Fizzyade

Since upgrading to 6.12 and updating the nvdia I've been getting GPU errors, like the following: [56386.612293] NVRM: Xid (PCI:0000:08:00): 31, pid=17147, name=ffmpeg, Ch 00000008, intr 10000000. MMU Fault: ENGINE GRAPHICS HUBCLIENT_FE faulted @ 0x1002_06807000. Fault is of type FAULT_PDE ACCESS_TYPE_WRITE [57317.960662] NVRM: Xid (PCI:0000:08:00): 69, pid=27124, name=ffmpeg, Class Error: ChId 0020, Class 0000c1c0, Offset 00001b0c, Data 00008000, ErrorCode 0000000c I am using Emby in a docker container, it's the only thing that uses the GPU and once the GPU fault occurs, it kills that ffmpeg which then causes Emby to either just stop playing the stream that was being played, or it jumps ahead to the next episode if it's a TV series that is being watched. Nothing else has changed and the machine was 100% stable prior to updating. I know when I was on a previous Unraid release, I had stuck to an older driver and I never had any issues, when I checked the driver being used it was using "latest", so as a test I changed the settings in the plugin to use " v530.41.03" and all the issues have now gone away, it's back transcoding without any GPU faults.

64GB. I really wish that Unraid supported lxc containers though, I have a NUC running Proxmox which has a load of containers set up which I use for CI (TeamCity), but I need more containers to handle other builds and the NUC has no more ram (it's got 64GB as well) and unraid would be ideal if it supported them.

64GB although I could do with more, I have multiple virtual machines spun up at once for development work (Windows 10, Fedora, the latest Ubuntu in addition to the oldest Ubuntu LTS which is used to build AppImages. I also have VM's set to be CI agents. My NUC which runs proxmox also has 64GB.

I just upgraded to the latest rc and this is still broken for me. Modify the port in the XML and then go make a change in the GUI and the port reverts back to auto and -1. This drives me up the wall, especially with a fair few VM's and trying to remember what the port setting actually was.

If you change the VNC port in the XML to be a specific port instead of auto (which is the behaviour that you want, you want to connect to the same VM on the same port) and then sebsequently change any setting of the VM using the gui the port reverts back to the auto port, losing your nicely configured static port setting. We need to be able to show and set the port in the gui so that it can easily be configured to a static port and more importantly that the mapping isn’t lost when you make a GUI change. im forever having to go back and edit the XML to restore the vnc port.

Probably already said it, but set the VNC port in the GUI so it doesn’t get lost when changing configuration xml. Drives me absolutely nuts.

It's running right now and now emby is running like a slug.

Ok, I removed the mover plugin and so forth, only share that's set to use the mover is my downloads. Last night I could tell it was running (I get pauses when trying to start playing something in Emby), went to the server and sure enough every core full red.

it doesn’t, the reason I initially installed mover tuner is because the mover pegs the processor.

Right, I’m ready to give the mover another go. the only thing I’m interested in using it for is my downloads. are there any dos, do nots. Any sane settings I should start with. is it possible to set the mover logging to log to the array, so that if I end up having to restart the server due to the mover, I can produce a log file? thanks.

So it all went wonky again last night, every CPU pegged to 100%, system unresponsive, in the end I had to reboot it, of course that meant I lost the logs. There was only my media folder using the cache, the actual downloads were not cached. No overclocking. I just stopped docker & vm's, ran mover and then disabled the couple of shares that were using cache, then removed the cache drive.

I had actually only turned that on a few hours before I posted the screenshot just as something else to try, it’s been running disabled forever. ive just updated the plugin (to the correct one) enabled logging and redisabled turbo write.

Here's a screenshot of the mover tuning settings. I looked in the syslog file and couldn't see any mover entries. Thanks for your response, I know this isn't your issue. it's the mover, it shouldn't kill a 16 core machine though, even if IO bound I would have expected setting the priority low that everything else would take priority.

I'll do that now. The mover is actually doing my head in, so much so that I'm contemplating removing the cache drives because my machine absolutely tanks when it runs. I'm sure at one point I had this plugin working to solve the issue, but I just can't seem to get it to work, the mover runs and until it's finished I can't play any movies from Emby. I was so happy when I found this plugin.

Does this still work? No matter what I do, mover absolutely kills my machine, every core showing red on the dashboard. tried changing priority’s and it makes no difference. Mover has just started and now I can’t play anything on Emby because it’s killed the computer. I’m sure this plugin used to help, but I just can’t seem to change anything in it that helps.

VM enhancements....mainly that the "basic" editor only change fields/attributes that have been edited. It's incredibly annoying to change something only to find that the VNC port has reverted to auto *yet again*. Heck I'd even settle for just that stopping happening.

Both were 64GB Sandisk UltraFit drives.

I've had 2 USB stick failures in less than 12 months, the first one I was surprised at after 5 months, but just put it down to bad luck. Then I had another one at the next 5 month interval, this then made me very nervous, I had to contact support to get them to issue the new key as I'd gone through my "grace keys". When I saw 29000 and then 35000 writes after 7 days I got very nervous again and this was when I decided to try and find out what was the root cause which I eventually tracked down to this Plugin. I've removed it now, I wasn't really using it anyway, but that volume of writes is not good! Surely this is something that can just be stored in ram and then only written to USB on an array stop or shutdown. Unscheduled shutdowns (power loss, kernel panic) are tough luck.

Following my USB failure last night, I tried to create a new USB key under Catalina and again was unsuccessful, disk was not bootable. The tool or make_bootable_mac script didn't work, I've disabled SIP. When I run the script, I get: INFO: The following drive appears to be the unRAID USB Flash drive: /dev/disk5 61.5GB Permit UEFI boot mode [Y/N]: y To continue please enter your admin INFO: Unmounting /dev/disk5 Forced unmount of all volumes on disk5 was successful INFO: Writing MBR on /dev/disk5 0+1 records in 0+1 records out 447 bytes transferred in 0.002103 secs (212544 bytes/sec) INFO: Mounting /dev/disk5 Volume(s) mounted successfully syslinux for Mac OS X; created by Geza Kovacs for UNetbootin unetbootin.sf.net shell-init: error retrieving current directory: getcwd: cannot access parent directories: No such file or directory "/dev/disk5s1" unmounted successfully. /dev/disk5s1: invalid media signature (not an FAT/NTFS volume?) INFO: the Unraid OS USB Flash drive is now bootable and may be ejected. Unsurprisingly the disk isn't bootable. I managed to borrow my daughters windows laptop this morning, used the USB creator and copied across a backup of the config and the disk booted up straight away, no issues. We desperately need another solution to this, Catalina has been out a while and the USB creator is still broken. If you can't fix it (and even if you can) in the mean time can you supply an img that can be written using balena etcher so that there's an easy way of creating a working USB stick with no fuss under Catalina.

I spent all morning trying to make a bootable disk on my mac and failed, disabled sip, added full disk acces. The disk is created but is not bootable. If I try manually, using the script I get errors from syslinux abount being able to unmount. If I run syslinux myself manually (same command that the script does) syslinux doesn't error and says it completed corretly, but again, the disk is not bootable. Eventually I booted into Windows, ran the utility there and it worked first time. It would be nice to have prepared images available so that we can use other software like etcher which has never failed me.

are the priority settings still working under the latest unraid? I'm sure they used to work, but now when the mover runs I get buffering on my server if it's moving (i had them set to low and it seemed to work before, never had issues with emby server under docker)? Is there any way of confirming it is Working?

It’s definitely a compatibility with the 10Gtek intel modules and the X520, no link flaps since I installed the genuine intel module. i just migrated my esxi server to Proxmox, using a thunderbolt 3 enclosure with another 10Gtek module plugged into an x520, after it installed I was pinging hosts on the network and it was was down up and down like nobodies business, had a 60% loss rate on ping. went searching in my magic networking box as I knew I had another intel module (but wasn’t sure what type it was), luckily it was a also MMF SR module, switched that in and straight away ping was behaving and the link stopped flapping. i’ve not Had any issues with the 10Gtek modules in MikroTik, UniFi or Aquantia devices, just the intel ones. you live and learn.

Ok, 12 hours in since I swapped out the 10GTek SFP+ module for the Intel SFP+ module and not a single drop since then. Actually hopeful now that this is cured.