[13:56:51] ATM this festival is on in Valencia https://en.wikipedia.org/wiki/Falles [13:57:11] == loud bangs all day and marching bands in the street, and despacito playing [13:57:28] hahaha [13:59:45] I think despacito is the worst of part [13:59:57] that is so sad, alexa play… nevermind... [14:04:05] haha [14:27:15] cdanis, godog i have just had my first attamnet and writing an mtail prpogram for ulogd as i couldn't find a pre-exisiting one. [14:27:48] i have added you both as reviewers however if you could let me know a way to test this it may save you time pointing out some opbvious problems [14:29:59] jbond42: *nod* there are tests in the mtail module, although hic sunt leones, to test locally you need 'mtail' installed [14:30:12] jenkins will DTRT though if you provide tests [14:32:46] do you mean in the puppet module? i dont see anything in puppet/modules/mtail [14:33:03] the tests are hidden under modules/mtail/files/test [14:33:12] some python there and then the 'logs' subdir of there has expected input [14:33:41] ok great that should be enough for me to move forward [14:34:10] the other thing i wanted to ask is how to theses logs make it to logstash. they are on the central syslog host but i cant seem to see them in logstash [14:36:02] ATM logstash is opt-in, so we'll need to opt-in ulogd if that makes sense and its logs are ok to be viewed by all NDA users [14:36:53] also worth double checking the local testing mtail version against the version in prod as the configs are not always compatible between versions [14:37:12] ok maybe we can descuss that in the wedensday meeting i think the data should be ok but i suspect opinions my vary [14:37:20] thanks herron [14:37:26] (we should probably look at upgrading mtail on the rsyslog hosts) [16:20:34] I put a bunch of placeholders for goals at the goal pad ( https://etherpad.wikimedia.org/p/SRE-goals-FQ4-FY1819 ) [16:22:02] db automation will have to wait for monday I think, to also coordinate with volans, joe, etc. [16:22:27] logging + metrics require a bit more coordination in this team I'd say to come up with a realistic proposal [16:22:51] at least godog, herron, shdubsh, and cdanis expressed some interest [16:23:03] and then we have the Kafka stuff [16:23:07] that is still a bit nebulous [16:23:33] herron: do you have a better sense of what to do there? could you perhaps coordinate that with e.g. ottomata? [16:23:47] sure would be happy to coordinate [16:24:06] from the sound of the task he was hoping for us to make the decision on when to expand, but yeah would be good to sync up [16:24:18] yes [16:56:16] So the database automation, this is clearly in line with general automation work [16:57:00] why is the primary teams a ? [17:03:43] I don't know the full extent of the task [17:04:03] depending on how it will look like, it depends whether e.g. data persistence will be primary [17:04:13] and us assisting/helping, rather than driving it [17:04:28] roger [17:08:29] my intention was to discuss across all three involved subteams on Monday [17:10:08] That makes sense [17:10:35] is that pad going to be the one used at the meeting / getting filled in by other sub-teams? [17:11:50] yes [17:11:58] cool :) [20:51:16] if anyone has any review bandwidth: https://gerrit.wikimedia.org/r/c/operations/puppet/+/496836 https://gerrit.wikimedia.org/r/c/operations/puppet/+/493774