[17:32:30] <jbond>	 hi all whats the best way to see when an alert last triggerd? specificly the "CPU frequency" check on any host
[17:32:44] <jbond>	 its currentl an nrpe check
[17:33:26] <denisse>	 Hi Jhon, I think you could see that information in Logstash by searching for that alert. :)
[17:44:39] <jbond>	 denisse: thanks howevr i think im mising something i have tried to look at evrything comming from alert1001 https://logstash.wikimedia.org/goto/b809b1e359d9831e378c2ef013be6fdc and it seems all the messages are about frtech hosts
[17:44:53] * jbond is a logstash novies so assume i missed something obviious :)
[17:45:56] <jbond>	 nevermind i worked it out :)
[17:46:24] <jbond>	  do yo know how far dose logstash go back, i have a feeling its either 30days or 90
[17:49:55] <denisse>	 Nice! Please let me know if I can help you with something else.
[17:50:18] <denisse>	 I'm investigating how long does we store that. I'll get up to you in a moment.
[17:51:37] <jbond>	 thanks :)
[17:52:58] <denisse>	 jbond: We retain logs for 90 days. https://wikitech.wikimedia.org/wiki/Logstash#Data_Retention
[17:53:10] <jbond>	 denisse: great thanks  <3
[19:44:14] <mutante>	 while replacing monitoring::service Icinga checks with prometheus::blackbox::check::http (you like this, right?:).. I notice that our good old "notes_url" parameter goes away. Hmm, I remember how we put quite some effort into "every check needs to have a runbook URL on wikitech". Should we keep that parameter?