[14:42:05] <_joe_> if you want to do some yak-shaving, have at it https://wikitech.wikimedia.org/wiki/User:Giuseppe_Lavagetto/Service_ports [14:42:09] <_joe_> ottomata: ^^ [14:42:30] <_joe_> we should change your service's port :P [15:27:45] ah! ooookkkkk [15:28:18] i might make them end in 92 if that is ok with you [15:28:34] <_joe_> ahah sure [15:28:35] probably 4392 for logging [15:28:37] 4192 for analytics [15:28:48] <_joe_> why do you love 92 so much? [15:28:51] trying to keep them matching their service ports but starting with a 4 [15:28:52] <_joe_> if I may :P [15:28:57] kafka is 9092 [15:29:01] <_joe_> oh I c [15:29:05] eventstreams [15:29:08] i think is 8092 [15:29:12] and eventgate default is 8192 [15:29:14] (defaults) [15:29:32] just trying to keep 92 as a un enforced convention for streaming stuff [15:29:34] but really doesn't matter [15:30:15] <_joe_> that's ok [15:30:18] <_joe_> we can explain it [16:21:38] _joe_: in https://wikitech.wikimedia.org/wiki/User:Giuseppe_Lavagetto/Service_ports do you mean 4000-5000 range? [16:22:12] <_joe_> yes [16:22:17] <_joe_> what did I write? [16:22:30] 4500 [16:22:34] <_joe_> lol [16:22:36] <_joe_> nope [16:43:04] hm, jbond42 i can't run puppet-merge [16:43:10] and i see your username in the output [16:43:17] E: failed to lock, another puppet-merge running on this host? [16:43:17] locking process tree: systemd---sshd---sshd---sshd(jbond)---zsh---sudo(root)---puppet-merge---python3(gitpuppet) [16:44:18] yes was just testing is it critical? [16:44:35] oh guess not, just some firewall rules, not critical but would be good to get merged soon [16:44:46] nothing will break unless we merge a subsequent patch too [16:44:55] ack if you are happy for me to merge ill do so in the next 10 mins [16:44:59] yes please thank you [16:45:02] let me know when you do [16:45:10] godog: you also happy for me to merge yours? [16:45:46] jbond42: woops, yes please [16:46:06] ack will ping here when done (likley 5 mins or less) [16:46:29] no worries, harmless change on my end [16:46:35] ack [17:00:49] lol the 'locking process tree' output is some of the most horrible code I've written, to the point where it's funny to see it work [17:05:08] cdanis: not sure if you saw in m_security but seems like there is something wrong with the locking see https://phabricator.wikimedia.org/P9878 & https://phabricator.wikimedia.org/P9879 [17:05:26] i was unable to recreate it but possibly something to do with a change on the private repo [17:05:39] yeah I meant to take a look after my next 2h of meetings ;) [17:05:41] er :) [17:06:12] ok cool thanks (not urgent) :), i didn;t see anything obvious so a second set of eyes is appreciated [17:13:12] jbond42: another patch waiting for ya [17:14:27] ottomata: can you send a paste i dont see anythin g? [17:15:21] https://gerrit.wikimedia.org/r/c/operations/puppet/+/558128 [17:15:35] hm, did it get merged? [17:15:38] OH [17:15:40] i can merge now [17:15:52] just puppet mergedd [17:15:59] ack [17:17:29] godog: herron FYI apache kafka 2.4 is out, and it has a cool new feature: [17:17:34] allow consumers to fetch from closest replica [17:17:46] which means consumers can be spread out between replicas, not just partition leaders! [18:03:03] ottomata: sweet! looks like a nice win