<tt>nagios-users-bounces@lists.sourceforge.net schrieb am 02.09.2008 19:12:01: </tt> <tt>Hello Ryan,</tt> <tt>first of all: please to not top-post and do a fullquote at the same</tt> <tt>time. It's considered rude on mailing lists and it's unnecessary</tt> <tt>waste of bandwidth and makes mails hard to read, as I do not know</tt> <tt>if you posted between the lines too.</tt> <tt>Either please don't quote at all, or post your replies between the</tt> <tt>quotes, like I do usually for example. Thanks for your understanding.</tt> <tt> > It seems that for every host, 3 processes are launched to do the > host ping check: sh, ping, and nagios. I currently have ~57 hosts > that are in an down state and have been acknowledged as out of > service. I would assume 57*3 plus the 30 second timeout could cause > this many processes at the same time.</tt> <tt>Ok, that seems like a possible explanation. If you have a lot of checks,</tt> <tt>which are expected to time out, then you will have a lot of "hanging"</tt> <tt>processes waiting to quit. There is really nothing you can do about</tt> <tt>it, but lower your timeout value.</tt> <tt>Another option is to refrain from using check_ping, but use check_icmp</tt> <tt>instead. check_ping is a wrapper script, which calls /bin/ping and</tt> <tt>parses its output. Therefor you have the extra "sh" fork. check_icmp</tt> <tt>is a plugin with native icmp implementation and doesn't require</tt> <tt>the extra fork. It's faster, less bulky and todays check of choice</tt> <tt>for host checks - at least for intranet hosts in my opinion.</tt> <tt>WAN hosts are another thing though... but that might be my personal</tt> <tt>opinion only.</tt> <tt>> I guess that brings me to my next question. I could disable active > host checks for these out of service machines which would most > likely alleviate my warnings about the amount of processes, but > would I have to re-enable them once the machines are brought back > up? I currently just acknowledge the problem and leave a comment > when a machine is put out of service, but this means that it will be > back at some point. When it does come back, acknowledgement is gone > and regular checks are still happening. Does anyone know of a > better way to do this?</tt> <tt>There is really no way to solve that "problem". It is like it is.</tt> <tt>Question is rather: how can it be, that you have 57 hosts, which</tt> <tt>are down for an extended period? I do not know your environment, but</tt> <tt>it seems questionable to me, why such hosts would deserve a monitoring.</tt> <tt>But that's up to you to decide.</tt> <tt>Regards</tt> <tt> Sascha</tt> -- Sascha Runschke IT-Infrastruktur GFKL Financial Services AG Limbecker Platz 1 45127 Essen Telefon : +49 (201) 102-1879 Mobil : +49 (173) 5419665 Fax : +49 (201) 102-1102105 GFKL Financial Services AG Vorstand: Dr. Peter Jänsch (Vors.), Jürgen Baltes, Dr. Till Ergenzinger, Dr. Tom Haverkamp Vorsitzender des Aufsichtsrats: Dr. Georg F. Thoma Sitz: Limbecker Platz 1, 45127 Essen, Amtsgericht Essen, HRB 13522