[13:20] <rbasak> I want to find the last activity on a bug by API. I found the bug_activity collection but don't see a defined ordering. Is there anything that will give me a most-recent-first (such as the bugs collection is defined to do)? Or should I just iterate and sort myself - or do I iterate and not need to sort?
[13:25] <cjwatson> bug_activity is ordered by ID, which should generally amount to most-recent-last
[13:27] <rbasak> Thanks. So since I can't slice from the end of a collection, I guess I need to iterate anyway?
[13:28] <cjwatson> I'd just iterate; the size is usually not prohibitive.
[13:29] <rbasak> OK, thank you.
[13:29] <cjwatson> In the common case it will fit in a single batch.
[13:29] <rbasak> I really like this API BTW. Very nice and straightforward.
[13:30] <cjwatson> It's rare but nice to get praise for the Launchpad API ;-)
[13:34] <muktupavels> cjwatson: do you have time to review merge proposal? https://code.launchpad.net/~mitya57/git-build-recipe/+git/git-build-recipe/+merge/311874
[13:41] <cjwatson> muktupavels: looks good to me, will land
[13:41] <muktupavels> thanks!
[13:49] <cjwatson> uploaded to Debian unstable, will see about getting it onto our builders a bit later
[13:58] <mitya57> cjwatson, thanks!
[16:54] <tsimonq2> cjwatson: Ping, I messed up. Kubuntu recently staged a good amount of packages, and we have a script that uses the API to retry the dep waits. So thay ended up triggering 200 some builds before some common deps were ready. Not only did I accidentally flood the queue, but that may be why half of the amd64/i386 builders are disabled, right?
[16:55] <tsimonq2> s/flood the queue/flood the queue with builds that are known to fail/
[16:56] <tsimonq2> Is the builder thing my fault? If so, I admit to it.
[16:59] <cjwatson> tsimonq2: not your fault
[17:00] <tsimonq2> cjwatson: Ok, so what's the deal then? :)
[17:00] <cjwatson> tsimonq2: lcy01's openstack infrastructure has been having some difficulty and is down for the time being; the rest is certainly worse than usual, but I don't have time to look today, I can just re-enable and hope
[17:01] <tsimonq2> cjwatson: Ok fair enough
[21:40] <mapreri> the job queue for the amd64/i386 builders is huge.  what's going on?  Also all of lcy01-* is "Cleaning"
[21:41] <nacc> mapreri: i asked this in #ubuntu-relesae as well -- i think lcy01's openstack is having issues (based upon cjwatson's comment ealier) and there was a glibc and perl5 upload (so a lot of stuff is getting rebuilt?)
[21:41] <mapreri> oh, I should have really read backlog
[21:41] <mapreri> nacc: there are no rebuilds for those uploads... (though there they do cause a lot of autopkgtests to be triggered :S)
[21:42] <nacc> mapreri: err, right, sorry!
[21:42] <nacc> mapreri: i meant queue entries
[21:43] <mapreri> nacc: what do you mean by "queue entries"?
[21:43] <nacc> mapreri: as you said, autopkgtests that need to run
[21:43] <mapreri> ah yeah
[21:51] <cjwatson> those don't affect LP builders
[21:51] <cjwatson> they run on the same cloud but there are quotas
[21:52] <cjwatson> I think the most recent problem was an internal DNS outage that probably took out a bunch of stuff
[21:52] <cjwatson> so we'll see how it goes as that all clears
[21:52] <cjwatson> looks better already than when I last looked
[21:53] <nacc> cjwatson: ah ok, thanks for that info (filing it away)
[21:54] <mapreri> cjwatson: well, looking at https://launchpad.net/builders seems lcy01 is still not doing anything :)
[21:55]  * mapreri got used to no build queue at all on launchpad :>  guess it's bad idea to get used to such performance
[21:57] <cjwatson> yeah, lcy01 is probably still dead but lgw01 is OK on its own once it's caught up
[21:57] <cjwatson> which it should if the resolver quits exploding
[21:58] <mapreri> then where would be the fun
[22:03] <acheronuk> has been trying to catch up all day, and got not very far with it. hopefully overnight it can