feraay

yes seems so. for my understanding the problem is only related to hdr tone mapping and not transcoding in general. so we need to wait for a 5.16+ Kernel. Hopefully that will work. Somewhere I read 5.17 also includes Fixes for igpu 9. Generation and above. But maybe 5.16 will fix it. https://www.phoronix.com/scan.php?page=news_item&px=Linux-5.17-More-Intel-TTM-Prep Highlights from this week's pull include: - A fix for GPU hangs caused by certain media and OpenGL workloads that were able to hang Skylake/Gen9 hardware and newer. Idk if it’s relevant for transcoding. @limetech 5.16+ need to be LTS before we can give it a try as RC Version?

Moin, ich steh etwas auf dem schlauch wahrscheinlich ist die Lösung super simpel. Ich habe einen VPN Container. Hier dann ein custom Network erstellt und den sabnzbd Container über dieses Netzwerk via --net=container:nordlynx. Funktioniert alles super. Im nordlynx Container Port 8080 gemapped. Jetzt würde ich gern aus kosmetischen Gründen via pi-hole custom DNS die sabnzbd Web-gui via sabnzbd.box erreichen. Dafür müsste ich ja entweder den Port 8080 auf 80 Mappen oder einen reverse Proxy in das --net=container:nordlynx Netzwerk hängen. Da der nordlynx Container aktuell im Bridge läuft könnte ich hier Port 80 und 443 ja auch nicht Mappen. Ziel ist einfach die Container im net=container:nordlynx ohne Angabe von Ports aufzurufen. Kann mir jemand einen Tipp geben wie ich hier am besten vorgehe. Ich habe ein zweites Docker net proxynet mit einem Swag Container für overserr via Cloudflare. Aber ich will die Services aus dem nordlynx net gar nicht nach außen erreichbar machen. Es geht tatsächlich nur um den Aufruf von lokal damit ich mir sparen kann die Ports hinten dranzuhängen. Vielen Dank.

it doesn't matter. Coffe Lake is 8. Gen 2017. Just because the result is the same for some people ...... random crashes. Doesn't mean the reason is the same. You can see in some Logs posted here that the Alder Lake IGPU hangs and that's clearly the cause for crashes with 12 Gen CPUs. Of cause there are 1000 other reasons for getting the same result (a system crash) but there is no connection. Hope you get my point. i have tried anything that was suggested here. Nothing is working for Alder Lake we need to wait for 5.17 Kernel. May we have Luck with 5.16 but I don't think so. See my last post in Alder Lake Thread. You can see the differences in the uptime. With my 12900K and HW Transcoding in Plex I will bring my server to crash in Minutes. The guys here with other hardware a talking about days till it happens. So clearly not the same reason just the same result.

https://www.phoronix.com/scan.php?page=news_item&px=Linux-5.17-More-Intel-TTM-Prep Highlights from this week's pull include: - A fix for GPU hangs caused by certain media and OpenGL workloads that were able to hang Skylake/Gen9 hardware and newer. This is Kernel 5.17 so I don't think 5.16 will solve the Problem.

are we sure that 5.16 Kernel will solve our problems? https://tomthegreat.com/blog/setting-up-ubuntu-20-04-lts-for-plex-with-intel-gen-12-cpu/amp/ He is using Kernel 5.15 on Barematel Ubuntu installation with plex and it sounds that he is not running into this transcoder problem. So how solid is it that 5.16 will solve the Problem with GPU Hangs on newer Intel IGpus?

just a small update i uninstalled Intel GPU TOP created just a empty i915.conf in /boot/config/modprobe.d and set i915.force_probe=4680 i915.enable_guc=2 in syslinuxconfig Two transcoding are running since 15 minutes and no GPU HANG in the logs till now. With the intel gpu top installed the error appears just in time. May its just luck I am not sure. ok it happened Mar 15 16:11:42 Mycroft kernel: i915 0000:00:02.0: [drm] Resetting vcs1 for preemption time out Mar 15 16:11:42 Mycroft kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 12:4:28fffffd, in Plex Transcoder [17657] Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 12:4:28fffffd, in Plex Transcoder [17657] Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] Resetting vcs1 for stopped heartbeat on vcs1 Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] Resetting chip for stopped heartbeat on vcs1 Mar 15 16:11:53 Mycroft kernel: [drm:__uc_sanitize [i915]] *ERROR* Failed to reset GuC, ret = -110 Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* Failed to reset chip Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm:add_taint_for_CI [i915]] CI tainted:0x9 by intel_gt_reset+0x276/0x29b [i915] Mar 15 16:11:53 Mycroft kernel: [drm:__uc_sanitize [i915]] *ERROR* Failed to reset GuC, ret = -110 Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] Plex Transcoder[17657] context reset due to GPU hang Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* rcs0 TLB invalidation did not complete in 4ms! Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* bcs0 TLB invalidation did not complete in 4ms! Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* rcs0 TLB invalidation did not complete in 4ms! Mar 15 16:11:53 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* bcs0 TLB invalidation did not complete in 4ms! Mar 15 16:11:58 Mycroft kernel: Fence expiration time out i915-0000:00:02.0:Plex Transcoder[17657]:7cfe! one transcode died but the server is still responsive and didn't crash and the second transcode is still running. second transcode also crashed and Plex docker crashed but server is still responsive. The WebGui was still accessable but server did not respond anymore. I will go with cpu transcoding and test again with Kernel 5.16

i am on 6.10 rc3 yes its alder lake correct. diagnostic is attached. mycroft-diagnostics-20220311-2201.zip

Had my first crash with 6.10 RC3. So in advice of @Ich777 I do the following. Plugged a HDMI Dummy to Onbard HDMI. Removed the chmod -R 777 /dev/dri from go file. Installed Intel GPU TOP created the /boot/config/modprobe.d/i915.conf File ok so after 3 crashes in a row and a damaged Plex config I can also see this in the logs. Mar 11 15:16:13 Mycroft kernel: [drm:__uc_sanitize [i915]] *ERROR* Failed to reset GuC, ret = -110 Mar 11 15:16:13 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* Failed to reset chip Mar 11 15:16:13 Mycroft kernel: i915 0000:00:02.0: [drm:add_taint_for_CI [i915]] CI tainted:0x9 by intel_gt_reset+0x276/0x29b [i915] Mar 11 15:16:13 Mycroft kernel: [drm:__uc_sanitize [i915]] *ERROR* Failed to reset GuC, ret = -110 so added i915.enable_guc=0 we will see ok i915.enable_guc=0 resulted in a crash also. will change it to 2 and test again. I was not able to geht a log from guc=0 the crash was faster ^^ with i915.enable_guc=2: ar 11 16:08:33 Mycroft kernel: i915 0000:00:02.0: [drm] Resetting vcs0 for preemption time out Mar 11 16:08:33 Mycroft kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 12:4:28fffffd, in Plex Transcoder [18057] Mar 11 16:08:44 Mycroft kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 12:4:28fffffd, in Plex Transcoder [18057] Mar 11 16:08:44 Mycroft kernel: i915 0000:00:02.0: [drm] Resetting vcs0 for stopped heartbeat on vcs0 Mar 11 16:08:44 Mycroft kernel: i915 0000:00:02.0: [drm] Resetting chip for stopped heartbeat on vcs0 Mar 11 16:08:45 Mycroft kernel: [drm:__uc_sanitize [i915]] *ERROR* Failed to reset GuC, ret = -110 Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* Failed to reset chip Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm:add_taint_for_CI [i915]] CI tainted:0x9 by intel_gt_reset+0x276/0x29b [i915] Mar 11 16:08:45 Mycroft kernel: [drm:__uc_sanitize [i915]] *ERROR* Failed to reset GuC, ret = -110 Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] Plex Transcoder[18057] context reset due to GPU hang Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* rcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* bcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* rcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* bcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* rcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* bcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* rcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* bcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* rcs0 TLB invalidation did not complete in 4ms! Mar 11 16:08:45 Mycroft kernel: i915 0000:00:02.0: [drm] *ERROR* bcs0 TLB invalidation did not complete in 4ms! what about 3? GuC submission and power management is enabled by setting the kernel module parameter: i915.enable_guc=1 HuC authentication only is enabled by setting the kernel module parameter: i915.enable_guc=2 Combine for both features together: i915.enable_guc=3

I am on Alder Lake so I just use i915.force_probe=4680. Deleted i915 File under modprobe.d. Works 👍 My VM with Nvidia passthrough throw a error internal error: PCI host devices must use 'pci' or 'unassigned' address type on start. I just recreated the VM it boots without a error. So for now looks good.

you need to use vbios I think. For me its only working with vbios. I'm not sure if Unraid legacy boot is needed but for me its working without uefi

do you use a vbios file? Two weeks I ago I was also struggling with Nvidia passthrough. Changed Unraid Boot from Uefi to legacy setup Vms completely new etc. Got always stuck in Tiano Bios logo. Seems like the GPU Output got frozen on the bootup of the vm. SO startet from scratch with Unraid Boot in Legacy Onboard Gpu enabled on MSI Boards its called IGPU Multimonitor I guess. Installed everything with vnc on first and gpu second and a vbios File. Plugged the Monitor into HDMI because DP just shows up when windows is booted.

same for me. 6.10 RC2 Intel Core i9 12900k. Intel GPU Top installed and IGPU passthrough to Plex Container. Sometimes the whole System is unresponsive via http and ssh. And sometimes the Transcoding Handler just don't stop and using 100 percent CPU on 12 Threads the webgui is still working then but thats all. When I do a fresh boot and test transcoding nothing happens its just running as expected. After a day or two the Server becomes completely unresponsive. I have /boot/config/modprobe.d/i915.conf with content blacklist i915 and Intel GPU Top Installed. In go file I have just chmod -R 777 /dev/dri Seems I am not using i915.force_probe=4680 or just to stupid to find it. So i sm not really sure if my settings are correct to be honest.

Hey Please check my script here: https://forums.unraid.net/topic/47160-how-to-automatically-wake-from-sleep/?tab=comments#comment-649452 Example Monday to Friday wakeup at 9.55 time=09:55 now=$(date +%s) other=$(date -d $time +%s) dayofweek=$(date +%u) if [ $now -ge $other ] && [ $dayofweek -lt 5 ] then echo `date '+%s' --date='tomorrow 09:55:00'` > /sys/class/rtc/rtc0/wakealarm elif [ $dayofweek -ge 5 ] then echo `date '+%s' --date='next monday 09:55:00'` > /sys/class/rtc/rtc0/wakealarm else echo `date '+%s' --date='today 09:55:00'` > /sys/class/rtc/rtc0/wakealarm fi

sorry for late answer. so you want the server to wake up every day at the same time? time=09:55 now=$(date +%s) other=$(date -d $time +%s) if [ $now -ge $other ] then echo `date '+%s' --date='tomorrow 09:55:00'` > /sys/class/rtc/rtc0/wakealarm else echo `date '+%s' --date='today 09:55:00'` > /sys/class/rtc/rtc0/wakealarm fi to be honest the better way to do it is set a alarm in bios and repeat every day. the script is only needed if you don't want the server to wake up every day.

That means a simple hdd passthrough is not enough for stable ZFS? I set the vdisk path to /dev/disk/by-id/ata-WDC_WD30EFRX-68EUZN0_WD-WCC4N7Ve23a. So he grabs the whole HDD. I read so much different stories the last days. I unterstand that the best method is to passthrough a hba. So it will be like bare metal installation. But in FreeNAS Forum they say for home use small machine it would be ok to do a vm without hba passthrough. So that is what I want so use it for. For this use Case they say to have redundant storage is the best way. Means do a raid before setup VM. But on the other site they say its not recommended to to a raid with ZFS but this only meant for a HBA passtrought? Its a bit confusing to be honest.

yesterday I tested freenas on top of that. at the moment only with vdisk this is really bad and don't make sense to use. But Freenas is running pfsense is running just have to passtrought some hdds and it should be fine. We will see. Everyone says the best way is to passthrough hba and lis but I can´t do so. would it make any difference for ZFS if I only passthrough hdds directly? OK it would be bad passthrough only hdd. So there is no way to work without a separate pcie sata controller right ? Read a bit more about that. So I need a 200 Dollar LSI Card to Passtrought for real? thats heavy.

but it should work. running pfsense since 2 weeks with no problem with br0 as lan and one passthrough pcie intel Nic as wan. ( testing only. recommended bare matel installation I know ^^) I also want to test a freenas vm with passed trought hdd´s. But I did not started yet. Have only a small Desktop System with 3 HDDs at the moment. So the only way it would make sense for my system, is to passthrough all hdds but noway to do so I think. Best would be a pcie card to passthrough whole controller. Or Proxmox. But don't want to miss unraid. what exactly is your problem with passtrough? you can just add it to syslinux.cfg and unraid would pass it on boot, so you could just select it in the vm edit page at the bottom.

cache pool ? same disk size? hpa detected? don't know this warning to be honest. but back to topic how do you passtrought the card? do you use other cards for unraid or Nvidia as single? i have also a problem with Sound. But beside of creators update or not. Also enabled MSI Mode but without effect. The Sound is distorted or warped after windows goes to standby or just disable the Screen. Tested it with a amd card on first slot and Nvidia second. Same result. So went back to Nvidia card on first single. With the vbios dumping method. But Problems with Sound are also the same^^ Maybe have to test MSI again.

Same here, passing trough a Gtx 1080 with a dumped vBios. Works very well. hyperv enabled. <hyperv> <relaxed state='on'/> <vapic state='on'/> <spinlocks state='on' retries='8191'/> <vendor_id state='on' value='none'/> </hyperv> No issues until now. Is it like it should be?

Nvidia Audio Controller is also Passed Trought?

Hello, i am running a win10 vm with a Nvidia passthrough, works very well. Now its my second or third try to passtrough some USB ports for plug and play functions. I´m started having doubts about me. Maybe I just understand the whole thing wrong but I think my problem is my motherboard MSI Z97S specifications. Bus 1 --> 0000:00:1a.0 (IOMMU group 4) Bus 001 Device 002: ID 8087:8009 Intel Corp. Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 2 --> 0000:00:1d.0 (IOMMU group 9) Bus 002 Device 002: ID 8087:8001 Intel Corp. Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 3 --> 0000:00:14.0 (IOMMU group 2) Bus 003 Device 003: ID 1b1c:0c03 Corsair Bus 003 Device 002: ID 046d:c52b Logitech, Inc. Unifying Receiver Bus 003 Device 004: ID 045e:02e6 Microsoft Corp. Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 4 --> 0000:00:14.0 (IOMMU group 2) Bus 004 Device 002: ID 0781:5583 SanDisk Corp. Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub it shows. bus 1 to 4 root@Watson:~# lspci | grep USB 00:14.0 USB controller: Intel Corporation 9 Series Chipset Family USB xHCI Controller 00:1a.0 USB controller: Intel Corporation 9 Series Chipset Family USB EHCI Controller #2 00:1d.0 USB controller: Intel Corporation 9 Series Chipset Family USB EHCI Controller #1 3 usb controllers. Ports: 2x USB 3.0 Front 4x USB 3.0 2x USB 2.0 but all ports are on Bus 3 --> 0000:00:14.0 (IOMMU group 2) except a single Usb 3.0 on Bus 4 --> 0000:00:14.0 (IOMMU group 2) Bus 3 and 4 are on the same controller, the same iommu group and have the same id. And bus 1 and 2 are unreachable for me because there are only onboard headers? So my problem is I can't passthrough the controller because my unraid stick is on it too right ? Bus 004 Device 002: ID 0781:5583 SanDisk Corp. If more informations are needed please let me know. I can't find a Layout description about my Motherboard. But I use 2 onboard USB 2.0 Headers one for a LED and one for All IN One Corsair Watercooling. Maybe thats the reason why the USB Ports are all on the same Controller? Really don't know. Would it be better to unplug the frontusb from the header? Or switch the onboard header from the AIO Cooling? I Read here that maybe it has something todo with the BIOS USB Settings xhci ehci legacy, but I already changed the settings. Don't really know what to set here to get the effect I want. That would be more than enough for my uses but i have only 4 settings in bios under the usb config: 1.disable usb controller off 2.xhci hand-off tested both enabled and disabled no effect 3.ehci hand-off tested both enabled and disabled no effect 4.usb legacy is set to auto , if I disable it my unraid stick won't be recognized. In Bios it tell me I have 2 usb root hubs. I don´t get it what did msi do with this motherboard? Would be very happy about every help and every hint. Thanks very much. Best wishes feraay

same here. s3_sleep version: 3.0.5 after reboot status is stopped. After re-save it shows running till next reboot. first thought it´s because of my random wake up script time=09:55 now=$(date +%s) other=$(date -d $time +%s) dayofweek=$(date +%u) if [ $now -ge $other ] && [ $dayofweek -lt 5 ] then echo `date '+%s' --date='tomorrow 09:55:00'` > /sys/class/rtc/rtc0/wakealarm elif [ $dayofweek -ge 5 ] then echo `date '+%s' --date='next monday 09:55:00'` > /sys/class/rtc/rtc0/wakealarm else echo `date '+%s' --date='today 09:55:00'` > /sys/class/rtc/rtc0/wakealarm fi but tested without it and same result. p.s this bad syntax is the only one that worked for me.

This Script should do it. Just paste it into S3 Sleep settings in the custom command box (post sleep) Be sure you disabled resume by rtc alarm in bios and set the bios time to rtc time. It will wakeup every day at 09.55. So you can set jobs for 10.00. time=09:55 now=$(date +%s) other=$(date -d $time +%s) if [ $now -ge $other ] then echo `date '+%s' --date='tomorrow 09:55:00'` > /sys/class/rtc/rtc0/wakealarm else echo `date '+%s' --date='today 09:55:00'` > /sys/class/rtc/rtc0/wakealarm fi Monday to Friday exclude Saturday and Sunday: time=09:55 now=$(date +%s) other=$(date -d $time +%s) dayofweek=$(date +%u) if [ $now -ge $other ] && [ $dayofweek -lt 5 ] then echo `date '+%s' --date='tomorrow 09:55:00'` > /sys/class/rtc/rtc0/wakealarm elif [ $dayofweek -ge 5 ] then echo `date '+%s' --date='next monday 09:55:00'` > /sys/class/rtc/rtc0/wakealarm else echo `date '+%s' --date='today 09:55:00'` > /sys/class/rtc/rtc0/wakealarm fi

This Script should do it. Just paste it into S3 Sleep settings in the custom command box (post sleep) It will wakeup every day at 09.55.( Monday to Friday) So you can set jobs for 10.00. Be sure you disabled resume by rtc alarm in bios and set the bios time to rtc time. time=09:55 now=$(date +%s) other=$(date -d $time +%s) dayofweek=$(date +%u) if [ $now -ge $other ] && [ $dayofweek -lt 5 ] then echo `date '+%s' --date='tomorrow 09:55:00'` > /sys/class/rtc/rtc0/wakealarm elif [ $dayofweek -ge 5 ] then echo `date '+%s' --date='next monday 09:55:00'` > /sys/class/rtc/rtc0/wakealarm else echo `date '+%s' --date='today 09:55:00'` > /sys/class/rtc/rtc0/wakealarm fi