[BUG]Pod kb-addon-prometheus-server in CrashLoopBackOff because too many open files

[ahlfors]

Describe the bug
Pod kb-addon-prometheus-server in CrashLoopBackOff because too many open files

To Reproduce
Steps to reproduce the behavior:

1. Install kbcli and init playground;
2. Create mysql cluster;
3. Keep the cluster running for a week;
4. Check pod kb-addon-prometheus-server status
5. See error:
   Pod kb-addon-prometheus-server in CrashLoopBackOff
   Defaulted container "prometheus-server-configmap-reload" out of: prometheus-server-configmap-reload, prometheus-server
   2023/03/27 11:27:28 too many open files

Expected behavior
Pod kb-addon-prometheus-server in Running status

Desktop (please complete the following information):

OS: [Linux 5.10.0-21-cloud-amd64 #1 SMP Debian 5.10.162-1 (2023-01-21) x86_64 GNU/Linux]

Additional context
prometheus-server CrashLoopBackOff
#15742

[ahlfors]

This issue exist on gcp vm, because its default process file size limits is too small 1024, while aws ec2 default process file size limits is 65535, ok for this case. User need to adjust the vm process file size limits before running kubeblocks. Close issue.