How can I get real-time health of the backend servers and possibly statistics, such as number of connections for each backend listener for load balancing in PRTG for HAProxy?

After adding my HAProxy instance into PRTG, I get the standard Ping, CPU load, memory, eth0 sensors. I cannot find any way, after searching, to get HAProxy statistics, only the host statistics.

I would like have any failed host checks from HAProxy to be logged in PRTG, and notified if the service, or backend host is down. Any ideas?

Ubuntu Server 12.04 LTS HAProxy 1.4.18

Thanks, Timm

Article Comments

Hi,
there is no native sensor in PRTG but you might check the HA Proxy documentation if there is some kind of SNMP integration. If so, a MIB file should be offered as well. The MIB file could then be imported to PRTG using the MIB Importer.
Best regards


Jan, 2014 - Permalink

You may wish to look into this. It appears that someone has scripted out monitoring haproxy with a PowerShell script:

http://lazic.info/josip/post/monitor-haproxy-via-prtg/


Jun, 2015 - Permalink

A new open source SNMP agent and MIB are now available for HAProxy. Loadbalancer.org have a blog with detailed instructions for importing the MIBs to PRTG:

https://www.loadbalancer.org/blog/loadbalancer-org-releases-open-source-snmp-mib-and-agent-for-haproxy/


Sep, 2020 - Permalink