at least 3.0 health checks are buggy

There was a recent change in health_check.sh with 5757b99cabd3e2d57231d11558d03530b9b3ff48

With a sematic which is not backward compatible but happened from 3.0.2-1 to 3.0.2-2 or 3.0.2-3.

http_publish_uri is not optional anymore and has to be set in the graylog configuration. The grayloog configuration states that this property is optional!

#### HTTP publish URI
#
# The HTTP URI of this Graylog node which is used to communicate with the other Graylog nodes in the cluster and by all
# clients using the Graylog web interface.
#
# The URI will be published in the cluster discovery APIs, so that other Graylog nodes will be able to find and connect to this Graylog node.
#
# This configuration setting has to be used if this Graylog node is available on another network interface than $http_bind_address,
# for example if the machine has multiple network interfaces or is behind a NAT gateway.
#
# If $http_bind_address contains a wildcard IPv4 address (0.0.0.0), the first non-loopback IPv4 address of this machine will be used.
# This configuration setting *must not* contain a wildcard address!
#
# Default: http://$http_bind_address/

This lets the health check fail with exit code 1 and the same log message and exit code (which is a bit useless when not looking into the check itself :/

So a few ideas/todos

make http_publish_uri again (as it is stated in the configuration documentation)
fix the health checks - better error messages/exit codes to decide what is wrong(btw where do this echos even go to?)
currently check_url gets overwritten in the check, this looks at least falsy or i don't get it.
the last exit 1 (when the curl fails) should also be distinguishable from the others
republish 3.0.2-1 if it ever was on docker hub, there is no tag with 3.0.x where this is not buggy

What did I do to fix my setup: added http_publish_uri = 127.0.0.1:9000 to the configuration

Graylog2 / graylog-docker

at least 3.0 health checks are buggy #100