notification_interval check vs. check_period interval check - config.c

[Marcel]

Now i'm getting confused too 8|

On Mon, Sep 15, 2008 at 5:13 PM, Marcel <webknowledge at gmail.com> wrote:

>
> On Sat, Sep 13, 2008 at 5:22 AM, Hendrik BŠäcker <andurin at process-zero.de>wrote:
>
>> Marcel schrieb:
>> >     If you have a service that would be checked every 10 seconds, and
>> >     notification interval = 60, then you should get the warning message.
>> >
>> > I think a bit and that would not be the case, but instead, not 10
>> > seconds, but like 300 (or five minutes if interval_length=1). Then the
>> > warning is issued.
>>
>> So, actual I'm a little bit confused about what is the fact.
>>
>> You have service object with check_interval = 300 (seconds, cause
>> interval_length=1) and a notification_interval of 10 (seconds) or vice
>> versa?
>
>
> No, i have notification_interval's default value of 60 (minutes if
> interval_length=60), but when using itnerval_length=1 and
> normal_check_interval of 300 (seconds), then the warning message saying that
> my notification interval is lower than my command_check_interval appears at
> nagios.log
>
> check_interval = 300 +
>> notification_interval = 10
>> The warning message is IMHO OK cause it's useless to resend notification
>> cause of a service state that isn't able to change.
>
>
> That's what Nagios thinks aswell.
>
> Please drop us a short note what you are doing exactly.
>>
>
> Notification_interval never gets multiplied by interval_length (nor
> divided), so it assumes interval_length = 60.
>
> Documentation says notification_interval is "time units", but never gets
> checked against interval_length, as command_check_interval is in line:
>    1344                 command_check_interval*=interval_length;
>
> notification_interval never gets multiplied by interval_length, at least at
> config.c, then when using the default notification_interval of 60, and
> setting a 300s normal_check_interval (five minutes, because
> interval_length=1), then the sanity check at line 1878 prints the wrong
> warning message.
>
> My solution was to multiply (not assigning
> notification_interval*=interval_length), in config.c at check time, but that
> is wrong because it did not assign notification_interval*=interval_length,
> as command_check_interval is.
>
> Well, i did get rid of the wrong warning messages, but i think
> notifications would be messed up (i did not get into notifications.c to
> confirm anything).
>
> Anyone seeing messed up notifications with interval_length < 60 ?
>
> Regards,
> Marcel
>

I've realized some mistakes when trying to understanding what I've said :)

I was assuming a wrong premise of the notification_interval type. All the
time I was referring to notification_interval as not being
interval_length-based, but it is, so my whole assumption of the bug is
wrong.

But, notification_interval did not gets multiplied by interval_length as
command_check_interval is. So, from that point, if command_check_interval
and notification_interval are all the same 'time units', then i would expect
notification_interval being assigned like command_check_interval
(command_check_interval*=interval_length).

Cheers,
Marcel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.monitoring-lists.org/archive/developers/attachments/20080915/da43dc5d/attachment.html>
-------------- next part --------------
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
-------------- next part --------------
_______________________________________________
Nagios-devel mailing list
Nagios-devel at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-devel