sage2050

same issue with brand new samsung bar plus 64gb

did some more digging around, THIS solved it. Still have no clue what caused this to begin with.

I think you missed the part where I put 70+ hours into the game in a similar (but not the same) VM. something changed in the last couple of weeks to cause this, and I cant figure out what it was.

i5 12400. I gave my VM access to all cores, and also tried giving all cores but 0/1 pair. The VM I was running when it was working had all 12 cores btw.

This just started happening to me in Monster Hunter Wilds after playing 70ish hours without any issue (RTX 3080). I took a break from playing and in the interim I updated my motherboard bios, changed a few settings, and swapped nvme drives (both installations were baremetal). Now once the game starts my GPU utilization drops to 20% and my fps goes down to 10-12 no matter what graphics settings i select. Oddly I get full utilization in on the title and character select screens. I've tried all the suggestions in this thread but nothing has worked. I'm at my wits end here. I figure it must be an unraid setting or VM config that's causing the problem, because the game runs fine when I boot directly into windows. I've even tried creating new VMs with very basic settings, and installing on vdisks instead of baremetal and it's the same everywhere. Of the games I'm playing I'm only having this issue with Monster Hunter Wilds. the other games are not very graphically intensive. Furmark benchmarking runs just fine as well.

Trying to run Fantasy Life in a windows vm and none of the xml edits are working for me Edit: nevermind, I changed the network model from e1000 to virtio and it worked. I was using e1000 for a very specific reason.

this is exactly what I came for, thanks

well i updated my motherboard bios this morning and the problem went away 🤷‍♂️

i honestly dont remember what the docker-shell line is for, i will look into that, edit: its from an ancient spaceinvader one video about bashing into a running docker container. it's probably no longer necessary. the second script is a vm virtiofs trick thats now deprecated, the bootimage looks like old iommu separation that's now native to unraid, that can definitely be cleaned up. those pkgs are tools from the now deprecated nerd pack that i use frequently (except openssl, i dont even remember what i needed that for) so I installed them myself, not sure why they would cause any issues. ive never run a custom kernel, just betas and release candidates.' the successful driver install you're seeing is possibly me removing the plugin and reinstalling it, as that's the only way to get it working after a reboot, like in pa77ys case.

I'm currently having the exact same issue as you (also on an asrock mobo) , all overclocking is disabled as I was fiddling with bios settings to try to hit deep idle cstates. Unfortunately for me this coincided with an update to rc1 so I'm not sure if it was the bios or the software update that caused it. Id really like to avoid doing a fresh unraid install. servbot-diagnostics-20250423-2318.zip

Denuvo games work in unraid 7.1

nevermind im an idiot. I forgot to update the gpu firmware.

i have Clever Access Memory enabled in bios and have done all the steps outlined in this post, but still in my VM it says that reBAR is not enabled in bios. I'm stumped.

here's a script that will check and restart your steamcmd servers if there's an update (kept getting pinged about satisfactory updates when my west coast friend was trying to play). I've got it on a 30 minute cron #!/bin/bash APP_ID= CONTAINER_NAME= APPDATA_FOLDER= LOG_FILE="$APPDATA_FOLDER/update.log" # Get the current installed version CURRENT_BUILD_ID=$(awk -F '"' '{if($2 == "buildid")print $4}' $APPDATA_FOLDER/steamapps/appmanifest_$APP_ID.acf) echo "Current Build ID: $CURRENT_BUILD_ID" # Get the latest version using SteamCMD inside the container LATEST_BUILD_ID=$(/mnt/user/appdata/steamcmd/steamcmd.sh +login anonymous +app_info_print '$APP_ID' +quit | grep -A 3 'buildid' | grep -oP '\d+' | head -1) echo "Latest Build ID: $LATEST_BUILD_ID" if [ "$CURRENT_BUILD_ID" != "$LATEST_BUILD_ID" ]; then echo "$(date): Update required: Current Build ID ($CURRENT_BUILD_ID), Latest Build ID ($LATEST_BUILD_ID)" > $LOG_FILE echo "Restarting the $CONTAINER_NAME server to apply updates" >> $LOG_FILE docker restart $CONTAINER_NAME else echo "The server is up-to-date: Current Build ID ($CURRENT_BUILD_ID)" fi

I started running the SMART test right after posting, attached here (completed without error) I checked logs of my docker containers that interact with that folder and didn't notice anything out of the ordinary, i'm very confused. fortunately it was non-critical data servbot-smart-20241104-1056.zip

I lost a bunch of files, specifically in /mnt/user/Media/downloads/_Music (seems like only here, files gone but folders still intact) on Oct 31, sometime between 9:30PM and 10:00PM. a quick look at my syslog shows an nvme error occurring at 9:50PM, but those files were not on the nvme. 1) what does this nvme error mean, and what should I do? 2) is this the culprit of my lost files? 3) is there anything else in the diagnostics I should be aware of? servbot-diagnostics-20241104-1035.zip

my psu is well oversized for what i'm running - is there a good way to test the rails?

I rebooted again and this time a *different* disk went missing this isn't looking good. edit: another reboot and all disks present. this is worrysome.

I swapped sata cables with the drive next to it and both appeared after reboot 🤷‍♂️

New disk only a few months old. Diagnostics from after reboot attached, dont have any from before unfortunately. servbot-diagnostics-20240924-2050.zip

Is this different from the version in the nerd tools pack? I -get- why people say python shouldn't be installed directly, but I frequently run into permissions issues trying to run scripts from docker containers. running scripts in venvs has always just seemed to work better.

mktorrent no longer working in unraid 7 beta. I've tried uninstalling and reinstalling.

- replacement of included virtiofsd w/ rust version - script to export shares to different drive letters - hugepages memory reservation to stop high cpu usage

Can any of these custom fixes be disabled in the 7.0 beta?

got hugepages setup and the cpu issues are completely gone. I had some docker containers using some memory apparently, I had to kind of guess around at the size until i saw the command above and resized accordingly. Is there any way to tell with certainty which containers are the culprit? i have a small postgres running but it doesn't account for all the reservations.