ksullivan86

I switched to unraid from windows thinking it would be more stable but WTF IDK if its Unraid being shitty or if its linux but windows was more stable for me when I ran my server on windows for years. Earlier tonight I got a notice that my Unraid server was offline and then when back online, later I checked my server and it only has an up time of 20 min and I am getting a warning about unsafe a shutdown. Then I noticed all my drives are writing and I check my parity and see its rebuilding parity WTF WTF!!!! This is the second time I have lost parity in about 6 weeks. I am at the point where I am thinking about getting my money back(canceling my recent credit card licence order due to faulty software) and going to proxmox. Has anyone else had issues with random reboots and loosing parity. This is really frustrating. im running 6.12.8 FYI

I run a few containers through a vpn(gluetun) with --net=container, but this doesn't seem to allow the use of container names for networking address that is possible when using containers on the same docker network. When using --net=container the docker network needs to be set to "none", this has to be the problem but when I add any of the containers going through the vpn to a network the containers are no longer reachable. I this is really easy to do in docker compose but I am hoping there is someway to do this with unraid templets. Can someone please point me in the right direction.

I just installed an Intel arc a770 16gb(asrock) but I have noticed the card is drawing about 60watts (my idle without the card is around 60w so the gpu is doubling my idle power while doing nothing except producing extra heat(the temps on one of nvme drives is about 8-12 F warmer) but even with that idle heat the cards fans stay stopped. I tested the card in plex transcoding 4k to 1080p and it worked fine, the cards fans only spun up a fraction of the time it was playing and I did notice power draw was about the same around 120w. I have gpu top enabled and was able to see the processor was being used. I also have powertop installed and switched all my turnables to "good" but that didn't seem to make any difference. I also have all the aspm setting enabled on the bios asrock z790 taichi lite. literally the only reason I put a gpu in and got and intel arc is because they can draw under 1 watt in idle, I didn't want a gpu because of the high idle power but when I read about intel's low idle I got one (mainly for AI as my 14700k handles about any transcoing amazingly well and efficient)

yeah IDK the flash drive is a newer one but I haven't noticed any issues with it that I can tell(idk if I could even tell). I rebuilt the parity and have rebooted a few times and now upgraded to 6.12.8, hopefully this never happens again but it was really annoying. Hopefully the dev team will dive into the diagnostic files and see whatever happened and fix the issue if this not the flash drive.

I have an intel arc 770 that I would like to use for in my unraid server. From what I read unraid need to have linux kernel 6.2 and from what I can see the new 6.12.8 is on 6.1.7. I am guessing that arc still wont work on the current version but I know there is a way to manually update the kernel but I am not entirely sure how this is done. Someone mentioned needing to compile the new file but I am not sure how to do this and more of the info is around a year old so I'm not sure how outdated it is. I think I need to use one of these kernels https://github.com/thor2002ro/unraid_kernel/releases but not sure what one I should use, but if I had to guess Id say use the latest release: 6.7.5 20240223 o2 gcc 6.12.8? After downloading and extracting the files what do I do next... do I need to compile? if so how is this done? Do I need all the files? I read somewhere if done correctly when the unraid is later updated unraid will update as normal.... how do I make sure this happens?? Do we have any idea when the 6.2 kernel will be added to unraid? Thanks in advanced

I shucked the drive and installed in to the server before the parity. I know the boards on the elements can change drive details so I made sure it was just the bare drive before I started to build parity. I basically did exactly what you suggested except I didn't preclear because I read that unraid does this automatically now but I did test the drive in the enclosure and then properly installed the drive. super lame, this is clearly a bug

I was having problems with some of the pages in unraid so I rebooted and except I when it came back up I my parity disk says "wrong". Its trial install of unraid that that I started to build parity 9 days days. Parity finished and everything was fine until I rebooted just now. The drive is a WD elements that I plugged into the server to test and make sure it was functioning before shucking the drive. Drive worked fine so I shucked it and installed it in my server. I didnt want any issues like I am having now so I made sure to only build parity after the drive was shucked. It seems like unraid is confusing the drive with before I shucked it(weird) If anyone has an idea that allows me to safely restore my parity that would be awesome as I really dont want to stress my drives any more than I need to media-center-diagnostics-20240222-0106.zip

I have been trying to find out what is the best file system type to run on a cache drive. I did my own testing using the windows in docker template and ran crystal disk mark for benchmarking. All test are done with only the system and appdata shares on the cache drive with only the windows in docker container installed and exclusive disk access for both shares. I ran crystal disk mark in the default path and a separate test in a shared folder linked to the appdata share so one test is fully on the docker VM and one is outside of docker. I also included a test from before my unraid switch with a test on windows 11 baremetal for comparison from what I can see from the test is there is no real clear winner. ZFS reaches the highest speeds but is not a clear winner as the read speeds are a lower (on folder outside VM disk) when compared to btrfs and xfs. I do find it confusing why the test on xfs outside of vm disk has higher write speeds than the hardware can actually perform, I am assuming any speeds higher than the drives hardware capabilities is actually taking place in ram... didn't think xfs used ram?? what do you think? Testing Hardware: taichi z790 lite 64gb corsair vengeance ddr5 (6400) intel 14700k Crucial P3 Plus 4TB PCIe Gen4 3D NAND NVMe M.2 SSD, up to 5000MB/s - CT4000P3PSSD8 windows for docker btrfs (c drive) Exclusive access windows for docker xfs (c drive) Exclusive access windows for docker xfs (outside vm disk) Exclusive access windows for docker zfs (outside vm disk) Exclusive access windows for docker zfs (c drive) Exclusive access windows for docker btrfs (outside vm disk) Exclusive access windows 11 bare metal

I am doing some testing to figure out how I want unraid formatted. This is my first unraid install but I have watched many of space invaders videos. I have a windows image backup folder(15GB) I tested transfering to a ZFS drive and a XFS in the same array from a unassigned nvme ssd. I also have turbo write speed enabled as well. You can see in the screen shots that XFS is about double the write speed of ZFS, is this normal? Also I do not have encryption and compression on...would having compression turned on help speed up ZFS? FYI both drives are shucked 12TB WD reds and have the following info: Model family:Western Digital Red Device model:WDC WD120EMFZ-11A6JA0 LU WWN device id:5 000cca 29bdb96dc Firmware version:81.00A81 User capacity:12,000,138,625,024 bytes [12.0 TB] Sector sizes:512 bytes logical, 4096 bytes physical Rotation rate:5400 rpm Form factor:3.5 inches Device:In smartctl database 7.3/5528 ATA version:ACS-2, ATA8-ACS T13/1699-D revision 4 SATA version:SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s)