<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> <HTML> <HEAD> <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=ISO-8859-1"> <META NAME="Generator" CONTENT="MS Exchange Server version 5.5.2658.24"> <TITLE>RE: [Nagios-users] Nagios memory Leaks</TITLE> </HEAD> <BODY> -----Original Message----- From: nagios-users-bounces@lists.sourceforge.net [<A HREF="mailto:nagios-users-bounces@lists.sourceforge.net">mailto:nagios-users-bounces@lists.sourceforge.net</A>]On Behalf Of Stanley.Hopcroft@Dest.gov.au Sent: 24 January 2007 02:07 To: nagios-users@lists.sourceforge.net Subject: [Nagios-users] Nagios memory Leaks Dear Sir, I am writing to thank you for your valuable letter and say, > From: Tobias Klausmann <klausman@schwarzvogel.de> > Subject: [Nagios-users] Memory leaks > > Hi! > > (First off: if this should also go to nagios-devel, just yell at > me.) > I don't think so because it deals with the aspects of the implementation that are visible (and in fact, the letter doesn't propose detailed solutions). > Nagios 2.6 and 2.5 have memory leaks. They are not that big that > within hours your machine will be swapping, but they degrade > performance in other ways. > > First off, their approximate extent. > > 2.5 and 2.6 without perl cache have the smallest memory leaks. A > fairly busy Nagios server (hardware quoted below) with about 3000 > services on about 330 hosts will degrade from 330M used (that's > *not* Nagios alone) to 368M used in about 16 hours. Or about 2.4 > MB per hour. The very same machine behaves neutral if Nagios is > not running, so it's definitely Nagios itself. Do you mean: 2.5 and 2.6 Nagios with embedded Perl but without the Perl plugin cache option ? If so, the fault is not Nagios, but the embedded Perl implementation and or Perl. Your next paragraph suggest that this is plain vanilla Nagios without any Perl options to configure. Is that correct ? > > Activating the embedded Perl interpreter and -cache will increase > the amount of lost memory to about 5-6M per hour. In this case, > however, sometimes the memory usage snaps back, i.e. some of the > lost memory is collected. I've not yet found out what triggers > the reclaim. Still, over the course of hours, more and more > memory is lost. Still, it's roughly linear memory loss. > I have never witnessed memory being reclaimed after ePN leaks it. I can't conceive of the process memory size being reduced while the process is running (free() and friends only return the memory to the process heap). I think the leak is caused by the ePN implementation. I a hoping to trying some measurements with several pilot implementations to see what is the most promising way of doing this. ... (snip) Yep. I agree. The leak is bad. > The question that remains is, if this can (and will) be tackled > before 3.0 is released. A related question is if Nagios 3 will be > prone to the same problem. > Certainly it will if the current ePN implementation remains. If (pretty big if) I can provide you stuff to try are you willing to repeat your measurements on candidate implementations (wrt 2.5 or 2.6 code base) ? I am not sure of my willingness/energy quotient but if they look Ok, I may not have anything to show until March this year. > Any thoughts, ideas etc. are appreciated. > > Regards, > Tobias Yours sincerely. Morning !! I have been reading with interest about these memory leaks. I see you mention 2.5 & 2.6. Does this happen with 2.4 as well ?? Thanks !!! John </BODY> </HTML>