Alexstrasza

Members
  • Posts

    58
  • Joined

  • Last visited

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

Alexstrasza's Achievements

Rookie

Rookie (2/14)

6

Reputation

  1. I think you've muddled your 🦈 and 🛡️'s 😉
  2. Ah... have you made sure that "Host access to custom networks" is set to "Enabled" in your UnRaid docker settings?
  3. Are you running the latest version of the container available (:latest)? If so our environments should be identical. The only thing I had to change to get mine working was turning IPv6 forwarding on for the host, which you've done too.
  4. Strange. Can you check cat /proc/sys/net/ipv6/conf/all/forwarding from the UnRaid console? I know it should return 1 based on your results above, but it's worth double checking.
  5. That's a pretty strange issue, that's not the error you should get if it's a SMBv1 problem from what I know, the error is normally much more specific. Have you tried posting a general thread in https://forums.unraid.net/forum/55-general-support/? They'll be much better prepared to help you there, this is more of a support thread for Tailscale-specific issues. Unfortunately I think if your non-Tailscale file transfers don't work properly, they are unlikely to work within Tailscale, as the method is exactly the same.
  6. Hi there, no, no more complete panics since then. I've had quite a lot of general instability issues, the X470D4U motherboard seems to be plagued with issues, in particular reliability varies hugely BIOS to BIOS, so if you're running that motherboard then I heavily recommend the P3.50 BIOS firmware and 2.20.00 BMC firmware, which have been the most stable I've seen. It's also worth noting that future UnRaid versions have also improved other issues a lot, so I heavily recommend checking you are on the latest version as well.
  7. There was a bug with older versions of QEMU where the host-passthrough CPU mode would not work. This is fixed in later versions of UnRaid. There's this really understated warning in the motherboard manual which has caught me out before and might be your problem too: The setting isn't called exactly the same thing as the manual says either (so that's helpful), but you need to manually switch the slots into 2x8 mode for either to work properly with two GPUs installed on a Ryzen 3000 series CPU.
  8. Correct on all three counts! However there is a bit more nuance to it. Whilst Wireguard can be used with a kernel implementation (which I believe is more efficient, so less CPU usage) it can also be implemented in software. Tailscale at the moment exclusively uses the software implementation to ease cross-platform compatibility, although there are plans in the future to link in with the kernel system on systems with support. This means it's technically not speaking with the system implementation at all at the moment. As for compatibility in general, as far as I'm aware any number of systems can use the underlying Wireguard technology, as long as they don't use conflicting address spaces (this is true with any VPN afaik and in my experience). Since Tailscale uses the rare 100. address range, it's incredibly unlikely to conflict with anything else provided you haven't manually specified that same range for the Unraid Wireguard tunnels.
  9. They can both run independently, so feel free to try it out!
  10. Good to hear! I'm glad I'm not going insane :). Unfortunately I still can't see the button because of another error: key [iManufacturer] doesn't exist Sounds like it's the same thing again but for this key 😅. I wonder what it's trying to read that lacks these values?
  11. Hi John, Am I going insane, or did there used to be a "Benchmark all" button or something to that extent? The only 'multiple benchmark' button I can find is under the controller info, and that seems to just check for a controller bottleneck. Also, when I look at the "USB Bus Tree" page, I see this error. Thanks for all your work!