Re: [Nagios-devel] [Fwd: Hundreds of passive checks in a second]
Posted: Mon Sep 06, 2004 8:04 pm
Hi Sean,
see if you can find the following msg id: [email protected]
if you can't find the message, have a look at
http://svn.webperf.org/WebSVN/listing.p ... ev=26&sc=1
it might provide you a hint on what it was doing.
the jist of the patch was to reduce the sleep time, and to immediatly
re-check the named pipe for
a new command before going on to something else.
On Mon, 06 Sep 2004 16:33:34 -0400, Sean Dilda wrote:
> On Mon, 2004-09-06 at 16:03, Andy Mayhew wrote:
> > Moving the named pipe off the file system and into ram (via a ramdisk) usually
> > resolves processing issues with handling hundreds (or thousands) of incoming
> > passive checks a second.
>
> This doesn't make much sense to me. As I understand it, although you
> connect to a named pipe through the filesystem, the actual buffering for
> the named pipe is all handled by the kernel in RAM. As such, I'm not
> sure how moving the named pipe to ram would increase the buffer size.
> Have you tried this before?
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1 today.
> http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
> _______________________________________________
> Nagios-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/lis ... gios-devel
>
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]
see if you can find the following msg id: [email protected]
if you can't find the message, have a look at
http://svn.webperf.org/WebSVN/listing.p ... ev=26&sc=1
it might provide you a hint on what it was doing.
the jist of the patch was to reduce the sleep time, and to immediatly
re-check the named pipe for
a new command before going on to something else.
On Mon, 06 Sep 2004 16:33:34 -0400, Sean Dilda wrote:
> On Mon, 2004-09-06 at 16:03, Andy Mayhew wrote:
> > Moving the named pipe off the file system and into ram (via a ramdisk) usually
> > resolves processing issues with handling hundreds (or thousands) of incoming
> > passive checks a second.
>
> This doesn't make much sense to me. As I understand it, although you
> connect to a named pipe through the filesystem, the actual buffering for
> the named pipe is all handled by the kernel in RAM. As such, I'm not
> sure how moving the named pipe to ram would increase the buffer size.
> Have you tried this before?
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1 today.
> http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
> _______________________________________________
> Nagios-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/lis ... gios-devel
>
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]