PUSH mechanism in Check_mk agent

Hello All,

We are having Check_mk RAW version 1.5.0p21, having some issues where active connections on port 6556 is not allowed from check_mk server.

Is there any way to push all check_mk agent data to check_mk server from client to server ? also let me know if any capability to recognize this check_mk agent data at check_mk server end.

Regards,
Dinesh

See Can we monitor a host without an active connection

Basically you need to setup a cronjob on the client that calls the check_mk_agent every minute, transfer that output to the checkmk server and store it in a file and have the server do a cat that_file instead of connecting to the client on port 6556.

1 Like

cat /path/to/file is basically the simplest datasource program you can write.

https://docs.checkmk.com/latest/en/datasource_programs.html has the information. Checkmk just needs the agent output. How this is transported to the monitoring server does not matter.

Yes. That’s how it appears from the Powerpoint perspective. :wink:

If you take a closer look then you’ll notice some nasty details and questions:

  • How do you make sure that no files get cat'ed that are only partially transmitted?
  • How do you recognise a host being down?
  • Does the file’s age matter? When is a file too old?
  • What about missing files? Is the client down or just the connection in-between?
  • Should the DS program delete the file after catting it?
  • Can the server cope with 500 incoming ssh/ftp/whatever connections from 500 clients?

I’m not writing against reversing the communication. I’m just mentioning the questions that might pop up when doing so.

1 Like

Thanks @r.sander @Dirk for your valuable response.

1 Like

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact @fayepal if you think this should be re-opened.