CMK version: 2.3.0p43
OS version: Monitoring Site Ubuntu 24.04 / Central Site 20.04
We are encountering the following problem: we can no longer perform service discovery management on hosts with a large number of services (approx. 2,800 services). The hosts are using the Push-Agent in the same version (2.3.0p43). When attempting a service discovery via the GUI (e.g., “Accept all”), we receive the following error:
Failed to fetch state [500]: INTERNAL SERVER ERROR. Retrying in 1 second. In case this error persists for more than some seconds, please verify that all processes of the site are running.
I have already adjusted parameters in apache.conf:
-
Timeout 600 -
LimitRequestBody 0
And in /opt/omd/apache/site.conf:
ProxyPass … timeout=600
Unfortunately, these changes did not resolve the issue. On the central Checkmk site, the web.log shows the following traceback:
Python
File "/omd/sites/zentrale/lib/python3.12/site-packages/werkzeug/formparser.py", line 242, in parse
return parse_func(stream, mimetype, content_length, options)
File "/omd/sites/zentrale/lib/python3.12/site-packages/werkzeug/formparser.py", line 282, in _parse_urlencoded
raise RequestEntityTooLarge()
werkzeug.exceptions.RequestEntityTooLarge: 413 Request Entity Too Large: The data value transmitted exceeds the capacity limit.
Is the traceback in the web.log also related to the Service Discovery Problem?
What needs to be optimized so that larger requests are processed?
I also tried accepting only a single service via the GUI, but the error still occurs.
I tried also on the monitoring site:
cmk -II destinationhost
cmk -R
cmk -O
Did also not lead to accepted services.
Somebody a idea? Many thanks in advance!
