Description of problem:

If collectd has been configured and is active it will continue to send data to the corresponding endpoint and thats what causes the issue as there's no endpoint listening and the queue keeps growing. And all of the memory will be used up

Similar BZ are and it seems that it still there in RHOSP 17.1.2

https://bugzilla.redhat.com/show_bug.cgi?id=1817124
https://bugzilla.redhat.com/show_bug.cgi?id=1798617
https://bugzilla.redhat.com/show_bug.cgi?id=1790928
https://access.redhat.com/solutions/4855731

It can be reproduced in Lab
~~~

1. podman stats collectd
   ID NAME CPU % MEM USAGE / LIMIT MEM % NET IO BLOCK IO PIDS CPU TIME AVG CPU %
   f61051856726 collectd 1.98% 53.41MB / 33.37GB 0.16% 0B / 0B 327.7kB / 81.92kB 39 13h12m51.737656s 1.98%

1. free -g
   total used free shared buff/cache available
   Mem: 31 8 13 0 10 23

=> 16.2

1. podman stats collectd
   ID NAME CPU % MEM USAGE / LIMIT MEM % NET IO BLOCK IO PIDS
   ad91dff0cdea collectd – 70.93MB / 536.9MB 13.21% – / – 1.839MB / 519MB 38

1. free -g
   total used free shared buff/cache available
   Mem: 31 12 3 0 15 18
   Swap: 0 0 0

Need to understand why in 17.x environment the limit of 500 MB is not configured or applicable
~~~
On the other hand, setting the memory limit does not work.

~~~

1. get the ID of the collect container
   id=$(podman ps | grep collectd | awk ' {print $1}

   ')

1. setting need to change swap to update mem
   podman update --memory-swap 300m $id

1. setting max memory to 100 mb
   podman update m 100m $id <-- adjust to the correct value needed if more than 100m

1. validating the new setting
   podman inspect $id | jq '.[].HostConfig.Memory'
   ~~~

~~~
parameter_defaults:
ExtraConfig:
collectd::write_queue_limit_high: 100
collectd::write_queue_limit_low: 100
~~~

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1.
2.
3.

Actual results:

Expected results:

Additional info: