Open metronidazole opened 3 years ago
Doing some further testing I have discovered that connman was the cause (i.e. fucking with my network configuration).
It seems to be installed by default when using the suggested image in the readme (under RPiOS64 autoinstall). It should probably be removed or disabled in the auto setup script.
I have now updated the installer script, to work with the new latest image https://downloads.raspberrypi.org/raspios_lite_arm64/images/raspios_lite_arm64-2021-11-08/
Test runs appriciated, =) works for me.
Hi, I am assuming should refer to the raw file here?
curl https://raw.githubusercontent.com/TuxfeatMac/pimox7/master/RPiOS64-IA-Install.sh > RPiOS64-IA-Install.sh
Also, I get the following result (however, I didn't run this in a Screen session; would that make a difference?)
...
Installing new version of config file /etc/default/networking ...
network config changes have been detected for ifupdown2 compatibility.
Saved in /etc/network/interfaces.new for hot-apply or next reboot.
Reloading network config on first install
client_loop: send disconnect: Broken pipe
I am using the 2021-10-30-raspios-bullseye-arm64 image
Thanks @chadai I've updated the comment!
this file will be delete by my script: /etc/network/interfaces.new <-------- Not any more, it will automaticaly configure the vmbr0
Have you chosen the same IP address, your pi got on firt boot up via dhcp, as the static IP in the script?
Thanks again @chadai for time and your try! For me there was no issue.
I've updated the installer script. Network changes will be applied at reboot, preventing the ssh connection getting closed during installation.
I am having a strange issue after setting this up. I followed the readme and used @TuxfeatMac 's script to update to bullseye and install pimox7. That all worked initially, but I wanted to change my networking to use a bridge interface so I could hook my testing containers/VMs into the bridge. However, after configuring this and rebooting I lost network connectivity to the pimox server (UI and via SSH, ping, etc). I have seen #13 which seems related.
After doing some testing, it seems that for some reason, the real physical network interface gains a local IP (169.254.36.32) and a second default route is created using that within a minute of networking starting (using ifdown and ifup). After resetting the network, I can use it until this second route is made. See the following debug log for further explanation.
Any ideas on what is causing this? I assume it's probably some rogue debian script but I'm not sure how to proceed. Can I intercept calls to route?