[19:05:21] cwhite: i think go.dog mentioned something about wanting to test out the auto-escalation in prod today? i'm happy to find a time and to do that [19:13:02] cdanis: I'm down. Let's do it. [19:13:57] how were you thinking of doing it? edit the business hours rotation to be empty, then let it batphone? [19:15:44] I suppose we can also edit each policy -- make the business hours policy reference an empty rotation, make the batphone policy reference the NA business hours rotation, temporarily [19:15:46] that seems best [19:16:14] it's just the policy ID that matters from the API's perspective, and it doesn't page everyone but still tests the feature [19:20:01] yeah, that seems it may cause the least disruption [19:22:16] cdanis: changed the policies around, look right to you? [19:22:40] looks good! [19:23:31] Great! Would klaxon create a page that would test this well? [19:23:35] yes [19:23:46] oh, actually, hm [19:24:16] so -- Klaxon will create a page that would get escalated fine, but, I think actually the change to the policies maybe wasn't right [19:24:52] there, I just had to swap where the timeoutw as [19:25:36] page is sent, haven't recieved yet [19:25:54] it worked for me :) I think you actually swapped in a "Dev Null" user in the rotation instead of yourself? [19:25:59] https://portal.victorops.com/ui/wikimedia/incident/2876/details [19:26:04] the timeline looks like what it should look like [19:26:13] there's first the people who opted-in for emails about all alerts [19:26:26] then it got rerouted (a couple times, but that's fine), and then it sent a push notif to me [19:26:52] hah, ok, not sure how devnull got picked. probably override isn't quite right [19:28:37] so now I think we should revert the setup to as intended, and make sure that a manual page goes to us but not to the batphone [19:28:55] sounds good [19:31:17] cool, I think that original setup looks good to me [19:31:32] great, firing again [19:32:11] got it here [19:32:36] yep, and it was outstanding for more than enough time [19:32:39] great :) [19:34:37] thanks for the assist and the code :) [19:34:43] np! [21:16:57] cwhite: *nod* that looked like a good idea from that wikitech page. I'll start there and come back for more help if I end up stuck [22:31:40] cwhite: I added you as the SRE reviewer for the 2 patches that I think will get the striker in docker logs flowing to ELK. If you're not comfortable with +2, but can +1 I'll round up another SRE to handle the +2 and deploy. [22:32:33] no huge urgency to review, but sometime tomorrow would be nice if you have time [22:58:09] bd808: will do! Thanks!