alexgaribay / libcluster_gae

MIT License
24 stars 1 forks source link

Problem with libcluster, unable to connect #2

Closed jtormey closed 4 years ago

jtormey commented 5 years ago

After following the README instructions, deployed instances are unable to connect to each other. The logs show warnings like this every 10s (the default polling interval):

2019-11-10 19:57:15 default[20191110t125255]  19:57:15.990 [warn] [libcluster:my_app] unable to connect to :"my_app@aef-default-20191110t125255-r2jt.c.my-project.internal"

I've set VmDnsSetting=GlobalOnly in Cloud Console > Compute Engine > Metadata, which did not seem to change anything.

What's strange is that now with -name configured in vm.args, I'm not able to attach to a running instance of my app from inside its container:

justin@aef-default-20191110t125255-r2jt ~ $ docker exec -it 17a5a0e7dbd1 /bin/bash
root@17a5a0e7dbd1:/app# ./bin/my_app remote_console
Could not contact remote node my_app@aef-default-20191110t125255-r2jt.c.my-project.internal, reason: :nodedown. Aborting...

However, the application is definitely running, it's accessible and can be interacted with. It seems like I'm missing something when it comes to configuring the Erlang VM, or maybe with configuring GAE. Any help would be appreciated!

alexgaribay commented 4 years ago

Duplicate of #1.