Issues when starting coordinator in ArangoDB using the Starter

Question

I am trying to create a cluster with three servers in Arango, by following the procedure stated in:

https://github.com/kvahed/arangodb-cluster-admin-course/blob/master/part-one.md

This give me some issues, when the Starter tries to start a coordinator (on all servers). I get the following error after running the arangodb command:

Error while starting coordinator: Cannot start coordinator, because port 8529 is already in use.

I guess this is just because Arango after installation is already listening on this port. (Agents and Dbservers are started without problems). So I stop this service by service arangodb stop and run the command arangodb again. This just results in the following error:

Failed to get master URL, retrying in 5sec (All 3 attemps resulted in temporary failure)

which is an issue I have seen some other people also experienced, when shutting down Arango.

My guess is that there is some simple configuration I am missing. I just can not figure it out.

I am running this on Ubuntu 16.04 servers on AWS.

Answer 1

this is just a port clash. If you would like to use this installation for a cluster, please stop and disable the single instance service like so:

sudo systemctl stop arangodb3
sudo systemctl disable arangodb3

After that your coordinator on port 8529 should do fine.
Cheers.