KnightRdr23

Intel lost me for a while. They’d have to do something totally spectacular to win me back at this point. My 9900K is fantastic for my Unraid but their new stuff is crap compared to what AMD is doing gaming wise. My advice, don’t wait! Prices only going to continue getting worse. I built in September last year and it’s amazing how much the same components are even now.

I went 9950x3d for the extra cores and future flexibility. It performs very nicely in games and will not bottleneck a 5070 Ti at all. In my rig the 5070 Ti is the bottleneck lol I’ve also found that while I did vm gaming for years, there are still definite advantages to running native. I stream the PC using Apollo/Artemis (sunshine/moonlight fork) to an Nvidia Shield connected to my 83” 4K TV in the living room. There was always latency doing this with a VM. Native is much smoother

Honestly I ended up keeping my old machine as Unraid server and turned the new machine into a straight gaming PC. Aside from the gaming VM my old rig is perfectly fine running as an Unraid server. Maybe one day I might change that but for now running two machines was the better option for me

Yes I did. Went with the MSI Tomahawk 870E. Decided to go big for the extra PCIE lanes for potential future expansion. AMD is very limited in this space

Thank you for the insight. If Intel hadn’t gone to crap the last couple generations I’d stick with them but AMD is just so far ahead of them right now.

I'm looking to upgrade my Unraid server which I also use for Windows 11 VM gaming. Currently have an Intel 9900K with 64GB and an ASUS PRIME Z-370a motherboard. I'm interested in the AMD 9950X3D because of the core/thread count to split with my VM and Unraid but unsure of what motherboard to get. Like I see most of these AM5 motherboards only have a couple PCIe slots. I would need one for a GPU passthrough (likely going to upgrade to an nvidia 5000 series), one for my LSI controller for 8 SATA drives off one card and then also a separate NIC adapter so I can pass through to VM without sharing with Unraid (it introduces too much latency when I use Sunlight/Moonlight for streaming). I'd also like to get the benefit of PCIe 5.0 cache drives. I've found some that have 3 PCIe slots but all sorts of notes that if you use more than one NVMe port, the main PCIe slot runs at x8 instead of x16 which for PCIe 5.0 may not be horrible. I've seen benchmarks that even the fastest GPUs still work fine at even PCIe 3.0 x16 speeds. I believe the LSI card is an x8 card though that could be a challenge. Any recommendations on a motherboard? Thanks!

I was able to resolve this by stopping docker, changing ipvlan to macvlan, saving then reverting and saving, and also changing host access to custom networks to disabled, saving then enabled, and saving again. Finally re-enabled docker and seem to be working again for now

Custom network, ipvlan

Realtek 2.5G card for Unraid, pihole setup with its own IP on the same subnet as my other devices. Other containers use shared networking

Bump for help please

I am using the DNS option to set my docker containers to use pihole also running in docker as their DNS. This has worked for a long time until 7.1.2. Since booting after 7.1.2, my docker containers cannot talk to pi-hole but pi-hole is working and other devices can access it. Help? nicknas2-diagnostics-20250515-2140.zip

So I am using a RAM risk as my transcode cache and gave it 24G which is fine for about 95% of what I'm doing. Where I seem to have an issue is that if it has to reorder the streams first, it makes a copy of the original file in transcode cache then proceeds to transcode it so between the reordered original file and then the transcoding, some files are too big to fit in the ramdisk this way. Is there any way to tell Tdarr to replace the existing file on disk with the reordered file and then proceed to transcode it so it doesn't use up so much memory? Or could I even get more fancy that if a file is larger than X, use a different directory for transcode cache?

Glad to hear it works for you. Curious if anyone from the Unraid team could comment on this. Minor bug but appears to be a bug nonetheless. Thanks

Darn, I'm usually pretty good about staying updated. I never got any notification that 7.0.1 was available. Also, when looking in my /var/log directory in terminal, the space is being taken up by the cache directory plugin logs. I disabled logging for that plugin and will install 7.0.1 now Thanks for fast response!

Got an alert from the common problems plugin that /var/log is at 89%. Server only had 34 days uptime and never saw this before. Looks like the dhcpd log lines filling up the log. Is there a fix? nicknas2-diagnostics-20250309-1244.zip

bump

I am noticing an issue with this and audio transcoding. I am playing an Atmos stream on Apple TV so it transcodes audio to FLAC. Video was direct play so nothing to really convert. I have 16GB tmpfs setup for Plex to transcode in RAM. When I start playing the media the usage goes up very fast until it hits 16GB and then stops playing with a transcoder error on the client. Once it crashes the transcode folder drops to 0 used. I tried to throttle the transcoding to 30 seconds ahead but that didn't even help. I have no idea what is using up all the storage in the transcode directly from simply transcoding audio. Any ideas?

I was experimenting with hugepages for a Windows gaming VM. Looks like Unraid defaults to 2M but if you want to set 1G, it doesn't seem to work unless you explicitly set a new default. Doesn't work: hugepagesz=1G hugepages=25 grep Huge /proc/meminfo AnonHugePages: 464896 kB ShmemHugePages: 0 kB FileHugePages: 0 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB Hugetlb: 26214400 kB However, if I put this in the append, it does work default_hugepagesz=1G hugepagesz=1G hugepages=25 grep Huge /proc/meminfo AnonHugePages: 131072 kB ShmemHugePages: 0 kB FileHugePages: 0 kB HugePages_Total: 25 HugePages_Free: 24 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 1048576 kB Hugetlb: 26214400 kB Wondering if that's a bug somewhere because I would think hugepagesz by itself should let you set the size to override the kernel default but it only appears to work if default_hugepagesz is present too.

Unplugged all 3 and rebooted and still seeing that IPv6 log spam every 2-3 mins I also have 2 Apple TV if that’s relevant

I do, 3 in fact

Nope haven't enabled that feature

Thanks, anything I should do with the IPv6 log lines? I am using native IPv6 on my router.

Had an interesting issue when I woke up this morning. My office sounded like a jet engine coming from my Unraid tower. I logged into the GUI and all looked normal, temps were low. I did leave my gaming VM running last night but that's never been a problem. The VM was definitely frozen and I think the fan noise was coming from the GPU. Power usage was no higher than normal so it appeared the fan was just going full blast for nothing. I could not shut down the VM and had to force stop it. That did not stop the fan noise so eventually I had to reboot. I did capture diagnostics before rebooting though in hopes someone here could diagnose what happened. Would greatly appreciate the help! Thanks nicknas2-diagnostics-20250121-1018.zip

virt-xml-validate /etc/libvirt/qemu/Windows\ 11.xml /etc/libvirt/qemu/Windows 11.xml:12: namespace warning : xmlns: URI unraid is not absolute <vmtemplate xmlns="unraid" name="Windows 11" icon="windows11.png" os="window ^ /etc/libvirt/qemu/Windows 11.xml validates