I'm a nagios user, but I'm having problems to find an "equivalent" solution to my elastic AWS opsworks stacks.
I mean, machines will be created, terminated, stopped, launched or for some reason down.<p>I'm trying to use check_mk WATO api to configure a "scheduled down time" when a machine is stopped for example. Or even using the opsworks time based instances to avoid false alarms. I also would like to start monitoring new instances, and stop when the instance is terminated.<p>The hosts are running docker containers started by fig/docker-compose. Each fig.yml has its own 'monitor' container which is configured to monitor all the containers running on that host. This way I can monitor normal things on the host (cpu, disk, load, etc) and also only one HTTP check to my monitor container.<p>This is configured using the "custom json" opsworks and chef.<p>I saw that there are a whole new world about monitoring out there (prometheus, boson, shinken, etc ) and the SaaS like boundary, datalog etc.<p>My primary concern is to alert sysadmin guys when an http service (the monitor container) from one of the hosts returns something diferrent of HTTP 200 OK status code, for example.<p>Which tool or service would you guys suggest me ?<p>Thanks
For HTTP monitoring, check out <a href="https://www.runscope.com/docs/radar" rel="nofollow">https://www.runscope.com/docs/radar</a> (disclosure: I'm a founder)