Problem extending the

william(at)elan.net william at elan.net
Thu Dec 14 18:45:36 CET 2006


Note - size increase for this patch is 2k, I increased if further manually.
And if you use it go through every line - it is internal patch I use to
increase several other limits as well (but some such as MAX_INPUT_BUFFER 
are needed to get larger MAX_PLUGIN_OUTPUT to work) and it also changes 
path in at least one place.

On Thu, 14 Dec 2006, william(at)elan.net wrote:

> If you're interested in increasing buffers, here is patch:
>  http://www.elan.net/~william/nagios/nagios25-largerbuffers.patch
> [warnings about it being experimental and use at your own risk apply]
>
> Regarding 512k FIFO limit, while its good to be safe and I've no
> problem with it, it really has been quite some time (over 10 years)
> when 512k was stopping factor; unfortunately however what are FIFO
> buffers is badly documented on unix systems you likely to use so
> doing it above 4k (even above 1k) like Ethan says is just not safe -
> do experiments on your own if you like to be adventurous though...
> Best I can tell you is that I got 3k plugin output with that patch
> on linux rhel3 & linux suse10 systems.
>
> On Thu, 14 Dec 2006, Nicolas COLIN wrote:
>
>>> Ethan Galstad <nagios <at> nagios.org> writes:
>>>
>>> The problem you'll run into is due to a limitation with FIFOs in
>>> general.  The POSIX standard only guarantees an atomic write of 512
>>> bytes max to a FIFO, so writing more data than 512 bytes at a time could
>>> land you in trouble.  AFAIK, most OSes bump this level up a bit, but the
>>> max buffer size on a FIFO is often limited to 4K, so your total write
>>> size can't really be above that level.
>>>
>>> Length limitations for plugin output will disappear with Nagios 3.x.
>>> There is a new IPC method for handling active host/service checks that
>>> allows for long plugin output (its capped at 4K for safety's sake) and
>>> there is a new external command that will allow passive checks to have
>>> longer (and multiline) output as well.  The new PROCESS_EXTERNAL_FILE
>>> command will tell Nagios to read other external commands/passive checks
>>> results from an external (flat) file.  Regular files don't have line
>>> length limitations, so this will be a useable workaround.
>>>
>>> So, your the problem you have will disappear once Nagios 3 is released,
>>> but it will still be a problem for Nagios 1/2.    Hope that helps.
>>>
>>> Ethan Galstad,
>>> Nagios Developer
>>> ---
>>> Email: nagios <at> nagios.org
>>> Website: http://www.nagios.org
>>
>> Hi,
>>
>> Thanks for you answer. It helps.
>>
>> I tried different values but get stuck to low. I am stopping investigating on
>> this issue.
>>
>> I will impatiently wait for the v3. By any chance, do you have an idea of the
>> release date or even of the first beta date ?
>>
>> In the mean time, I am thinking of using Nagios with MySQL (instead of default
>> files) which will allow me to update the database directly. Do you think this
>> could be effective ?
>>
>> Thanks again.
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys - and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Nagios-devel mailing list
> Nagios-devel at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/nagios-devel

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV




More information about the Developers mailing list