadra P400 GPU installed, x16 slot. Drivers are insatlled & running, Nvidia Driver Version: 515.76 (had to replace the USB boot disk because it didn't have enough room for them). I have a GUID for the board. I also have GPU Statistics Plugin installed & it can see the GPU. So far, so good... But I don't seem to be able to use it in either Emby or Tdarr. (I think) I've found all the extra variables I'm supposed to be using (according to some convincing looking sources), screen grabs attached However, there's no clues that the emby or tdarr can see the GPU, not listed explicitly, and definitely not being used, according to the CPU stats & the GPU stats. I'm clearly missing something, or have got it wrong somewhere, please help me & nudge me in the right direction? Those are the only 2 dockers I'm likely to be using it with, not looking to make a games VM or anything. Thanks in advance.

Ah spit. I ran xfs-repair on all the disks & disk3 complained. So did waht it say, & mounted & dismounted & re-ran, no errors now. But the disk used sizes didn't change. so Didn't achieve very much at all. But lots of system gone now. No community apps, no user scripts, ahh spit. Can't download cummunity - unraid version too low. Really screwed it. Well so be it, it's only hours. grr Next time - LEAVE IT ALONE !!!!

Gave up on finding an answer, so deleted everything. When "empty" the disks still had some data in ther, system, iso images for vms etc, but not much: 8.5G ./disk1 64.6GB 8.2G ./disk2 9.08GB 61G ./disk3 120GB 0 ./disk4 55GB CLI GUI However, the GUI still reported somewhat less than empty...when compared to the "du" command on CLI. I don't care all that much about such a small amount, but do I need to do a "disk check" to clear them down ??

Cripes, that took a loooong time & got me not very far. That was 11 days spent trying that suggestion, and not successful. I guess it eliminates a variable, but...crikey. So the rsync with delete option has finally finished, no difference in disk size. n2 = 27.5TB n1 = 23.6TB So what is difference? I've been through the whole disks. If I don't get some useful suggestion abotu where to look, I'll have to trash the whole server & start again. Not very happy about that, really dents my appreciation of UNRAID, which has been very positive until this...

Still bashing away at this. Have run that script a dozen times now, takes a looooong time to run, and keeps breaking for one reason or another. It's deleting a bunch of stuff, but still recording as 27.9TB compared to 23.6TB The server just lost all its marbles, all the shares, just a blobby mess, so I had to reboot it & restart the script (yet again). THis will be the last attempt, it wil be quicker to scrap the whole thing & start again! Update to follow.

Hiya JorgeB Number of files: 411,199 (reg: 380,126, dir: 31,073) Number of created files: 1,927 (reg: 1,866, dir: 61) Number of deleted files: 10,757 (reg: 8,620, dir: 2,137) Number of regular files transferred: 379,786 Total file size: 22.88T bytes Total transferred file size: 22.76T bytes Literal data: 0 bytes Matched data: 0 bytes File list size: 524.23K File list generation time: 0.001 seconds File list transfer time: 0.000 seconds Total bytes sent: 13.20M Total bytes received: 2.21M sent 13.20M bytes received 2.21M bytes 125.81K bytes/sec total size is 22.88T speedup is 1,484,964.81 (DRY RUN) rsync error: some files/attrs were not transferred (see previous errors) (code 23) at main.c(1330) [sender=3.2.3] Script Finished Mar 07, 2024 19:39.59 Full logs for this script are available at /tmp/user.scripts/tmpScripts/__rsync-delete-test/log.txt I can see that there's a LOT of files it's trying to delete. I will go through the list it's generated & see if they're OK to delete & then let it go ahead & do it's thing. Thanks for the nudge - looks like the right direction. I'll take a couple of days at least to go through the list & then report back. Ta sdd

HI All, Love unraid, have 2 servers n2 is just a backup of n1 RSYNC is run regularly to copy data across. I don't delete missing stuff during rsync but I clean it up manually from time to time (bcompare helps), so there's always a few leftovers & a small amount extra, but not 4.9TB worth. However, despite runing bcompare to confirm they are pretty similar, n2 shows disk space used = 28.4TB whereas n1 shows it as 23.5TB I've checked on the CLI using du, and only found around 0.6TB difference, when looking at the array share /mnt/user/filestore However, checking the individual disk shows a very different story: n1 = /mnt/disk1 = 5.3T ./filestore n2 = /mnt/disk1 = 6.6T ./filestore All 4 disks are the same sort of thing. (XFS) I also ran xfs_repair on one of the disks, as I got an error on it, but didn't fix this though. The other directories on these disks don't account for very much, system, appdata etc. Interesting the GUI and CLI don't agree in the detailed numbers, but still generally in overall amount. Not sure where to start looking now. I know the last resort is to wipe & restart, but that's a whole week's worth & is scary. Any pointers anyone can suggest?

Update...fell over again this morning, NOTHING in the syslog. maybe the GPF happened at a lower level than syslog was capable of? Locally connected screen says "kernel panic" Have removed offending (probably) memory module & rebooted., oh joy. Give it a week & then send the memory off for warranty.

Hi All, Interesting update... Rigged up the cables & motherboard header & got the speed right & the settings etc... Boot process showed the same menu on the console as the main screen - which is a positive step. Interestingly (almost) there were more detailed outputs from the serial console than the main screen. However, once the boot process finished, the serial console didn't respond to any keyboard entry. Maybe there's a 2nd level listener process which I haven't enabled or setup? I've also rigged up a terminal server (Lantronix EPS2-100) which will be connected - already tested, and the cabling is really easy - it's just RJ45 cisco flipover flat cable & the standard Cisco db9 adapter, simples. (edit - I too used the "xterm-256color" and the boot menu came up in "glorious technicolor" how fabulous.) Still can't login after it's boot ed though ;-/ I will re-read the above notes & make sure I've done all the steps but if anyone has a nudge I'd be grateful.

Update... 4 days in & no General Protection Faults anymore... So I will now close this as "maybe solved" by just re-seating the RAM sticks <?!> I'll also stop looking at syslog on a daily basis... It's still running, so if there's a crash, I will look through & see if there's a clue... fingers crossed! YMMV

Update... Single RAM stick = several days test = 0 errors. (Server WAS headless, no graphics card, but change in memory forced temp use of graphics card.) Replaced 2nd RAM stick now memory is good again, BIOS recignised it, but refused to boot. Long story short, new SATA PCIEx1 adapter, but now it refuses to boot without the graphics card. Slightly annoying, needs looking into, must be BIOS setting, but it can wait. Anyway, 12 hours after booting with both RAM sticks, still OK...no new GPF errors yet. If it re-errors, it confirms original diagnosis & mempry can go back for warranty, if not, end of job. Update to follow.

Thanks very much to all above for this info. Am in process of trying it, but I wanted to add a small detail about NULL MODEM cable, for anyone watching...who is a networking techie with loads of Cisco console cables in his bag (er....like me). Cisco Console cable = most of a null-modem cable. There are two versions, logically/effectively identical, just mechanically different. Old version = flipover RJ45-RJ46 flat cable (for serial comms only) + DB9-RJ45 adapter, usually grey. - can be separated. New version = Light Blue moulded cable - same connectors - cannot be separated. So you can make a null-modem cable with a pair of them. (DB9---RJ45 cable )(either version) + DB9-RJ45-adapter (old version) Or ifyou have 2 new ones, you can connect them together with a RJ45-RJ45 Coupler, just quite long & unwieldy (& has to be a straight coupler) I hope that makes sense... Or if you're not a hoarder of such things, then do as the man say & buy one, ebay has them for a fiver... Good luck.