monit-general
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Is monit deterministic?


From: Tino Hendricks
Subject: Re: Is monit deterministic?
Date: Fri, 23 Dec 2022 11:11:49 +0100

Hi Leonard,

I think I can answer 2) 
As per
https://mmonit.com/monit/documentation/monit.html#SERVICE-POLL-TIME
you have several options. The "cron style" might come closest to your need.

Happy Christmas days to all!

Tino


Am 23.12.2022 um 11:00 schrieb Leonard Teng <leonard@ackcio.com>:

Hi all, thanks for answering my question previously. I have 2 further questions:

        1) Are there currently any implementations/solutions for auto renewal of client certificates when they become invalid or expired?  
        2) Is monit able to perform different polling cycles for different processes organically? Currently it seems like I can only specify 1 value for the poll cycle length, but there are processes that I want to check less frequently than others (every 1 minute vs daily for example). I know that it's possible to have multiple users on a system run multiple instances with their own monitrc files, but I don't want to run into the complexity of managing different access levels for users. 

Thanks in advance!

On Fri, Dec 23, 2022 at 7:34 AM Lutz Mader <lutz.mader@freenet.de> wrote:
Hello Leonard Teng,
from my point of view monit is not the problem sometimes.

> I just wanted to inquire on whether monit is guaranteed to perform service
> restarts in high CPU/RAM usage situations?

From the monit point of view, monit can handle low free RAM or RAM
shortage, but restarts can not handle sometimes. I find "out of memory"
conditions on Linux and AIX systems with a RAM shortage all the time
monit try to start a command to try a restart or anything else.

> Wanted to know if I need to be
> conservative in setting thresholds for CPU/RAM usage for services.

You should use conservative thresholds.

But it depence to your systems,
Lutz

Appendage:
On AIX systems I use cpu and memory threadholds up to 95%.

check system $HOST
  every 5 cycles
  if cpu usage > 95% for 10 cycles then alert
  if memory usage > 95% for 20 cycles then alert
  if swap usage > 50% for 20 cycles then alert
  if swap usage > 30% for 5 cycles then exec
"/usr/local/etc/monit/scripts/aixpgsp.sh 500"
     repeat every 20 cycles



reply via email to

[Prev in Thread] Current Thread [Next in Thread]