=== attente_ is now known as attente
[02:23] <Logan> can someone please reject my musescore upload in Xenial? I did the version numbering incorrectly :(
[02:26] <Logan> please reject the first one and approve the second one :3
=== jamespag` is now known as jamespage
[11:36] <flexiondotorg> infinity, Morning
[11:36] <flexiondotorg> infinity, I've got a couple of package deletions for Yakkety.
[11:36] <flexiondotorg> https://bugs.launchpad.net/ubuntu/+source/mate-netspeed/+bug/1584769
[11:36] <ubot5> Launchpad bug 1584769 in mate-netspeed (Ubuntu) "Please remove mate-netspeed from the Yakkety archive, it is now included in mate-applets" [Undecided,New]
[11:36] <flexiondotorg> https://bugs.launchpad.net/ubuntu/+source/gnome-main-menu/+bug/1584767
[11:36] <ubot5> Launchpad bug 1584767 in gnome-main-menu (Ubuntu) "Please remove gnome-main-menu from the Yakkety archive, it is obsolete" [Undecided,New]
[11:48] <xnox> hello, please reject "s390-tools (xenial-proposed/main) [1.34.0-0ubuntu8 => 1.34.0-0ubuntu8.1] (no packageset)"
[11:48] <xnox> will reupload with one more bugfix
[11:57] <apw> xnox, done
[11:58] <xnox> thanks
=== yofel_ is now known as yofel
[13:24] <sergiusens> cjwatson hey, sorry to bug you but I uploaded a package on Friday and can't find it, this is all I have "[ubuntu/xenial-proposed] snapcraft 2.10 (Waiting for approval)"
[13:25] <cjwatson> sergiusens: it's in https://launchpad.net/ubuntu/xenial/+queue?queue_state=1 awaiting SRU team approval
[13:26] <sergiusens> cjwatson I swear I checked there :-/ Thanks!
[14:28] <Trevinho> bdmurray: I got emails about possible SRU regresisons (multiple mails actually :-)), but I think they're nothuing new.
[14:34] <bdmurray> Trevinho: I agree about the regressions not being new issues.  I'll look into why you got multiple emails too.
[14:49] <sergiusens> infinity hi there, can I get snapcraft allowed into xenial-proposed?
[15:47] <apw> sergiusens, the exception for that has all sorts of QA requirements, where are you documenting they are complete?  I don't see anything in the first couple of bugs.
[15:50] <sergiusens> apw are you talking about https://wiki.ubuntu.com/StableReleaseUpdates#Snapcraft ?
[15:55] <apw> sergiusens, yep that one
[15:55] <sergiusens> apw everything change lands with a test in code (snapcraft.tests for unit tests, integration_tests and/or demo_tests)
[15:56] <apw> sergiusens, right but the SRU is supposed to indicate that all of the steps on https://wiki.ubuntu.com/SnapcraftUpdates have been completed
[15:57] <elopio> apw: QA here. Everything for this release is tested in autopkgtests. Once it gets to proposed, I do a manual verification.
[15:57] <sergiusens> apw but our package hasn't been built yet, it is stuck in the unnapproved queue
[15:57] <apw> sergiusens, i was unable to find that easily, so i was asking where as i was sure infinity will ask
[15:57] <sergiusens> apw https://launchpad.net/ubuntu/xenial/+queue?queue_state=1&queue_text=snapcraft
[15:58] <apw> sergiusens, ahh ok the write up there is not well worded as i took it to mean before queueig, and yet in there is says -proposed, so ... ignore me
[15:59] <sergiusens> elopio mind rewording a bit ^ ?
[15:59] <elopio> apw: we do parts before it gets to proposed, and parts after it gets to proposed. I thought that by proposing it, we were saying that we followed the process.
[16:01] <elopio> apw: I'm happy to make this clearer, just not sure how. Should I comment on every bug the tests we did before proposing it?
[16:19] <xnox> elopio, apw, sergiusens - autopkgtests do run from silos. you should get a silo assigned, upload for xenial sru, which will build and run packages.
[16:19] <xnox> then sru team can copy that into -proposed.
[16:20] <xnox> autopkgtests will run again, and things will publish in -proposed
[16:20] <xnox> and then migrate to -updates if all is good.
[16:20] <xnox> please use silos in the future, and then you can satisfy everything upfront.
[16:21] <elopio> xnox: we run the autopkgtests in jenkins. That works the same as putting the package in the silo, afaik. I'm not sure what we will win by that extra step.
[16:22] <xnox> elopio, not needing jenkins =) and the fact that SRU team has no visibility into jenkins, but do have visibility into http://autopkgtest.ubuntu.com/
[16:22] <xnox> and that you don't need to block on stable release updates team accepting your package, to see if it will be accepted. as silo assgniment is self-service.
[16:23] <xnox> and it's the same binary that will be used into the -updates pocket, as the one build in silo.
[16:23] <xnox> things built in jenkins, are obviously thrown away =)
[16:23] <elopio> xnox: so if we put the package in a silo, and it passes the autopkgtests there, it will get to proposed without any manual step?
[16:23] <xnox> elopio, there is manual step still. but there are less questions about those =)
[16:24] <xnox> as in, it's more trivial to process then opaque thing. as above.
[16:24] <xnox> *than
[16:25] <elopio> xnox: if it simplifies things, I'm not against doing that. But I still don't understand some things, like, I don't see any results from silos in autopkgtest.ubuntu.com.
[16:26] <elopio> what we are trying to explain in the SRU exception page is that no PR lands into master if the autopkgtests don't pass.
[17:26] <sergiusens> xnox I cannot dput to silos
[17:29] <xnox> sergiusens, i know i can.
[17:30] <sergiusens> xnox we are trying to move people away from the equation, not add them
[17:30] <xnox> sergiusens, do you release from git/bzr branches? silos can just take that, merge branches for you and throw things into a silo...
[17:30] <sergiusens> xnox I can dput to the archives but not that train ting
[17:30] <xnox> sergiusens, and e.g. i think it's easy to get right sto dput thing sinto silos.
[17:30] <sergiusens> xnox yes, git branches on gitub
[17:30] <xnox> sergiusens, e.g. i can dput to silos.
[17:30] <sergiusens> github
[17:31] <xnox> silos can take things from github direct, without $someone in the middle generating things and dput things
[17:31] <sergiusens> xnox I don't want to depend on people being around, I already have to chase people down to get the darn package accepted.
[17:31] <xnox> sergiusens, talk to sil2100 to get dput permissions into silos.
[17:31] <sergiusens> xnox does it still mangle changelogs and versions?
[17:31] <xnox> sergiusens, yes, that's what i'm trying to reduce too!
[17:31] <sergiusens> xnox last I talked to him he said it wasn't possible
[17:31] <xnox> silos are entirely self-service.
[17:32] <xnox> i'm sure it is, cause i can dput stuff =/
[17:32] <xnox> into silos
[17:32] <sergiusens> xnox yeah, they let core devs dput to silos, no one else
[17:32] <xnox> sergiusens, oh y, r u not core dev? =)
[17:33] <sergiusens> xnox no, just ppu
[17:41] <sil2100> sergiusens: hey! Policies changed a bit, if you have PPU access for some packages then we can give you direct-silo upload permissions
[17:41] <sil2100> sergiusens: we generally do not want to give those powers right now to people that we don't know
[17:41] <sil2100> sergiusens: let me add you in a minute
[17:49] <sergiusens> sil2100 xnox ok, I'll do that for the next upload then
[17:49] <sergiusens> I consider this one running under the agreed upon course
[17:49] <sergiusens> it would be good to see the StableReleaseUpdates wiki page mention the fact that silos are now required
[17:49] <sergiusens> slangasek ^
[17:49] <xnox> sergiusens, they are not required.
[17:50] <xnox> but they help you, satisfy your requirements.
[17:50] <sergiusens> xnox ah, then I'll probably pass. We already satisfy our requirements
[17:50]  * apw regrets muddying the water
[17:50] <xnox> general srus are one-off, for unique bugs, and hence upload to unapproved queue, wait forever, test whenever, release whenever is good enough.
[17:50] <xnox> sergiusens, well, not quite.
[17:50] <sergiusens> xnox we have an exception
[17:51] <xnox> sergiusens, silo is self-service, and will allow you to do all the testing sans las publications, and you will find your reviewes from a silo copy of binaries to be done reqlly quickly.
[17:51] <xnox> instead of always being stuck for 1-5 weeks in the SRU unapproved queue.
[17:51] <xnox> even though you do satisfy requirements.
[17:51] <sergiusens> xnox so far we've gone from initial upload to publication in 3 days
[17:52] <xnox> #toolong =)
[17:52] <infinity> xnox: Err, since when do silo SRUs get reviewed "really quickly"?
[17:52] <xnox> silos get same day publication, and you essentially can do manual testing straight away, and autopkgtest testing is done straight away.
[17:52] <xnox> infinity, you are back =)
[17:52] <infinity> xnox: The SRU team tends to hate reviewing copies. :P
[17:52] <xnox> infinity, how alive are you? =)
[17:52] <xnox> or just really good meds?
[17:53]  * apw hits the rewind time button, and keeps quiet
[17:53] <infinity> I'm about 60% alive.
[17:53]  * xnox passes infinity +5 hp healing potion
[17:54] <xnox> infinity, i guess it's up to you if you want to see bi-weekly snappy SRU as a source dput, or a copy.
[17:54] <xnox> one might not build, might fail tests, might fail autopkgtest - the other one managed to pass all that, just before a copy.
[18:00] <cjwatson> xnox: it might be a good idea to check what the SRU team wants before spending ages advising people what to do :-)
[18:00] <cjwatson> the problem with copies is mainly the age-old one that it's a pain to get diffs for them
[18:01] <cjwatson> (which is an LP bug, and so my problem, sure, but it's Hard)
[18:09] <infinity> cjwatson: Soyuz Redesign!