=== Ursinha is now known as Ursinha-zzz === Ursinha-zzz is now known as Ursinha [12:12] gary_poster: I happened upon something reading my feeds last night that tied into our discussion yesterday: http://leanandkanban.wordpress.com/2011/07/15/demings-14-points/ [12:13] benji, heh, Francis gave me that same post. Great minds.... Yeah, it is good. [12:13] heh [12:31] bac benji danilos call rsn [12:31] ack [12:31] a-ok [12:32] sorry, i'm in the echo chamber again this morning: http://instagr.am/p/JZQKy/?ref=nf [12:49] gary_poster, looking at verify_pofile_stats.py, it seems the fix would be trivial [12:50] danilos, cool! You want to take it to up your critical bug count (sounds like an RPG stat), or do you want to coach someone? up to you [12:51] gary_poster, heh, I'm fine with either, and I don't think any coaching is needed: move one line fetching a template from after DBLoopTuner.run() to be before it (not sure how to test for "LP tree has just disappeared on you", though) [12:52] heh [12:52] gary_poster, I'll take it then [12:52] ok danilos, cool [14:21] oh, only now I notice my office computer is back on the internet, good choice coming back home instead of waiting for internet to come to the office :) [14:30] gary_poster, re your recent questions about "cold cache" timeout bugs, do you think we should just run with the "cold-cache" bug tag or something? bug 811447 in particular looks to be the victim of that (last seen OOPS is on qastaging with a time of 9s or so) [14:30] <_mup_> Bug #811447: BugTask:+index timeout - death by sql in PersonSubscriptions(user, bug) < https://launchpad.net/bugs/811447 > [14:32] (that was not in progress, I had a brainfart and marked that bug instead of 810116 :/ [14:35] danilos, sure sounds like a plan [14:37] gary_poster, not sure how much of a plan it is, but these are especially hard (as you noted in your email to list) [14:38] gary_poster, basically, the metric is going to be "does not time out on either staging or qastaging" [14:55] danilos, my plan is "ignore until unavolidable" [14:56] gary_poster, right, though I seem to be wrong about 810116: the query seems to be slow only for that user, I got a >100s run time on staging (cold caches, 2.7s warm caches) [14:56] uhm, not 810116 but 811447 [14:56] huh weird [14:58] gary_poster, probably not weird, that person is a member of a bunch of teams that might be subscribed to the bug, whereas we are not [14:58] ah [14:59] gary_poster, query for me on staging is 0.7s cold, 170ms warm [14:59] gary_poster, "for me" == when I substitute my own user ID [15:02] so a non-cold bug agreed. I suggest you note on bug if you don't tackle [15:12] gary_poster, right, I will [15:13] gary_poster, likely related to person being in a team ubuntu-bugs which is subscribed to 50-60 of the 450 duplicates [15:17] heh [15:26] benji: i just got suggestions on linking a branch for the first time. very cool! [15:27] bac: cool! I just wish the search was faster, that would make it much nicer. [15:27] it's faster than i can type [15:29] :) [16:28] bac, going to lunch, but https://code.launchpad.net/~gary/launchpad/bug821487 I think will be fast and easy, when you have a moment [17:32] gary_poster: just back from lunch and will look at it now [17:36] gary_poster: done [18:12] thank you bac. [18:13] np [21:36] gary_poster: sorry if i waded into that discussion and made things worse [21:37] bac, np, I thought you said good stuff. maybe I was too prickly. I just have heard that all before from him, so its a rehashed conversation. and maybe it's the end of the week, I dunno. :-P [21:37] s/its/it's/ [21:37] yeah, i got the impression the discussion had some history to it [21:38] :-) [21:38] have a good weekend [21:39] you too bac