[00:42] <Laney> nacc: it should retry until it gets an allocation
[00:42] <nacc> Laney: ok, that's what I figured
[00:42] <Laney> some tests require m1.large instances, and it's possible for others to sneak in underneath them
[00:43] <nacc> Laney: sure
[00:43] <Laney> AFAIK the linux failures are real ones
[00:43] <nacc> Laney: yeah, I don't think they are strictly related to the packages in question, but it does seem to be a real failure
[00:45] <Laney> Mmm - I think that it was allowed in at some point and now it continues to fail when triggered by other things
[00:45] <Laney> Or something like that
[00:46] <nacc> Laney: sounds about right
[01:26] <xnox> nacc i am confused about it too
[01:29] <Unit193> xnox: I don't suppose you'd care to comment on LP 643623?
[01:47] <jbicha> pitti: it looks like you can drop the badtest hint for gjs/armhf, no idea what's not working with s390x
[01:48] <jbicha> (gjs 1.47.90 was broken until -0ubuntu3)
[04:00] <doko> rbasak, caribou: that most likely comes from -O3. either adjust the symbols files, or force to build with -O2 on ppc64el
[09:34] <Unit193> in src:gocryptfs, disable the first patch and you fix the ftbfs.