Mon Debugging
I hacked up the standard ping.monitor to send heartbeats to our in-house monitoring front-end upon a successful ping. We started having problems recently with mon not getting the heartbeats out in a timely manner. I turned on the debug option and started digging through the log. I noticed that usually, once per second, there's an entry in the log that's just a number, and it increments by one...usually every second... usually. So, on to my question: Is the counter in the debug log SUPPOSED to count up once per second? If it doesn't, is that a symptom of too much load on the system? # tail -f /var/log/mon | egrep 'mon\[[0-9]+\]: [0-9]+ ' Jul 14 12:58:03 ops-inet-mon mon[3430]: 222348Jul 14 12:58:05 ops-inet-mon mon[3430]: 222349Jul 14 12:58:06 ops-inet-mon mon[3430]: 222350Jul 14 12:58:09 ops-inet-mon mon[3430]: 222351Jul 14 12:58:10 ops-inet-mon mon[3430]: 222352Jul 14 12:58:13 ops-inet-mon mon[3430]: 222353Jul 14 12:58:16 ops-inet-mon mon[3430]: 222354Jul 14 12:58:20 ops-inet-mon mon[3430]: 222355Jul 14 12:58:24 ops-inet-mon mon[3430]: 222356Jul 14 12:58:29 ops-inet-mon mon[3430]: 222357Jul 14 12:58:33 ops-inet-mon mon[3430]: 222358 Sorry if this has been covered, but I couldn't find anything in the mailing list or man page... ___ mon mailing list mon@linux.kernel.org http://linux.kernel.org/mailman/listinfo/mon
RE: Mon Debugging
[EMAIL PROTECTED] mon]# mon -v $Id: mon,v 1.4.2.14 2004/11/19 18:27:47 trockij Exp $ $Name: mon-1-0-0pre5 $ So, for future troubleshooting efforts, the debug counter *is* supposed to update once a second? I will try upgrading to 1.1.0pre1. Thanks for the info. -Original Message- From: David Nolan [mailto:[EMAIL PROTECTED] Sent: Thursday, July 14, 2005 4:27 PM To: Aaron Segura; mon@linux.kernel.org Subject: Re: Mon Debugging --On Thursday, July 14, 2005 12:23 PM -0500 Aaron Segura <[EMAIL PROTECTED]> wrote: > I hacked up the standard ping.monitor to send heartbeats to our in-house > monitoring front-end upon a successful ping. We started having problems > recently with mon not getting the heartbeats out in a timely manner. I > turned on the debug option and started digging through the log. I > noticed that usually, once per second, there's an entry in the log > that's just a number, and it increments by one...usually every second... > What version of mon are you running? Mon prior to the 1.1pre* series blocked during the execution of an alert script, so I'd check to see if you have an alert script which is hanging. -David Nolan Network Software Designer Computing Services Carnegie Mellon University ___ mon mailing list mon@linux.kernel.org http://linux.kernel.org/mailman/listinfo/mon
RE: Mon Debugging
--On Thursday, July 14, 2005 4:42 PM -0500 Aaron Segura <[EMAIL PROTECTED]> wrote: So, for future troubleshooting efforts, the debug counter *is* supposed to update once a second? I will try upgrading to 1.1.0pre1. Thanks for the info. I wasn't sure, but I just went and read the code. Its a loop counter, incremented once on every pass through the main loop. In a normal situation mon sleeps for 1 second on every loop after processing any pending I/O. So unless mon is either really busy (lots of jobs being queued, lots of I/O to process) or hanging (due to alert issues with older mon), yes, the counter should go up approximately once a second. -David David Nolan<*>[EMAIL PROTECTED] curses: May you be forced to grep the termcap of an unclean yacc while a herd of rogue emacs fsck your troff and vgrind your pathalias! ___ mon mailing list mon@linux.kernel.org http://linux.kernel.org/mailman/listinfo/mon
RE: Mon Debugging
On Thu, 14 Jul 2005, David Nolan wrote: process) or hanging (due to alert issues with older mon), yes, the counter should go up approximately once a second. i'm not sure why anyone is paying attention to that counter or how long it takes to tick, since it's not used for anything other than to count how many times the main loop has iterated. ___ mon mailing list mon@linux.kernel.org http://linux.kernel.org/mailman/listinfo/mon
