[01:54] <xnox> bedtools adt run is odd -> requested to test src+binary version 2.19.1-1, jenkins executes source 2.17.0-1 which doesn't have adt tests.
[01:55] <xnox> similar issue with epsilon.
[01:57] <xnox> ipset
[02:02] <xnox> haha retry does not help, it retries old version again none-the-less...
[07:16] <darkxst> should daily iso'
[07:16] <darkxst> s be spinning again now?
[07:17] <darkxst> i.e. for Ubuntu GNOME
[07:17] <darkxst> cjwatson, infinity, stgraber ^
[12:56] <infinity> darkxst: Not yet.
[21:49] <cjwatson> darkxst: No, we'll probably try turning them on tomorrow (assuming the livefs chroot creation ticket got resolved; haven't checked)
[21:57] <darkxst> cjwatson, ok
[21:59] <cjwatson> wgrant: Could you switch http://qa.ubuntuwire.com/ftbfs/ over to utopic when you get a chance?
[22:03] <infinity> cjwatson: It's not resolved yet, AFAIK.
[22:03] <infinity> cjwatson: Also, if you're around, do you happen to know a workaround for britney losing its mind over massive recursive sets?
[22:03] <infinity> AIEEE: counter overflow: 7309 involved packages.
[22:03] <infinity> AIEEE triggered by: libav
[22:03] <infinity> That sort of thing.
[22:05] <cjwatson> I don't
[22:05] <cjwatson> You could block stuff until it's all ready per a transition tracker ...
[22:05] <cjwatson> Then at least probably only the last run will take forever
[22:06] <infinity> Oh, kay, so this doesn't end up being a failure to process things, it's just telling me it hates me?
[22:06] <infinity> Blocking the three problematic packages for now would be simple.
[22:06] <infinity> gst-plugins-bad1.0, libav, rtmpdump
[22:07] <infinity> Let's try that for now and revisit in the $later.
[22:07] <cjwatson> Yeah, I don't understand it properly but I think the effect is that it gives up on some paths and then spends $forever backtracking
[22:08] <cjwatson> But it doesn't actually cause a failure directly
[22:08] <infinity> I expect xnox will yell at me later about blocking one of his transitions, but we'll sort that out in a bit.
[22:08] <cjwatson> However it does tend to cause p-m runs to take hours, so it's best avoided
[22:09] <xnox> infinity: cjwatson: do block libav, and portions of ghc.
[22:10] <xnox> they can be easily worked of a tracker.
[22:10] <cjwatson> ghc doesn't usually cause counter overflows?
[22:10] <xnox> i think it's something else though, libraw or some such?
[22:10] <wgrant> cjwatson: Oh yeah, I should do that.
[22:10] <xnox> infinity: do it as a bug, that way anybody can cancel it once ready.
[22:10] <infinity> xnox: It's the three I mentioned.
[22:11] <cjwatson> xnox: I'll babysit ghc as usual, I imagine; it shouldn't need blocking
[22:11] <xnox> infinity: yeah, block, there are logs to see past runs anyway to figure out what needs doing.
[22:12] <infinity> cjwatson: Any cleanup need doing if one kills britney mid-run?
[22:12] <infinity> Cause I'm sick of waiting on this one to finish. :P
[22:12] <cjwatson> infinity: Shouldn't be
[22:12] <infinity> Oh, it might be near done anyway.
[22:13] <infinity> They've been ~4h each since this started, and this is 2.5 in...
[22:13] <infinity> Err, 3.5
[22:13] <infinity> Meh.
[22:13] <infinity> Screw it.  Killing.
[22:20] <infinity> Oh, there were a few other triggers too.  Bother.
[22:21] <cjwatson> They might be enough smaller than libav not to matter as much
[22:21] <infinity> AIEEE: counter overflow: 3217 involved packages.
[22:21] <infinity> AIEEE triggered by: openjpeg
[22:21] <infinity> Still seemed pretty big. :P
[22:22] <infinity> I've just blocked every trigger for now to see if that happies things up, then we can peel back the damage.