mozilla :: #perf

16 Mar 2017
13:34rwoodigoldan, jmaher|afk: I have to head out to the dentist, hopefully won't be too long
13:34igoldanrwood: ok
13:35igoldanrwood: take care
13:35rwood|afkthx
14:22igoldanjmaher|afk: hi, Joel
14:22igoldanjmaher|afk: I'm configuring Mercurial locally
14:23igoldanjmaher|afk: I'm running mach mercurial-setup
14:23igoldanjmaher|afk: and want to know: did you enabled the history rewriting commands?
14:23igoldanjmaher|afk: do you use them?
14:51rwoodigoldan: if you haven't seen this, this is worth a look: http://mozilla-version-control-tools.readthedocs.io/en/latest/hgmozilla/installing.html
14:53rwoodand yep you will want the history rewriting extensions as noted in the docs
14:55igoldanrwood: thanks, Robert
14:57rwoodnp
15:52igoldanrwood: bye, guys
15:52rwoodigoldan: see you tomorrow
15:53igoldanrwood: have a great day!
16:20jmaherrwood: I will answer igoldan's email questions today (assuming you are not in the middle of it)
16:27rwoodjmaher: I just had a look, ok great thanks
18:03jmaherrwood: on those alerts that were on the wrong changeset, can we work on filing new bugs for the correct changeset?
18:04rwoodjmaher: yes I did a bunch of retriggers, they should be done soon
18:05jmaherrwood: ok, I should have known you would be on it!
18:06rwood:)
18:20rwoodjmaher: I don't know how to deal with this one, it's a merge commit on autoland, and on the graph it looks like it happened on inbound first (but there's no alert). Do I need to bisect all the commits in the merge commit?
18:20rwoodhttps://treeherder.mozilla.org/perf.html#/graphs?timerange=7776000&series=%5Bautoland,117a4f9a29d95c4b5f2de267475fffacc53f5944,1,1%5D&series=%5Bmozilla-inbound,117a4f9a29d95c4b5f2de267475fffacc53f5944,1,1%5D&zoom=1486489200382.0225,1489681463000,50,89.53722917094933
18:21jmaherrwood: oh the 'details'
18:21rwoodhaha
18:21jmaherrwood: interesting, it is bi-modal, yet we switched dominant modes
18:22rwoodright
18:22jmaherrwood: ignore alerts, find the root on inbound
18:22rwoodok, yeah looks like inbound to me also
18:22rwoodthanks
18:22jmaherand then manually create an alert there- once done, you can downstream the autoland alert to the new alert you created
18:22rwoodok cool!
21:49digitaraldRyanVM ddurst: any input on bug 1346865?
21:49RyanVMdigitarald: I've got about 3 different things to reply to you on right now
21:50RyanVMlong story short is that I'm not sure why this is a war room request vs. a general "Quantum needs data engineering" request
21:50digitaraldRyanVM: we can jump on vidyo if that streamlines it
21:50RyanVMwhich IIUC is what the meeting in Toronto on the 30th is about
21:50digitaraldRyanVM: there is definitely some overlap
21:50RyanVMI think "War Room" means different things to different people at this point and it's not clear we're even on the same page anymore
21:50digitaraldRyanVM: probably a major point for the agenda that answers most other questions
21:51RyanVMand yes, that needs significant clarification
21:51digitaraldRyanVM: for me its a place to wear army boots and smoke cigars oh wait
21:51RyanVMas do my team's Q2 priorities as I'm assuming they're going to be changing significantly soon
21:53digitaraldRyanVM: sounds like we need to figure things out fast
21:53RyanVMyep
21:53RyanVMgood thing I haven't even been formally invited to the meeting on the 30th yet ;)
21:53digitaraldshould we huddle before Toronto to have less open qeustions in Toronto and do deeper dives
21:53RyanVM\_()_/
21:54digitaraldI could give an overview of quantum perf/metrics work
21:55RyanVMif you think it'll help, I'm not sure that's what's slowing things down right now
21:56RyanVMdigitarald: it's 6pm here, send me an invite and we can chat tomorrow
17 Mar 2017
No messages
   
Last message: 104 days and 15 hours ago