Flavio Crisciani
Flavio Crisciani
Do you guys have some logs to share? would be super helpful to have a way to reproduce and grab logs with the engine in debug mode Engine in debug:...
@mpepping thanks, the error message is clear. The current suspect that I have is a race condition during the deletion of the sandbox that leak the vxlan interface behind it....
I'm also already trying to reproduce it locally, but if you guys narrow down a specific set of steps that are able to reproduce with high probability let me know
@tfasz @antoinetran this behavior got fixed by https://github.com/docker/libnetwork/pull/1792 will be available on 17.06 and also will be backported to 17.03
@antoinetran so the weird part is that the ping to the service name is answered locally, there is an iptables rule that reply to the ping to the VIPs directly...
@antoinetran I was thinking that C1 was still not ready, but if you can exec into it should not be the case. Do you see anything unusual in the daemon...
@antoinetran I see memberlist complaining about one node and marking it as suspect, this means that the physical network is failing delivering the health checks. if the networkdb is not...
@antoinetran in 17.12, I added several logs that you can use to check the status of the networkDB when you see weird behavior. First is: `NetworkDB stats`, it will print...
@abhinandanpb can you take a look?