May-28th-2019, 11:39 AM
(May-22nd-2019, 02:18 PM)Armin@netPI Wrote: Hello Andreas,
the point is as follows. The reason why you are getting access to portainer.io via "cifx0" but in the end this portainer.io not to DockerHub is the following:
If you remove the "eth0" or "wlan0" interface physically then the Host Linux will remove the DNS server from its local "/etc/resolve.conf" file.
If now portainer.io wants to request an image from "registry-1.docker.io" then the Host Linux cannot resolve this name and give you back an error like
Code:Failure
Get https://registry-1.docker.io/v2/: dial tcp: lookup registry-1.docker.io on 127.0.0.1:53: server misbehaving
And even if a container is started in Host and priviledge mode, it cannot access the file "/etc/resolve.conf" on the host, else you could easily insert a new DNS server entry like "nameserver 8.8.8.8" for example.
Armin
Hello Armin,
I did insert an nameserver 8.8.8.8 in /etc/resolv.conf in the host mode container (not the host system). This was the way to get DNS working in the container. Is there a way to get this line in the host systems /etc/resolv.conf?
My solution was now to additionally install Node-Red in the host mode container manually like described here https://nodered.org/docs/hardware/raspberrypi
The new issue now is that it seems that the NetPi crashes completely after a couple of hours of operation. This varies between 3 hours and 12 hours of operation. Last time when I observed that I could not connect to the container via ssh and I could not reach the Edge Gateway Manager web interface and got no ping to the device. Where should I start to investigate?
Thank you for your Help,
Andreas
Andreas