snidera

Just so you're aware, I did about the same thing last week - 6.9.2 -> 6.10rc3 in order to get the AMD reset plugin to work & it did just fine. I have another issue passing through that card that will require more testing on different hardware, but windows 10 & 11 rebooted just fine with the AMD reset plugin installed.

Exactly, you can use the GPU only in a VM (Guest) or on Unraid (Host). I would recommend that you are bind the card to VFIO in your System Devices page, reboot and try it again. Please clarify: Is it suggested to *not* have "AMD Vendor Reset" and "Radeon Top" plugins installed together for a GPU that will be passed through to VM? I realize Radeon Top isn't needed, but will it cause issues once VFIO grabs the GPU?

VM is Q35-6.1 If it resets 50% of reboots, I'm fine with that. Previously, I had downloaders & media library management on VM, so it was a pain to find out it had tried to update & been shut down for days due to no reboot. AMD Vendor Reset, Radeon Top, Intel GPU Top & Intel GVT-g are all installed. Plex docker can use iGPU to transcode. Yes, but it has still paused 1-2 times since. Just after typing the previous, UnRaid stopped responding & I had to hard reset. I think the Sata3_0 port causing my issues with the cache & possibly the VM, crossing fingers. It is shared with a m.2 & this MB has always had some odd issues. I moved the cache to the last SATA port, other controller. We'll see I haven't tried with this hardware, but previously my other server would not put console on iGPU & let me pass through the RX460. If the parity check goes through tonight, I will test with another monitor

I'm trying to use a RX460 that I had semi-working previously with a win10 VM. It had the reset bug & would not reboot without a power cycle. I grabbed a GT1030 & solved that problem. I have a fresh win10 VM running & updated via VNC. Added the GPU, mouse & keyboard after it was running & stable for a day or 2. It took a couple of tries & I'm not sure what I changed to make it work, but it is now working & will survive windows reboot with GPU. However, sometime between 4PM 3/14 & 11AM 3/15, the VM went into a 'paused' mode & will not resume. I have to force stop, then restart. No GPU, but it boots to windows via VNC. If I power cycle the server, it will come back up. I can't find the error in logs (but I'm not an expert by any means). This is my backup unraid server, the MB has always been a little wonky, the cache keep throwing UDMA errors. I will try a different port as time permits. Do you have any suggestions? Up to & including starting from scratch on this server, I'd rather not re-copy 7.5TB, but that's not out of the question. I've read this thread & googled 'reset bug' a lot (as well as in the past when I gave up grabbed an nVidia). The only thing I haven't figured out is how to use the iGPU as console. Monitor connected to GPU displays console during boot until the card is grabbed by VFIO. I've always had blank from iGPU when passing through anything. ich777, Thanks in advance for any support you can provide. Also, thanks for making this possible & relatively painless - it's really great work. backup-diagnostics-20220315-1137.zip

I have UnRAID & a Win10 VM that has very slow write speed from the vdisk to any shares, even with cache turned on on the share. Vdisk is also on the cache (500GB Samsung EVO) I only get 35-40MB/s max, but it ebbs & flows from max all the way down to stalling out completely. Transfers usually look like rounded hills, ramping up, then fading to 0MB/s, then repeat. Sometimes it will start out at 150MB/s or more, but drops off rapidly. Is this par for the course, or do I have something setup incorrectly? Would adding a 2nd SSD for the vdisk fix this? (should I pass it through or use unassigned devices?) I followed gridrunners' setup suggestions pretty much to the letter & this is the 2nd or 3rd VM that has had this issue, I'm just now getting annoyed with it enough to do something about it.

I followed the Plex instructions & it worked - going from Win7 to LSIO docker. Basically copy the library folders & paste into the new server. Some of the 'recently added' got a little out of order, but all my watched/unwatched transferred over, which was what I really wanted.

YGBSM. I tried everything I could think of to get the damn thing to work before posting here so I wouldn't feel dumb when someone pointed out the answer. And you suggest, correctly, that I need to use electrical tape. I guess duct tape (or electrical tape) and zip ties make the world go 'round. At least I work for an automotive wire harness company - toughest choice was what color tape to grab out of the ~10k rolls in my junk room. I chose gray fire-resistant - it sticks to damn near anything. Many thanks. after 2 months of fighting me, this thing is finally falling into line.

I have 2x M1115 cards which work just like M1015 (used same flashing instructions/files). When I install in new server it won't POST. Asus Maximus VIII Hero I5-6500 4x4GB RAM 500GB SSD (SSung EVO) 2x 6TB Toshiba 1x 4TB WD Red RX460 passthrough to win10VM Unraid 6.3.5 Plex Docker, community apps are the only 2 other things running Computer has been rock-solid, other than this issue & some VM oddities. -MB LED Q-Code stops at 16 - 'Pre-memory system agent is started' -If I clear CMOS, it stops at 55 - 'memory not installed' after stopping for ~45sec at B1 - Runtime Set Virtual Address MAP End -If I re-enable VT-x & VT-d (must haves) after clear CMOS, it goes back to stopping at 16 -Both cards boot/function normally in the Asrock 970extrem4 MB that the 1 came out of. -I JUST flashed the 2nd card into IT mode (I've had it for 3 years & not used it) It didn't post before or after flashing P20 firmware. -The used card had P15 firmware and I suspected it had the boot ROM, so I flashed it to P20 w/o boot rom. after doing so, I suspect that I didn't have the boot ROM anyway. -I've flashed the newest MB firmware (3504 / 6/29/2017) -I've tried the card in all slots X16/x8/x4 - won't be re-trying the x16 after installing the RX460, but I don't think that's the issue. -I've changed all the PCIe options I can find to use Gen2 speeds (rather than Gen3) For now, I have plenty of SATA ports on the MB - I want to pass the M1115 through to win10 & have DVD/BluRay drives for ripping. Any ideas? I haven't found any BIOS options that look like they'd deal with the pre-memory issues.

I have: Stubbed out all 5 PCI addresses in syslinux checked both the USB3 & SATA in the VM config page BR drive shows up in win 10, Device manager recognizes proper DR Model, recognizes a disk (DVD & CD audio), I can see the files in explorer, but I cannot PLAY anything. MakeMKV says it cannot find an optical disk it can use. Google says this controller doesn't play nice with Optical drives. bought this & a USB/SATA adapter that both don't work. geeesh.

OK, so it was fully hung up. shutdown -h now via ssh failed, so I did what I should have done first thing - held the power button down. Now it somewhat works - USB & SATA controller show up in the 'edit' page of the VM. I was going to make a change to overwrite the xml... SATA BR Drive shows up in windows, but I can't read data from the disk - the folder structure works, but I can't access files - do I need vertio drivers? ETA - thanks Bob, thought I'd try comma first, but hadn't killed power yet.

I want to add Bluray drives to my Win10 VM for ripping. Tried a couple methods that didn't work, so I bought a PCI-E x1 card that has 4x USB3.0 & 2x SATA. I want to pass through the whole controller. Currently, I've tried the <hostdev> method from here Simpler / Easier PCI Device Pass Through for NON-GPUs There were 5 devices & I was suspect about not seperating IOMMU groups, but I tried it anyway - now Unraid webgui is unresponsive (chrome shows 'waiting for 192....' I can SSH in & SMB shares are still functioning. Going to reboot using ssh. I have a monitor hooked to it, it was at the login prompt before, now it has scrolled through some output ending with a 'stop trace....' statement. I can't decipher any of it, but it was definitely related. IOMMU groups /PCI info that I want to pass through (group 12/13/14 - these were new after installing the card & I believe are all that are new) I already have a vfio-pci.ids entry in my go file(to get the audio to pass through with iGPU), how do I separate multiples? I've never seen an example with multiple entries to know whether it's a space, comma, etc. IOMMU group 0 [8086:191f] 00:00.0 Host bridge: Intel Corporation Skylake Host Bridge/DRAM Registers (rev 07) IOMMU group 1 [8086:1912] 00:02.0 VGA compatible controller: Intel Corporation HD Graphics 530 (rev 06) IOMMU group 2 [8086:a12f] 00:14.0 USB controller: Intel Corporation Sunrise Point-H USB 3.0 xHCI Controller (rev 31) [8086:a131] 00:14.2 Signal processing controller: Intel Corporation Sunrise Point-H Thermal subsystem (rev 31) IOMMU group 3 [8086:a13a] 00:16.0 Communication controller: Intel Corporation Sunrise Point-H CSME HECI #1 (rev 31) IOMMU group 4 [8086:a102] 00:17.0 SATA controller: Intel Corporation Sunrise Point-H SATA controller [AHCI mode] (rev 31) IOMMU group 5 [8086:a169] 00:1b.0 PCI bridge: Intel Corporation Sunrise Point-H PCI Root Port #19 (rev f1) IOMMU group 6 [8086:a16a] 00:1b.3 PCI bridge: Intel Corporation Sunrise Point-H PCI Root Port #20 (rev f1) IOMMU group 7 [8086:a112] 00:1c.0 PCI bridge: Intel Corporation Sunrise Point-H PCI Express Root Port #3 (rev f1) IOMMU group 8 [8086:a114] 00:1c.4 PCI bridge: Intel Corporation Sunrise Point-H PCI Express Root Port #5 (rev f1) IOMMU group 9 [8086:a118] 00:1d.0 PCI bridge: Intel Corporation Sunrise Point-H PCI Express Root Port #9 (rev f1) IOMMU group 10 [8086:a144] 00:1f.0 ISA bridge: Intel Corporation Sunrise Point-H LPC Controller (rev 31) [8086:a121] 00:1f.2 Memory controller: Intel Corporation Sunrise Point-H PMC (rev 31) [8086:a170] 00:1f.3 Audio device: Intel Corporation Sunrise Point-H HD Audio (rev 31) [8086:a123] 00:1f.4 SMBus: Intel Corporation Sunrise Point-H SMBus (rev 31) IOMMU group 11 [8086:15b8] 00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection (2) I219-V (rev 31) IOMMU group 12 [1b21:1182] 02:00.0 PCI bridge: ASMedia Technology Inc. Device 1182 IOMMU group 13 [1b21:1182] 03:03.0 PCI bridge: ASMedia Technology Inc. Device 1182 [1106:3483] 04:00.0 USB controller: VIA Technologies, Inc. VL805 USB 3.0 Host Controller (rev 01) IOMMU group 14 [1b21:1182] 03:07.0 PCI bridge: ASMedia Technology Inc. Device 1182 [1b21:0612] 05:00.0 SATA controller: ASMedia Technology Inc. ASM1062 Serial ATA Controller (rev 01) IOMMU group 15 [1000:0072] 07:00.0 Serial Attached SCSI controller: LSI Logic / Symbios Logic SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] (rev 03) I added this to the (functioning) win10 xml. <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x02' slot='0x00' function='0x0'/> </source> </hostdev> <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x03' slot='0x03' function='0x0'/> </source> </hostdev> <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x04' slot='0x00' function='0x0'/> </source> </hostdev> <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x03' slot='0x07' function='0x0'/> </source> </hostdev> <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x05' slot='0x00' function='0x0'/> </source> </hostdev>

Well, 22hrs & 4TB of data transferred later, it looks like a crappy powerstrip/surge protector might have been the issue. Either that or my server knows I made this thread & is screwing with me.

I checked logs when it happened last time & couldn't see any trace. I thought they were cleared on reboot (I'm a bit dumb in this area). Attached are from after a hard reset. tower-diagnostics-20171004-1930.zip

I'm having an issue where UnRAID becomes unresponsive, as well as my entire network loses connectivity. Ping unresponsive on multiple (probably all) devices connected to the router. Remove network cable from UnRAID box & everything goes back to normal instantly (or as fast as I can run over to a PC & check). Plug it back in, everything fails. yellow LED on UnRAID ethernet adapter is flashing steadily/slowly - green is activity, it is solid. Even after removing the ethernet cable, UnRAID seems frozen. Power button doesn't respond, have to cut power (hold button or unplug AC) This happens ~4+ hours after boot, has happened 3 times in about 12-15hrs. It happened 1x before the 3 week win10 run, but I chalked it up to lightning/power blip + it was on a different power supply (same power supply is currently in use & stable on another system) Same HW/SW config was steady for 8+ days on a different router/location (Asus RTn-10) Same MB/proc/RAM was steady for 3+ weeks running Win10 off a HDD (thrashing CPU with handbrake) killed my win10 VM before it happened this last time - don't think that is affecting it. This last time I was copying files from my other server (win7), the HDD activity LED is solid on. The times before, server was idle & HDD LED was off. Archer C7 3150 UnRAID 6.3.5 Asrock H170M Pro4 i5-6500 (intel cooler) 16GB RAM 2x 6TB Toshiba's 1TB WD blue (Cache) HAF912 case - fans running in 'performance' mode. HDDs never get above 35* Corsair TX550m No IP conflicts on network - I have set static DHCP for all devices & place devices in ranges for easy QOS control I thought about power (AC) - but the 3+ weeks was on the same surge protector, along with another PC that never missed a beat (using 2 win PCs to compress DVD/BRs via handbrake). I now have it plugged into UPS to try to rule out dirty AC. Suggestions to diagnose? MB was used/ebay, it's my prime suspect - suggestions on diagnosis/fixes? All other components are new. But again, stable with windows & for 8 days on another router Router? How do you get ethernet wrong? This router has been great for 8 months when my other one got struck by lightning. UnRaid - this USB has been stable & in use for over a year? I wiped it, downloaded 6.3.5 & set this server up fresh to take over main duties. Thought 8+ days of burn in was enough. RAM? Ran memtest for a while, not my first suspect CPU? Same as RAM - it's been pushed via handbrake & was stable Heat? doubt it, but never know.

I'm finally replacing my "server" (Win7+Flexraid) with Unraid. My intention was to do this 6-7 years ago when my friend & I split a 2-key license. His is still running 4.7 rock solid. Primary Uses: File server for Movies (BluRay/DVD/DL), TV (.mkv) & music Plex Server (probably docker) Win10 VM (MakeMKV, MediaCenterMaster, General surfing/office duties, light fusion360 CAD use) Future: OwnCloud/nextcloud/backup Couchpotato Headphones We'll see how it performs & add uses as CPU/stability permit. I bought Asrock H170M Pro4 i5-6500 16GB RAM 2x 6TB Toshiba's Have - 1TB WD blue (currently cache) After Compressing my DVD/BluRays to H265 & finding that 3TB+ of space was being 'eaten' by flexraid not emptying the supposedly non-existent recycle bin, I now have ~3.5TB of data (thought I was close to 11....) I plan to re-use a newer 4TB WD Red so I should have plenty of space for the time being. Once data is copied off the old server I'll also have available: 2x BR drives (ripping) RX 460 M1115 (same as M1015 SAS adapter, flashed to IT mode) - I'm out of SATA ports with my planned uses, so this will be going in even if not utilized right away. Pros/cons of splitting drives off of on-board SATA to the HBA? 240GB SSD - got some age on it, so I'm leary of beating on it. 2x4TB, 1x3TB, 2x2TB drives (probably going to make an off-site backup/plex server at a location I have MUCH faster internet speed). Questions: 1. Cache: 1TB WD Blue (7200rpm) or 240GB SSD? Both are used. My test built had a 60GB cache which died after light use (but was 5yrs old & died during a lightning storm, so I can't say UnRAID killed it). 2. VM drive - is there a large advantage to using SSD for vhd? (using unassigned devices) vs on the cache (HDD or SSD). I haven't been able to tell a difference in limited light use. 3. CPU cores. I only have 4 - how should I split them among VM & UnRAID. I'll rarely transcode even 1 stream, but don't want to choke unraid/plex (FireTV/firesticks direct play currently). I'm impatient with windows, would like to have horsepower when needed.... 4. Video for VM. I've passed through the iGPU for testing - it was a bit of a pain. I think I'd rather have the RX460 for the Win10 VM. Once Plex adds gpu transcoding, how hard is it to get igpu working with the docker? searching around says it's possible with the special 1.8 build, but it didn't get into 1.9 main & I stopped at 'you need to compile a new kernel'. 5. Anybody successfully moved from Plex on windows to Plex on unraid & kept the library intact? I've done it win->win. I have almost 1500 movies, hate to lose my watched/unwatched flags. 6. USB hubs - any suggestions/tips for making USB ports available to the VM as needed (thumb drives, USB HDDs, etc)