For scenarios when you are spinning up various services at once on boot, registrator will error out if the consul backend is unavailable, and does not retry. This may be that Consul has not started yet (split second behind etc).
It would be nice if there was a (potentially configurable) amount of retries and retry interval/s (seconds). I am solving this in Bash for now, but would be nice if it was natively supported :)
Example:
2015/10/08 00:11:16 Starting registrator v6 ...
2015/10/08 00:11:16 Forcing host IP to 10.20.0.117
2015/10/08 00:11:16 consul: Get http://172.17.42.1:8500/v1/status/leader: dial tcp 172.17.42.1:8500: connection refused
For scenarios when you are spinning up various services at once on boot, registrator will error out if the consul backend is unavailable, and does not retry. This may be that Consul has not started yet (split second behind etc).
It would be nice if there was a (potentially configurable) amount of retries and retry interval/s (seconds). I am solving this in Bash for now, but would be nice if it was natively supported :)
Example: