<div dir="ltr">Thanks! That seems like it just may be the ticket. I have enabled the debug output and changed the time calculation to > mytime = str(int(time.time())) So far so good, but the real test will be over the next few nights. I'll keep you guys posted. Thanks again for all the help! <div class="gmail_quote">On Tue, Aug 12, 2008 at 2:17 AM, Thomas Guyot-Sionnest <<a href="mailto:dermoth@aei.ca">dermoth@aei.ca</a>> wrote: <blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <div class="Ih2E3d">-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 </div><div><div></div><div class="Wj3C7c">On 11/08/08 11:23 PM, Thomas Guyot-Sionnest wrote: > On 11/08/08 04:57 PM, fevin Kagen wrote: >> So... is a date sent in the SNMP trap or is the date set to the >> time that Nagios receives the alert? > > It's in the command sent to the Nagios command pipe. > > See: > > <a href="http://nagios.sourceforge.net/docs/3_0/extcommands.html" target="_blank">http://nagios.sourceforge.net/docs/3_0/extcommands.html</a> > <a href="http://www.nagios.org/developerinfo/externalcommands/commandlist.php" target="_blank">http://www.nagios.org/developerinfo/externalcommands/commandlist.php</a> > >> Is there a way to output to both the existing output (the nagios >> command file) and a debug output? It doesn't happen all the time so >> trying to capture a bad entry would mean taking the whole system down >> for a week plus. > > I never touched python before, but looks plain simple object-oriented > programming. I'd try making the post_results function looks like this: > > def post_results(host, job, mondata_res, return_code): > mytime = time.time() > mytime = str(mytime) > mytime = mytime[:-3] > #print mondata_res > output = open('/usr/local/nagios/var/rw/nagios.cmd', 'w') > debug_out = open('/tmp/services-debug.out, 'a') > results = "[" + mytime + "] " + "PROCESS_SERVICE_CHECK_RESULT;" \ > + host + ";" + job + ";" \ > + return_code + ";" + mondata_res + "\n" > output.write(results) > debug_out.write(results) > > The change are oppening a 2nd file in append mode: > debug_out = open('/tmp/services-debug.out, 'a') > > And then writing the same thing as the nagios out to it: > debug_out.write(results) > > As soon as it fails check the end of that file, you should see the same > thing as what nagios received. If the [time] part is the problem thoug I > can't really help. This is the part that generated the time part: > > mytime = time.time() > mytime = str(mytime) > mytime = mytime[:-3] > </div></div>Well, python is a language I wouldn't refuse to learn, and I just started looking at it. On the fun side there's an interactive mode that let you learn as you try live... After a few minutes of reading <a href="http://docs.python.org/tut/" target="_blank">http://docs.python.org/tut/</a> ... - From what I understand in the above commands, time.time() returns you a float object that returns unix time with two decimals, ex: 1218520859.75. In this plugin the author strips the last three characters from the string'ified value ([:-3] to get a normal timestamp. Whenever the time is rounded (ex: 1218520859.7, 1218520859.0) this strips too many characters and you get a date far in the past. The proper way to do it is simply by converting it to an integer, i.e. replace the three lines above with: mytime = str(int(time.time())) Since I believe the str() was only to use the hooks on it, this probably works as well: mytime = int(time.time()) Hope this helps. <div class="Ih2E3d"> - -- Thomas -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - <a href="http://enigmail.mozdev.org" target="_blank">http://enigmail.mozdev.org</a> </div>iD8DBQFIoSr86dZ+Kt5BchYRAmZ5AJ9zH2akrPQfhsLhve74jgxtdicLjwCfWsIo qt0WgLTig7uVQvf1JwYcCMA= =KMAD -----END PGP SIGNATURE----- </blockquote></div> </div>