Re: How to repeat alert notification

[Riccardo Leonardi]

Hi,

unfortunately my case is different because the monit process running onto mailserver to be monitored "talk" with a centralized m/monit.
My fault, I didn't specified clearly before, sorry.
In fact is the notification of the failure event generated by monit to m/monit that is NOT repeated.

In short, the syntax you reported would do the trick:

if status != 0 then alert repeat every 10 cycles

but unfortunately it's not accepted by monit:

# /usr/local/monit/bin/monit -t /usr/local/monit/conf/monitrc
/usr/local/monit/conf/monitrc:51: syntax error 'repeat'

Alert notification from monit to m/monit is what I want repeat every x cycles until failure status is cleared.

Il 26/01/2017 17:55, Guillaume François ha scritto:

Hi 

In my case, I use a custom script to alert and Notify at the same time like below

if does not exist then exec "/var/lib/monit/scripts/notifyAndExecute.sh /bin/mount /path/to/remount" else if succeeded then exec "/var/lib/monit/scripts/notifySuccess.sh"

This way I can alert by using what ever channel (email;,sms, webhook etc...) and run command at the same time. 

I did this because in the the past monit didn't supported several rules but Monit evolved and you can now defined several rules for one event like below (in the past it was overwriting/ignored)

if does not exist for 3 cycles then alert

if does not exist for 5 cycles then start 

if failed port 8080 for 3 times within 4 cycles then alert

if failed port 8080 for 6 times within 8 cycles then restart

if failed port 8443 for 3 times within 4 cycles then alert

if failed port 8443 for 6 times within 8 cycles then restart

if failed port 9990 for 3 times within 4 cycles then alert

if failed port 9990 for 6 times within 8 cycles then restart

if does not exist for 10 cycles then alert

if does not exist for 5 cycles then exec "/bin/mount /mnt/nfs"

I don't have samples, but from my memory, you can also define several actions for the same "event" several times with newer version like below

if does not exist for 5 cycles then alert repeat every 2 cycles

if does not exist for 5 cycles then exec "/bin/mount /mnt/nfs" repeat every 3 cycles

if does not exist for 5 cycles then exec "/some/script/notify" repeat every 3 cycles

In your case it would be something like (not tested)

if status != 0 then exec "/usr/local/monit/conf/dummy.sh" repeat every 10 cycles

if status != 0 then alert repeat every 10 cycles

Hope this help.

2017-01-26 17:16 GMT+01:00 Riccardo Leonardi <address@hidden>:

Hi,

thank you for your answer.
I've checked the syntax about repeat feature, and it works (I have 5.20.0).
Unfortunately what I need is not the script execution: that's is ok.
What I need is repeat of alert notification, and as far as I know this notification is sent when the state changes only: OK->Failure and Failure->OK.
As workaround I could explore some configuration at server side or forging some explicit signal by the script that is run every x cycles (see dummy.sh in my example), but I don't know how to do it.

Thansk again

Il 26/01/2017 16:56, Guillaume François ha scritto:

Hi,

Syntax is ok https://mmonit.com/monit/documentation/monit.html#ACTION

Could it be an issue related to the version of Monit you're using ?

Changelog (https://mmonit.com/monit/changes/) shows the behavior was changed in 5.16.0 as before it was forced to "repeat every 1 cycle".

Therefore this syntax was not supported before.

Best Regards.

2017-01-26 16:27 GMT+01:00 Riccardo Leonardi <address@hidden>:

Hi,

I have a script checking mail queue returning alert if queue lenght is greather than some threshold.

The alert is correctly forwarded to slack channel, but when the threshold is reached only, not when the queue remains over the threshold.

In fact the behaviour I'd like to implement is:
- receive the alert when the queue goes over the threshold
- continue receiving the alert (possibly with current queue lenght) every x cycles
- stop receiving when the queue returns under the threshold

Currently I have implemented the following:

check program qlen_check with path "/usr/local/monit/conf/check_qlen.sh 200"
    if status != 0 then exec "/usr/local/monit/conf/dummy.sh" repeat every 10 cycles

The script check_qlen.sh echoes queue lenght and return 1 if queue > 200.
As soon as that happens I got 1 slack alert only, while I'd like to keep receiving it every x cycles (when dummy.sh is correctly run).

Does anyone have any suggestion how to achieve that behaviour?

Thank you in advance.

Rik

--
To unsubscribe:
https://lists.nongnu.org/mailman/listinfo/monit-general

--

-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS/IT d(+) s++:- a C++$ ULC(+)>+++$ !P--- L+>$ !E---? W+++$ !N* !o-- K--? w(+) !O---? !M- !V--? PS+? !PE Y+ PGP++>+++ !t-- !5 !X- R(+)>++* tv-? b(-) DI !D- G(+)>+ e+++ h--() r->$ y?*
------END GEEK CODE BLOCK------

--
To unsubscribe:
https://lists.nongnu.org/mailman/listinfo/monit-general

-- To unsubscribe: https://lists.nongnu.org/mailman/listinfo/monit-general

--

-----BEGIN GEEK CODE BLOCK----- Version: 3.1 GCS/IT d(+) s++:- a C++$ ULC(+)>+++$ !P--- L+>$ !E---? W+++$ !N* !o-- K--? w(+) !O---? !M- !V--? PS+? !PE Y+ PGP++>+++ !t-- !5 !X- R(+)>++* tv-? b(-) DI !D- G(+)>+ e+++ h--() r->$ y?* ------END GEEK CODE BLOCK------

--
To unsubscribe:
https://lists.nongnu.org/mailman/listinfo/monit-general