[12:09] <gary_poster> oops
[12:09] <gary_poster> bac benji frankban call in 1 or 2 (ASAP)
[13:07] <gary_poster> benji, did your fresh setuplxc run give you PG 8.x or 9.x?
[13:08] <benji> gary_poster: I haven't looked yet, been doing reviews.  Let me take a peek.
[13:09] <gary_poster> thx
[13:10] <benji> gary_poster: 8.4
[13:10] <gary_poster> ok thanks benji
[13:11] <gary_poster> benji, I guess hat means there's still a chance you might have a fix for the bug relatively quickly, so taht's good news :-)
[13:11] <gary_poster> two chances to spell "that," and two misses. :-)
[13:11] <benji> heh
[13:15] <frankban> benji: thanks for the review. gary_poster: I can land the branch benji reviewed also removing the slice used to avoid buffer overflows if you agree: the fix has been released
[13:15] <gary_poster> frankban, +1
[13:15] <frankban> ok cool
[13:33] <bac> yay AAA
[13:35] <benji> bac: a big fan of the american anthropological association are you?
[13:35] <gary_poster> heh
[13:36] <bac> i'm a big fan of big-ass flatbed tow trucks coming to my house for "free".
[13:36] <gary_poster> ah.  Land Rover?
[13:40] <gary_poster> benji, fwiw, I'm starting my daily test run with pg 9.1 installed (hopefully this does the trick http://pastebin.ubuntu.com/973543/) just to see how it affects things
[13:41] <bac> yes, blown brake slave cylinder.
[13:41] <benji> k
[13:42] <gary_poster> bac, have you ever added up how much repairs for that thing have totaled over the years, or have you merely cowered in fear? ;-)
[13:43] <bac> nope
[13:43] <gary_poster> :-)
[13:44] <bac> but it has been pretty good for the last few years after i put in electronic ignition.  some parts, like condensers, were consistently bad.  since no one uses them anymore they sit on shelves for a really long time.  so replacing those parts with modern tech really improved things.
[13:45] <bac> meanwhile, i've had my MINI for 8 years and have spent >$500 total in repairs
[13:45] <bac> er, <$500
[13:58] <gary_poster> wow
[14:38] <gary_poster> benji, tests fail miserably with PG 9.1 in general, from what I can see (http://ec2-23-22-4-132.compute-1.amazonaws.com:8010/waterfall) but maybe the test you need will still pass.  The failures have a trivial smell to them, but I'm not sure what the various necessary fixes are,
[14:38] <gary_poster> .
[14:39] <gary_poster> ow, something just killed the tests cold...
[14:40] <benji> gary_poster: interesting.  If I get done with understanding these translations reviews before the heat-death of the universe I'll try to see if just my test passes with 9.1 reliably, then we can at least know how to dispose of the one bug
[14:40] <gary_poster> cool.  yeah, I was feeling optimistic about the change generally until everything just died right now...
[14:41] <gary_poster> oh I see
[14:41] <gary_poster> a lot of those failures were layers not starting up
[14:42] <gary_poster> so therefore the associated tests didn't run
[15:00] <gary_poster> benji, mm, pg was running on wrong port.  trying to fix things up.  maybe will be better after that
[15:00] <benji> mmm
[15:02] <benji> ok, I just finished the outstanding reviews; trying to get PG 9.1 running on my container now
[15:10] <gary_poster> that diff to setuplxc is wildly insufficient :-P
[15:17] <benji> gary_poster: I don't follow.
[15:18] <gary_poster> benji, above I gave a pastebin to a small change I had made to setuplxc to try and get postgres 9.1 working (http://pastebin.ubuntu.com/973543).  It is wildly insufficient.
[15:18] <benji> ah!  I thought you were talking about the "diff" of the MP that I reviewed earlier.
[15:21] <gary_poster> nope :-)  I saw those sliding by though.
[15:22] <frankban> gary_poster: so, we need to change the port number in /etc/postgresql/9.1/main/postgresql.conf? Maybe the port number will be fixed in the package itself once the transition is done?
[15:24] <gary_poster> frankban, yeah, I hope so.  That would make sense to me.  Also, you need to actually remove 8.4; this will force the system to get all the other 9.1 dependencies.  That also will probably be handled autiomatically
[15:27] <frankban> gary_poster: is 8.4 still installed using your diff?
[15:27] <gary_poster> frankban, yeah. :-/
[15:27] <gary_poster> not sure why
[15:27] <gary_poster> I think I have things fixed up now.  retrying tests...
[15:28] <gary_poster> argh
[15:28] <frankban> gary_poster: you could try replacing "postgresql-9.1" with "launchpad-database-dependencies-9.1"
[15:29] <gary_poster> frankban, ah ok.  worth a try.  I may need to do that anyway because I keep hosing my ec2 machine for one reason or another...
[15:34] <gary_poster> I see a lot of this building too.  http://pastebin.ubuntu.com/973765/
[15:34] <frankban> gary_poster: if that works, maybe you have a fix for the port number issue too.
[15:34] <gary_poster> frankban, cool, good point
[15:38] <gary_poster> things are working much better now, at least...
[15:52] <benji> gary_poster: how did you get PG 9.1 to start?  the "service" command appears to ignore my taunts
[15:52] <gary_poster> benji, heh, service postgres st[op|art] worked for me
[15:53] <benji> hmm
[16:05] <frankban> gary_poster, benji: am I corrct if I assume a shared library in ubuntu is installed in different paths depending on the system to i386 or amd64? e.g. /usr/lib/lxc/liblxc.so.0 vs /usr/lib/x86_64-linux-gnu/lxc/liblxc.so.0
[16:06] <benji> frankban: I don't know.
[16:07] <gary_poster> frankban, I don't either, but I am in amd64, and I have the former (/usr/lib/lxc/liblxc.so.0) and not the latter (/usr/lib/x86_64-linux-gnu/lxc/liblxc.so.0).
[16:09] <frankban> thanks. i have the latter after installing the latest quantal version of lxc (that I build with my changes).
[16:12] <frankban> gary_poster: maybe I should ask Serge about this. It can be an error in my build, or a change in the quantal version. In the latter case, I think I should change lxc-ip to try the first path and fallback to the second.
[16:13] <gary_poster> frankban, gotcha.  yeah, sounds like a good plan.
[16:29] <gary_poster> gmb, hey.  https://talkgadget.google.com/hangouts/_/extras/canonical.com/goldenhordeoneonone when you are ready
[16:30] <gmb> gary_poster: Sure. Just getting bootstrapped.
[16:31] <gary_poster> cool
[16:34] <gmb> gary_poster: I think hangouts are a no go on this connection. Can we skype?
[16:34] <gary_poster> sure gmb
[17:24] <gmb> gary_poster: Couldya tell that everyone just hopped on the wifi? :)
[17:30] <gary_poster> gmb, heh, yes I could
[17:32] <gary_poster> benji, so is that the end of the road for that bug?  You can tell me in an hour or so--I'm going to lunch :-)
[17:32] <benji> gary_poster: pretty much, I'm going to try to find a workaround for just a little bit after lunching and relocating
[17:43]  * benji relocates.
[18:12] <bac> hi gary_poster
[18:12] <bac> re: bug 993467
[18:12] <_mup_> Bug #993467: lib/lp/bugs/browser/tests/distrosourcepackage-bug-views.txt fails when run alone (on lxc?) <paralleltest> <Launchpad itself:In Progress by bac> < https://launchpad.net/bugs/993467 >
[18:13] <bac> i have traced it back to schema/patch-2209-16-1.sql not being applied.  (new since 2012-04-23)  where did you see the original failure?
[18:23] <gary_poster> bac, locally!  ugh, sorry. :-(
[18:23] <gary_poster> bac, so, mark it in invalid and please accept my apologies.
[18:23] <bac> gary_poster: easy mistake.  sadly i was in the same boat.
[18:23] <gary_poster> oh!
[18:23] <bac> so i replicated the failure
[18:23] <gary_poster> heh, and :-/
[18:24] <bac> but i brushed up on storm debugging and some other stuff.
[18:24] <gary_poster> well, that's cool at least :-)
[18:25] <gary_poster> not what we want though :-/
[18:25] <gary_poster> glad it is resolved.
[18:25] <gary_poster> two green runs in a row
[18:25] <bac> \0/
[18:27] <bac> gary_poster: did you get your ec2 bill?  was it crazy?
[18:28] <gary_poster> bac, not as bad as the month before.  This was maybe $120, and the month before was > $300.  I think that only having a single gigantic machine helps
[18:28] <bac> mine was $211.31!
[18:29] <bac> and that didn't include any runaways
[18:34] <gary_poster> wow
[18:34] <gary_poster> that sucks
[18:35] <bac> gary_poster: expense submitted
[18:36] <gary_poster> ack...logging on...
[18:37] <gary_poster> bac, why the small two additional separate EC2 items?  what are they?
[18:37] <bac> march and april
[18:38] <bac> i indicated that in the email
[18:38] <bac> they really drive down my average!
[18:38] <bac> gary_poster: ^
[18:38] <gary_poster> oh, gotcha bac, cool.  I'll mention that in the line items too just for future reference.
[18:38] <gary_poster> average: heh
[18:41] <gary_poster> bac, approved
[18:52] <benji> gary_poster: I haven't been able to find a workaround for bug 992184 so I think I'm done.  I'm going to assign it to Stuart and take the ID off the card and move it to done-done, I guess.
[18:52] <_mup_> Bug #992184: lib/lp/services/database/doc/textsearching.txt fails intermittently/rarely on parallel tests <paralleltest> <Launchpad itself:In Progress by benji> < https://launchpad.net/bugs/992184 >
[18:52] <gary_poster> benji, how 'bout tracking
[18:52] <gary_poster> it's not forward, but it's truth :-)
[18:52] <benji> ooh, I like tracking
[18:54] <benji> gary_poster: any time you have a minute, I'm not ready to discuss goals again, but I'd like to have a (very short) follow-on to Friday's discussion
[18:55] <gary_poster> benji, sure thing.  getting ready.  meet you in oneonone
[18:55] <benji> k
[19:59] <bac> gary_poster: for the launchpad-lxc-build card did you have anything fancier in mind than http://pastebin.ubuntu.com/974325/
[19:59] <bac>  (both for setuplxc and lp-setup)?
[20:00] <gary_poster> bac, no, that's all I had in mind, ty
[20:04] <bac> gary_poster: would you put your Good Housekeeping Seal of Approval on https://code.launchpad.net/~bac/launchpad/lxc-build/+merge/104971
[20:04] <gary_poster> sure
[20:08] <gary_poster> bac, a thing of beauty and a joy forever.  approved.
[20:08] <bac> excellent
[20:14] <gary_poster> We have six clean runs in a row so far, which is 50% more than our previous record.
[20:30] <bac> gary_poster: you think they've been stalling us on our build machine waiting for these to become available?  https://plus.google.com/116015965439782966698/posts/2a87SCyWVyN
[20:31] <gary_poster> bac, heh heh, yeah, I think that must be it
[20:31] <bac> yes, it is the only plausible answer.  they were very convincing, though.
[20:32] <gary_poster> :-)