Vagrant install, latest version, rancher-server host disconnection

Hi,

I had noticed this earlier on yesterdays version at the office, and now again with the new 0.50 version…

I did a clone and vagrant up, and everything seemed ok… it put 2 VMs up, rancher-01 and rancher-server… Sicne I didnt have my docker/rancher-compose files handy, I started toying around with the catalog and was installking and ELK stack with logspout to play around… then while I thought it was downloading stuff and getting containers started, I noticed rancher-server appeared “disconnected”… Deactivating/reactivating is now got it in a “Activating…in progress” state for the better part of an hour… While all this is happening, however, the rancher UI is accessible…

Accessing the machine with cagrant ssh rancher-server, and then looking into the rancher-agent container’s logs yields: http://pastebin.com/TM6SB42j

bunch of connection error… Quite weird…? this is just a brand new git clone of the rancher repo, and trying to launch some stuff from the catalog…

fyi: I had a similar behavior today on a completely different machine, with version 0.49 where hosts just went into a disconnected state…I must be doing something wrong?

After 1 night online, woke up to the rancher-01 disconnected…

Something seems def amiss? It cant be only me lol…

I’ll try another computer, just a simple vagrant up from cloned rancher/rancher repo…

No one else having any issues?

Whenever I try to launch something its bringing everything down now… Tried re-pulling, deleting the VMs from vagrant and vbox, after a while the interface shows both rancher-01 and rancher-server offline, and eventually even the web interface stops… Can someone point me in what direction I should go to get correct logs to debug this? I can get to the console on the vms…Its not showing anything abnormal… gonna try to login and see if I find something

(cant login using rancher/rancher so I’m looking for a local cred if it exists… anyways I’m pulling 50.1 to see if I have better luck… :confused:)

It seems something quite strange is going on…

going in by vagrant-ssh into the machine, I can access the api locally using 127.0.0.1, but not using the “machine ip” 172.19.8.100…

I’m pretty sure now its something vagrant related where I’m losing access to the vbox interface… Apparently restarting the itnerface should fix it but its not working for me… I’m going to look further to the vbox side of things… if anyone has any ideas I’m open :slight_smile: its weird how this happens on both my work machine as well as my laptop tho…

I think it may be similar/same as https://github.com/rancher/rancher/issues/2196