Reverse proxy timeouts with Check_MK

In some distributed environments and with a lot of configured users and connection to Microsoft active directory configured in Check_MK times to distribute the changes from the master to all slave sites grow up strongly. Check_MK distributed monitoring uses Apache web server acting as reverse proxy. Sometimes one of Apache web server’s instances doesn’t return data within a pre-defined time range. In such cases you can get the error like this while trying to replicate your slave with the master site:

HTTP Error - 502: 

502 Proxy Error

Proxy Error
The proxy server received an invalid
response from an upstream server.
The proxy server could not handle the request POST /your_slave_site/check_mk/automation.py.
Reason: Error reading from remote server

In this case the following change helped me. I edited the file /omd/sites/your_slave_site/etc/apache/proxy-port.conf and changed the line that contains the directive ProxyPass from

ProxyPass http://127.0.0.1:5000/your_slave_site retry=0 disablereuse=On

to

ProxyPass http://127.0.0.1:5000/your_slave_site retry=0 disablereuse=On acquire=10000 timeout=600 Keepalive=On ttl=120 connectiontimeout=10

After I saved the edited file I had to re-start the main Apache web server by executing the following command:

service httpd restart

If you have more slaves, you have to repeat the procedure for each of them.

3 thoughts on “Reverse proxy timeouts with Check_MK

  1. Hank

    hi, do you know how to get the popup graph working on pnp4nagios for services that’s being monitored on the remote site without using the rsync method?

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.