Re: how does monit every cycle statement works?

[Jiguang Bai]

Is the 'cycles' have the same meaning for RESTART LIMIT?

Say the daemon is set to 1min, does the following def alerts when the process restarts 5 times in a row within 10 mins?

check process pro with matching "pro"

every 2 cycles

if 5 restarts within 5 cycles then alert

Best regards,

Jiguang

On Sat, Aug 13, 2016 at 12:39 AM, Martin Pala <address@hidden> wrote:

These statements are equal (see monit manual for more details: https://mmonit.com/monit/documentation/monit.html#FAULT-TOLERANCE):

#short format:

if status !=0 for 2 cycles then alert

# long format:

if status !=0 for 2 times within 2 cycles then alert

The long format ("X within Y cycles") is useful if X != Y, which you can achieve using short format, for example:

if status !=0 for 3 times within 5 cycles then alert

If X == Y, you can use short or long format - they work the same.

Best regards,

Martin

On 11 Aug 2016, at 16:57, Vikram Kone <address@hidden> wrote:

I'm a bit confused. According to the issue page you linked to, what I have originally is the right thing.

if status !=0 for 2 cycles then alert

But you are saying, this is right

if status !=0 for 2 times within 2 cycles then alert

Aren't both of them same?  Can you explain the difference?

From what I understand,  the way i can read the first setting is 

if status !=0 for 2 attempts then alert

Is that wrong?

On Aug 11, 2016 12:50 AM, "Martin Pala" <address@hidden> wrote:

yes

On 11 Aug 2016, at 08:54, Vikram Kone <address@hidden> wrote:

check program test_pgm with path "/bin/bash /var/run/test.sh"

with timeout 3600 seconds
every 2 cycles
if status !=0 for 2 cycles then alert

In this case let's say test.sh takes 45 mins to finish (either success or failure) each time. Will the above config, still holds correct ie, will it alert if the pgm failed twice consecutively?

On Wed, Aug 10, 2016 at 11:51 PM, Vikram Kone <address@hidden> wrote:

and how does timeout comes into play here

On Wed, Aug 10, 2016 at 11:38 PM, Vikram Kone <address@hidden> wrote:

Do you mean this one?

if status !=0 for 2 times within 2 cycles then alert

On Wed, Aug 10, 2016 at 11:06 PM, Martin Pala <address@hidden> wrote:

The second configuration is correct.

The combination of "every" with "X cycles" is confusing, we'll modify the syntax: https://bitbucket.org/tildeslash/monit/issues/174/the-for-x-cycles-is-confusing-if-the-test

Best regards,

Martin

On 11 Aug 2016, at 04:48, Vikram Kone <address@hidden> wrote:

I have a monit check program as follows

check program test_pgm with path "/bin/bash /var/run/test.sh"
every 2 cycles
if status !=0 for 2 cycles then alert

Let's say my monit daemon is set to run every 60 seconds which means each cycle is 1 minute. I want to run my program every 2mins and alert if it fails twice in a row

Is what I have above correct? or should I say

if status !=0 for 4 cycles then alert

or should it be

if status !=0 for 2 times within 2 cycles then alert

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

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

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

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

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

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