[07:03] good morning [07:03] good morning [07:04] hi fgimenez [08:59] Good morning all; happy Hug Your Cat Day! 😃 [11:25] good morning. [11:26] hey fgimenez: what do you think about the latest version, ready to promote to stable? [11:27] elopio, hi, yes, seems good so far [11:28] elopio, i'm finishing the script for the complete upgrade path proposed by rsalveti, that will give us more confidence [11:28] fgimenez: nice, thank you. [12:00] rsalveti: mind doing a no change rebuild upload of the archive version of 'ubuntu-snappy' to get the powerpc binaries built? It seems we never built them [12:13] rsalveti: where was the 15.04.x blueprint? [12:13] Chipaca: https://trello.com/c/wxQzJ9uT/18-create-15-04-1-stable-image [12:13] sergiusens: wonder if we would need a no change upload [12:13] yeah, that's the one [12:13] Chipaca: and the bug task i there [12:13] sergiusens: since it was copied from vivid [12:14] k [12:14] https://launchpad.net/snappy/+milestone/15.04.1 [12:14] as well [12:14] rsalveti: yeah, it was copied from a ppa that probably didn't build for powerpc [12:14] rsalveti: yeah, I just gave him the parent link ;-) [12:14] Chipaca: from the work items, do you know of anything that can be crossed out? [12:16] sergiusens: some of them should be landing about now [12:17] Chipaca: do you know anything about the backporting ones? [12:18] sergiusens: i've jused approved several backporting MPs [12:18] s/jused/just/, and by just i mean an hour or so ago [12:18] before lunch [12:18] Chipaca: these specifically Backport lp:~jamesodhunt/ubuntu/vivid/ubuntu-core-upgrader/add-sync-calls [12:18] Chipaca: and yeah, I saw :-) [12:18] and mvo's 15.04 moar-errors one could land if he's not going to be around [12:23] sergiusens: want to do another release and I can sponsor your upload? [12:23] sergiusens: or just a rebuild is fine? [12:24] rsalveti: just a rebuild, a release forces a lot of catching up on u-d-f [12:24] sergiusens: right, sure [12:24] rsalveti: I can confirm in a bit though [12:25] to triple check why powerpc isn't there [12:25] sure, will wait then [12:26] do we need to do anything with bug #1449625 for it to be SRUable? [12:26] Bug #1449625: systemd and bin-path exported variables are not in sync [12:26] bug 1449625 in Snappy 15.04 "systemd and bin-path exported variables are not in sync" [High,Fix committed] https://launchpad.net/bugs/1449625 [12:27] marked it "done" in the trello, but then thought to ask, because it explicitly says SRU there :) [12:28] Chipaca: we should probably SRU everything that's making it into lp:snappy/15.04 [12:28] * Chipaca marks it "done" [12:30] do we really support "upgrading to edge"? [12:30] i thought that was a corner case brought about by fiddling with files [12:30] edge 15.04 or rolling? [12:41] Chipaca: we want to support upgrading to edge when we have proper channels and os/kernel snaps [12:45] sergiusens, you mentioned being able to set up tarmac for ubuntu-core-launcher? Can you also do it for snapcraft? === rickspencer3_ is now known as rickspencer3 [12:47] mterry: yeah, I saw tedg's request :-) [12:47] will do it in a bit [12:47] sergiusens, oh didn't know he asked too :) [12:58] Chipaca: nagging welcome ;-) https://code.launchpad.net/~sergiusens/snappy/YouShallNotPass/+merge/261079 [12:58] * Chipaca prepares the nag [13:02] sergiusens: +1'ed with a nit [13:04] Chipaca: nice catch ;-) [13:04] sergiusens: which makes me realise [13:04] Chipaca: pushed [13:04] sergiusens: you're not testing the use of it [13:05] uh oh :-P [13:05] Chipaca: right! Not the error, which I guess I'll do now [13:05] :) [13:08] Chipaca: o, now I need interpackage architecture stubbing, yay [13:08] ... [13:09] no, no you don't [13:09] just test with "xyzzy" architecture [13:09] that is, in the package, put architectures: foo, bar, baz [13:09] but properly ;) [13:10] Chipaca: oh, I was going to build a test snap, not test the error only ;-) [13:10] sergiusens: you're already testing that you detect the "right" arch in the unit test [13:10] sergiusens: you just need to test that you report the error in the integration test [13:11] sergiusens: yes, i meant in the package.yaml [13:11] Chipaca: right, but I rely on UbuntuArchitecture() which depends on the system where you run the test ;-) That's why I said interpackage :-) [13:11] so I'll do what I was going to do ;-) [13:12] sergiusens: not sure i follow [13:13] helpers.UbuntuArchitecture() uses the systems arch, it's all fine if we all run on amd64 but this will all fail when using a different system [13:13] meh, I understand myself ;-) [13:13] I just need to mock UbuntuArchitecture [13:13] sergiusens: um [13:13] sergiusens: i don't see that need [13:13] hence my questioning it [13:14] Chipaca: I'm a lazy typer today https://plus.google.com/hangouts/_/canonical.com/seetheneed [13:35] Chipaca: see if you like it now [13:37] sergiusens: top-approved [13:37] purrfect :) [13:38] Chipaca: thanks [13:38] Chipaca: I always spend some time figuring out which "test helper" we want :-P not sure if the same happens to you [13:46] sergiusens: "write a new one every time" is probably a suboptimal strategy [15:01] Chipaca: https://code.launchpad.net/~sergiusens/snappy/YouShallNotPassBeforeEither/+merge/261097 [15:25] rsalveti: do you think we will need to test something on the phones in the next months? [15:25] I'm wondering if I can leave them here so somebody else gets them. [15:25] elopio: you never know [15:25] elopio: I'd suggest you to keep them with you [15:25] unless requested by someone else [15:28] rsalveti: elopio you only really need one; I guess we will mostly be doing the core work (os and kernel snap) [15:28] but would leave ui work to others [15:29] right [15:30] yes. That's what I was thinking. I need a phone to call my mother too :) [15:30] I'll ask how hard would it be to get the others back if we happen to need them later. [15:55] Chipaca, ping [15:59] jdstrand, so it is it totally legit for me to let to snaps to talk to each other over localhost:whateverports? [15:59] rsalveti: how was https://bugs.launchpad.net/snappy/+bug/1424586 Fix Released? Asking in the sense of backporting ;-) [15:59] Ubuntu bug 1424586 in Snappy 15.04 "snappy-selftest: fake newer version for upgrade test" [High,Triaged] [16:00] or is that what fgimenez is working on? [16:00] if that's the case, mterry just took over the last bug in that list that is !apparmor [16:01] rickspencer3: there is language to prevent that in package.yaml but it is not enforced yet (internal vs external) [16:02] sergiusens, i'm using the upgrade script just for testing one of the upgrade paths, anyway seems to be a different approach from that bug [16:03] rickspencer3: yes, for external ports (see https://developer.ubuntu.com/en/snappy/guides/package-metadata/). for internal ports there are ideas in the vision doc to create firewall rules that could interfere [16:04] sergiusens, the bug fix implements what we were talking about in yesterday's standup, modifying channel.ini to allow an update from the current version [16:04] rickspencer3: so internal today would work too. also note, ports is in meta.md but we don't do anything with it yet afaik [16:05] fgimenez: yeah, that's in selftests iirc [16:05] rsalveti: btw, is there a card for that ^. it seems like we should at least have ingress filtering for anything in 'ports/internal' [16:05] actually, that gives me a thought [16:05] sergiusens, yep, it's one of them [16:06] I thinks ports is not implemented in part because we wanted a ports service to say whether or not something was assigned [16:07] what if we used netfilter as the database? ie, if you specify 'ports/external' then you get an ACCEPT rule with an id/comment as the PKGNAME_APPNAME and an DROP rule for external interfaces for ports/internal with an id/comment [16:08] sergiusens, rsalveti: ^ idea for whoever implements something more with 'ports' [16:08] kyrofa: pong [16:09] sergiusens, rsalveti: in this manner, snappy can parse iptables output and determine if something on the system is assigned a port [16:09] Chipaca, I'm just getting started on a dbus daemon in golang, and I see you've contributed to go-dbus. Do you recommend that lib? [16:10] kyrofa: last time i checked, it was the better of the go dbus wrappers [16:10] kyrofa: what's your daemon interfacing with? [16:10] Chipaca, the Unity8 QML progress widget [16:12] kyrofa: so you're binding qt/qml? [16:14] rickspencer3: in case it isn't obvious, if you open a port, anyone can connect to it, so things would have to be designed with that in mind [16:14] Chipaca, no, I mean that's what it's communicating with via dbus. Perhaps I misunderstood the question? [16:14] kyrofa: ah, ok :) [16:14] jdstrand: so DENY by default and if declared as an external port we add an ACCEPT and if negotiable we find a free slot and assign it there and somehow send that back to the package [16:14] kyrofa: i was just asking because if you were binding qt or glib anyway, it might make more sense to use their dbus implementation [16:14] future internal ports intend to help-- you declare an internal port and what apps can connect to it [16:15] jdstrand: but iptables doesn't solve the app1 <-> app2 problem though [16:15] sergiusens: yes, essentially [16:15] Chipaca, ahh, gotcha :) [16:15] sergiusens: how "stable" are data dirs? can i tell a guy to ship a symlink to their data dir in their app? [16:15] Chipaca, alright, I appreciate your help! I'll dive into using go-dbus :) [16:15] sergiusens: no it doesn't yet-- that would come later [16:15] jdstrand: we should probably add this to the backlog [16:15] sergiusens: that is in the vision doc [16:15] kyrofa: enjoy! holler when it breaks :) [16:16] Chipaca, ha! Will do ;) [16:16] sergiusens: it would probably make sense to look at ufw's default setup, how it sets up chains, etc for inspiration [16:17] if you want, I can help with the design [16:17] or at least review/comment on it [16:18] sergiusens: alternatively to default DROP, you setup only the rules that are specific to what is declared in ports [16:18] that might be a safer first step until we understand frameworks better [16:19] but we can talk about it [16:19] jdstrand: tracked now https://trello.com/c/p8g9U2cV/80-internal-and-external-port-acl [16:21] sergiusens: awesome (commented) [16:22] sergiusens: it probably will make sense to discuss with th architects team since there will be interactions between snaps and frameworks [16:22] sergiusens: eg, a router framework and an unrelated snap. I'd like to be invited to that call with the architects [16:26] sergiusens: so, for the vision work (additional filtering of internal ports for app to app filtering), the basic idea is create a net cgroup that tags the network traffic, and then you have rules that reference the tag [16:27] sergiusens: I think that should be done as a second step [16:38] sergiusens: ok, I'll stop talking here-- I added my thoughts to the card [16:38] since it is clear that is what you want ;P [16:43] jdstrand: oh, sorry, I'm preparing lunch ;-) [16:44] jdstrand: we can discuss here in a bit and then I'll summarize neatly in the card [16:44] but food first since it's almost 2pm and I need to eat ;-) [16:45] sergiusens: I was just teasing. Everything I have time to discuss today is in the card so I think we are good for nowe [16:45] now* [16:46] sergiusens, added a .tarmac.sh to lp:snapcraft [16:46] Missed that in lp:snappy [17:11] sergiusens: for bug 1424586, only trunk is fix released [17:11] bug 1424586 in Snappy 15.04 "snappy-selftest: fake newer version for upgrade test" [High,Triaged] https://launchpad.net/bugs/1424586 [17:11] the 15.04 task is still opened [17:21] sergiusens: can you check if we can backport the selftests as well? [17:21] then we drop the older separated branch completely [17:22] and I believe the current self tests are compatible with both 15.04 and rolling [17:23] jdstrand: sergiusens: thanks for creating the card [17:23] I think ricmm will own that from the arch side of things [17:23] ricmm: https://trello.com/c/p8g9U2cV/80-internal-and-external-port-acl [17:23] he's our network architect :-) [17:24] let me add this as a topic for the archs call to make sure we don't forget [17:51] rsalveti: yea please add that to go over it tomorrow [17:52] already did [17:52] thx [18:08] rsalveti: right, but there is no link to any code so I have no idea what work was done for that [18:08] wrt to the bug [18:08] right, mvo just added a comment pointing out the branch [18:08] but didn't link to the bug [18:08] ricmm: rsalveti I also need an 'am I online API' where online can be, lan, wan, internet [18:08] https://bugs.launchpad.net/snappy/+bug/1424586/comments/6 [18:08] Ubuntu bug 1424586 in Snappy 15.04 "snappy-selftest: fake newer version for upgrade test" [High,Triaged] [18:08] rsalveti: ok [18:09] sergiusens: hm, right [18:09] rsalveti: there, I linked to the bug [18:09] thanks [18:10] for 15.04 we hopefully can just backport the selftests merge [18:10] rsalveti: selftests though, those are cross release still, so there shouldn't be any backport work [18:10] rsalveti: since this landed in the selftest series and not in trunk [18:10] sergiusens: right, but we still need to merge under the 15.04 branch, right? [18:10] sergiusens: but thought selftests were later merged [18:10] rsalveti: I don't think selftests are in trunk yet [18:11] man, I hate this code browse interface for bzr [18:12] yeah, thought mvo had merged that [18:13] rsalveti: nope, and I don't see an MP anymore [18:13] so all good then [18:13] we'll merge for 15.04.2 I guess [18:13] rsalveti: I know there were commens from elopio in an original let's "merge it" MP [18:13] right [18:13] I guess this may have died in relation to his work into converting to go [18:13] could be [18:14] sergiusens: https://code.launchpad.net/~mvo/snappy/snappy-merge-integration-tests/+merge/259592 [18:14] yeah, still wip [18:15] @reviewlist [18:15] https://code.launchpad.net/~mterry/ubuntu-core-launcher/tmpdir-15.04/+merge/261104 | No reviews (less than a day old) [18:15] https://code.launchpad.net/~stephen-stewart/webdm/tidy-up-list-row/+merge/260837 | No reviews (1 day old) [18:15] https://code.launchpad.net/~mterry/snappy/rollback-reboot/+merge/261114 | Needs Information: 1 (less than a day old) [18:15] https://code.launchpad.net/~chipaca/snappy/mangle/+merge/260934 | No reviews (less than a day old) [18:15] https://code.launchpad.net/~mvo/snappy/snappy-lp1449032-poor-mans-rsync/+merge/260931 | Approve: 1 (less than a day old) [18:16] why is it not there? [18:25] rsalveti: the webui isn't showing it either (which is why I thought it was destroyed) [18:25] weeeeeird [18:26] hm, showing fine here [18:26] https://code.launchpad.net/~snappy-dev/snappy/snappy/+activereviews [18:26] lp:~mvo/snappy/snappy-merge-integration-tests ⇒ lp:snappy [18:26] rsalveti: oh, right, I'm looking at lp:snappy/+activereviews [18:26] rsalveti: you don't see it in the reviewlist because it's WiP [18:27] that should be @worklist ;) [18:28] oh, got it [18:28] makes sense [18:28] it's just that it's really uncommon for people to let the status as wip [18:37] I'm getting "operation not supported" and a failure to install hello-world on latest rolling image [18:38] Anyone else seeing that or is my image just borked? [18:41] iirc latest rolling is still not properly in shape [18:41] so might just be a valid bug [18:44] mterry: rsalveti that's a go 1.4 bug (from Chipaca's findings) [18:45] rsalveti: nah, this team does the right thing ;-) If it's not ready for review WiP is the right status :-) [18:48] @reviewlist [18:48] https://code.launchpad.net/~mterry/ubuntu-core-launcher/tmpdir-15.04/+merge/261104 | No reviews (less than a day old) [18:48] https://code.launchpad.net/~stephen-stewart/webdm/tidy-up-list-row/+merge/260837 | No reviews (1 day old) [18:48] https://code.launchpad.net/~chipaca/snappy/mangle/+merge/260934 | No reviews (less than a day old) [18:48] https://code.launchpad.net/~mvo/snappy/snappy-lp1449032-poor-mans-rsync/+merge/260931 | Approve: 1 (less than a day old) [18:48] https://code.launchpad.net/~mvo/snappy/snappy-oauth-quoting/+merge/260909 | Needs Fixing: 1 (less than a day old) [18:51] sergiusens, I'm looking at this rollback code [18:51] sergiusens, and actually, we output "Setting %s to version %s" AFTER we rollback [18:51] sergiusens, so there's no feedback for a bit [18:51] sergiusens, I'm going to modify my branch to move that up [18:52] mterry: meh [18:52] mterry: I think it's too late :-/ [18:53] sergiusens, I caught it :) [18:53] sergiusens, no i didn't [18:53] sergiusens, sight [18:53] oh well [18:53] mterry: just leave it, we should have a proper spinner in place [18:53] sergiusens, fair enough [18:54] mterry: we can reset trunk if you want ;-) [18:55] yes pls [18:55] mterry: ok, one sec [18:55] :) [18:56] mterry: done [18:57] sergiusens, wait what? That is a thing we do? [18:57] sergiusens, I thought you were joking [18:57] mterry: sometimes, yes [18:57] mterry: we once had someone (not naming names) do something bad with trunk [18:57] sergiusens, you crazy. OK, let me set branch status [18:57] mterry: anyways, no smilies or anything and I take you seriously ;-) [18:58] sorry [18:58] sergiusens, that's fair. I just didn't think it was a real thing, so I didn't take you seriously :) [18:58] I mean [18:58] I know you can always push --overwrite [18:58] But I thought we had protections against that [18:59] mterry: oh, yeah, I know what you mean; but yeah, it's doable and very easy to make a mistake [19:00] a common mistake, bzr branch ; cd ; bzr merge ; bzr push ; sleep (a lot); write_core; bzr commit; bzr push (oops) [19:02] mterry: it's solvable by putting trunk under a different team and only having the merge bot there [19:03] sergiusens, not sure if you are still here, but are you saying that if I try to send data over localhost:n ... that it will stop working when the security policy is fully implemented? [19:03] sergiusens, yeah I try to never reuse branch directories for that reason [19:04] mterry: I can't wait for our git migration ;-) [19:04] rickspencer3: I am [19:04] rickspencer3: and yes, it would stop if you declare your port as internal (I don't recall the spec completely from the top of my head right now) [19:08] sergiusens, ok, I'll wait until the proper implementation for socket communication [19:09] is there a standard way that folks are parsing yaml in Go? [19:10] rickspencer3: we use gopkg.in/yaml.v2 [19:10] rickspencer3: http://godoc.org/gopkg.in/yaml.v2 [19:22] rsalveti, I think we may want bug 1457183 in 15.04.1 too [19:22] bug 1457183 in Snappy Launcher 15.04 "TMPDIR is being lost for snap commands" [Undecided,In progress] https://launchpad.net/bugs/1457183 [19:23] mterry: yeah, makes sense, thanks [19:35] When is lp:snappy/selftest run? [20:05] sergiusens should know if enabled as part of the merging process [20:05] but generally we want to run it as part of proposed-migration and after the image gets published [20:05] but that is currently disabled, since we had the ps4 outage [20:06] rsalveti: I can't run it yet as part of the merging process, well I could but it would take a couple of hours I guess (vm inside a vm) [20:07] right [20:07] so the goal is to have that, eventually [20:07] mterry: so your answer is only when manually testing it [20:08] rsalveti, fair [20:08] but the plan is more [20:28] sergiusens: suggestions for "integrate()"? [20:28] sergiusens: generateIntegration()? [20:29] sergiusens: ohblahdeeOhblahdahIntegrationLalalalalalala() [20:29] sergiusens: antiderivative() [20:34] Chipaca: doStuff() [20:34] sergiusens: doPerformHelpWorker() [20:34] Chipaca: legacyIntegrationHooks? [20:35] @bugs [20:35] sergiusens: No such command! [20:35] @help [20:35] "list" To see the available commands ; "help cmd" for specific command help [20:35] well, some of it is legacy, some isn't [20:35] @help list [20:35] Lista las órdenes disponibles. [20:35] @list [20:35] The available commands are: ['bug', 'critical', 'help', 'last', 'list', 'more', 'ping', 'reviewlist', 'seen'] [20:35] @critical [20:35] yay, half translations [20:35] @bug [20:35] hmmm [20:35] bug is for “bug # yadda” [20:35] Chipaca: !hal broke [20:36] but @critical isn't [20:36] these python programs, coming here, using up all the bugs [20:36] @help bug [20:36] Search a bug and show it's id, title and status [20:36] Chipaca: yeah, I want @bugs :-P [20:36] @help critical [20:36] show the crtical bugs for self.project (could be a meta project) [20:36] verterok: @critical ne marche plus? [20:37] Chipaca: do you know about https://bugs.launchpad.net/snappy/+bug/1460152 ? [20:37] Ubuntu bug 1460152 in Snappy 15.04 "apparmor cache not updated when apparmor.d rules change (breaks 15.04/stable -> 15.04/edge updates)" [Critical,In progress] [20:37] verterok: also, why doesn't @reviewlist list everything? [20:37] ubottu: i know since i last lucked, mvo picked it up somehow? [20:37] Chipaca: I am only a bot, please don't think I'm intelligent :) [20:37] or maybe somebody assigned to it [20:37] ubottu: sorry, i meant sergiusens [20:37] Chipaca: I am only a bot, please don't think I'm intelligent :) [20:37] ubottu: i never assume that which i am interacting with is intelligent [20:37] Chipaca: I am only a bot, please don't think I'm intelligent :) [20:38] Chipaca: I am only a human, pleaase don't think I'm intelligent :-) [20:38] ubottu: it's merely a convenient abstraction [20:38] Chipaca: what do you mean it doesn't list everything? [20:38] Chipaca: I am only a bot, please don't think I'm intelligent :) [20:38] verterok: compare [20:38] @reviewlist [20:38] https://code.launchpad.net/~mterry/ubuntu-core-launcher/tmpdir-15.04/+merge/261104 | No reviews (less than a day old) [20:38] https://code.launchpad.net/~stephen-stewart/webdm/tidy-up-list-row/+merge/260837 | No reviews (1 day old) [20:38] https://code.launchpad.net/~mterry/snappy/rollback-reboot-15.04/+merge/261135 | No reviews (less than a day old) [20:38] https://code.launchpad.net/~mvo/snappy/snappy-lp1460152-workaround/+merge/261144 | No reviews (less than a day old) [20:38] https://code.launchpad.net/~chipaca/snappy/mangle/+merge/260934 | No reviews (less than a day old) [20:38] verterok: and code.launchpad.net/snappy/+activereviews [20:38] https://code.launchpad.net/~chipaca/snappy/removeClick/+merge/260560 | Needs Information: 1 (5 days old) [20:38] https://code.launchpad.net/~mvo/snappy/snappy-lp1459212-check-required-fields/+merge/260509 | Needs Fixing: 1 (6 days old) [20:38] https://code.launchpad.net/~sergiusens/snappy/frameworkPath/+merge/260202 | Needs Information: 1, Approve: 1 (8 days old) [20:38] https://code.launchpad.net/~mvo/snappy/snappy-more-errors-15.04/+merge/260111 | Approve: 1 (8 days old) [20:38] https://code.launchpad.net/~mvo/snappy/snappy-merge-integration-tests/+merge/259592 | Needs Information: 1, Approve: 1 (14 days old) [20:39] verterok: in particular, two of mvo's branches and two of my branches don't show up [20:39] Chipaca: help me here, which ones? :) [20:40] verterok: of mine, lp:~chipaca/snappy/setActiveClick and lp:~chipaca/snappy/unsetActiveClick [20:40] verterok: of mvo, dunno, i just counted [20:40] * Chipaca is bad at making set differences [20:41] Chipaca: k, one example is ok [20:41] verterok: this is without looking at mvo's git one :) [20:42] yeah, no git branches/repo support last time I checked launchpadlib [20:42] no worries [20:42] * Chipaca hopes that will come [20:43] sergiusens: before i started the ubottu nonesense, did you read my reply? [20:43] Chipaca: I think rsalveti said it was on cjwatson's list [20:43] Chipaca: about me not being intelligent? [20:44] Chipaca: or this new goodie? https://code.launchpad.net/~mvo/snappy/snappy-lp1460152-workaround/+merge/261144 [20:44] sergiusens: yes. I mean no, about mvo being on the bug [20:44] Chipaca: which shows up on the list ;-) [20:44] yeah, that one [20:44] sergiusens: noticed it suddenly had a patch and a branch on the milestone list [20:45] Chipaca: wrt to integrate() add some comment or the word legacy there, we are planning on getting rid of it, right? [20:45] sergiusens: we had a drive-by mvo it seems [20:45] sergiusens: maybe :) it's mostly for generating the click manifest, so yes [20:45] Chipaca: I know of people that had enough mana to summon him :-) [20:47] i found a good place to have a sprint! [20:47] http://www.waterwaysholidays.com/detail/alvdove15.htm [20:47] jdstrand, where is docker packaging? I wanted to get the same error you did in bug 1442410 [20:47] bug 1442410 in Snappy 15.04 "insufficient logging with snappy install errors" [High,Triaged] https://launchpad.net/bugs/1442410 [20:48] mvo: boo! [20:51] hey Chipaca [20:51] Chipaca: how are you? [20:51] me? I'm fine. My irc client doesn't crash all the time. [20:51] lol [20:56] so, given http://www.washingtonpost.com/blogs/the-switch/wp/2015/06/04/fbi-official-companies-should-help-us-prevent-encryption-above-all-else/ [20:56] when do we start encrypting snappy? [21:00] Chipaca: it seems the bot was in a weird state, kicked it a bit [21:00] @reviewlist [21:00] https://code.launchpad.net/~mterry/ubuntu-core-launcher/tmpdir-15.04/+merge/261104 | No reviews (less than a day old) [21:00] https://code.launchpad.net/~stephen-stewart/webdm/tidy-up-list-row/+merge/260837 | No reviews (1 day old) [21:00] https://code.launchpad.net/~mvo/snappy/snappy-lp1460152-workaround-15.04/+merge/261148 | No reviews (less than a day old) [21:00] https://code.launchpad.net/~mvo/snappy/snappy-lp1460152-workaround/+merge/261144 | No reviews (less than a day old) [21:00] https://code.launchpad.net/~chipaca/snappy/mangle/+merge/260934 | No reviews (less than a day old) [21:00] * Chipaca kicks nothal [21:00] * nothal kicks Chipaca [21:01] @reviewlist [21:01] https://code.launchpad.net/~mterry/ubuntu-core-launcher/tmpdir-15.04/+merge/261104 | No reviews (less than a day old) [21:01] https://code.launchpad.net/~stephen-stewart/webdm/tidy-up-list-row/+merge/260837 | No reviews (1 day old) [21:01] https://code.launchpad.net/~mvo/snappy/snappy-lp1460152-workaround-15.04/+merge/261148 | No reviews (less than a day old) [21:01] https://code.launchpad.net/~mvo/snappy/snappy-lp1460152-workaround/+merge/261144 | No reviews (less than a day old) [21:01] https://code.launchpad.net/~chipaca/snappy/mangle/+merge/260934 | No reviews (less than a day old) [21:02] Chipaca: something is broken. sorry. will redeploy it [21:02] k... [21:03] @critical [21:04] sergiusens: afaict lp:~jamesodhunt/ubuntu/vivid/ubuntu-core-upgrader/add-sync-calls and lp:~jamesodhunt/ubuntu/vivid/ubuntu-core-upgrader/use-os-sync [21:04] * verterok kicks harder [21:04] hey Chipaca, sorr,y looks like my network is a bit unreliable [21:04] sergiusens: are meregd already into ubuntu:vivid/ubuntu-core-upgrader [21:04] Chipaca: did you say anything? [21:05] Chipaca: how are you? [21:05] * mvo has quit (Client Quit) [21:05] me? I'm fine. My irc client doesn't crash all the time. [21:05] lol [21:05] my irc client is fine, my network is not ;) [21:05] fair enough [21:05] I heard ubuntu-core-upgrader, whats up with that? [21:05] seems to be ok now though (the network), fingers crossed ;) [21:05] mvo: afaict lp:~jamesodhunt/ubuntu/vivid/ubuntu-core-upgrader/add-sync-calls and lp:~jamesodhunt/ubuntu/vivid/ubuntu-core-upgrader/use-os-sync are meregd already into ubuntu:vivid/ubuntu-core-upgrader [21:06] cool, yeah, quite possible [21:06] * mvo can't remember but it rings a bell [21:06] mvo: that is, i bzr branch'ed ubuntu:vivid/ubuntu-core-upgrader, and merging those two branches gave me no diff [21:07] mvo: does this mean the task wrt the upgrader for 15.04.1 is done? [21:07] mvo: or is there more to it? [21:07] * Chipaca has no idea [21:07] Chipaca: let me quickly double check [21:07] mvo: should i be letting you work at all? [21:08] Chipaca: heh :) [21:09] Chipaca: so its merged but it seems like its not uploaded yet, I will push to the PPA for now and we need a proper SRU [21:09] mvo: ok. I thought it landed in ubuntu:yadda after being in the archive proper, but obviously not [21:14] Chipaca: I land it now [21:16] * rsalveti waves to mvo [21:16] yeah, we can take care of the SRU side later [21:17] guess that would include core-upgrader and core-launcher [21:18] hey rsalveti [21:18] rsalveti: yeah [21:18] sergiusens, you said earlier that install problems on rolling were "a go 1.4 bug (from Chipaca's findings)" [21:18] sergiusens, is there a workaround or a bug to track? [21:18] mterry: well, well [21:18] mterry: there is no bug for the 1.4 thing afaik, maybe there should be [21:19] mterry: we'd have to check with a "proper" go 1.4 first [21:19] as this was just with my locally-compiled go 1.4, which might have other issues [21:19] (it hasn't been a problem before, but) [21:19] mterry: what exactly are you seeing? [21:20] Chipaca, when I try to "sudo snappy install" something, I get "operation not supported" and then a message about unpack from /tmp to the real place failed [21:20] ah, yes, that sounds exactly like the thing that happened with my 1.4 [21:20] Chipaca, mterry: AIUI lp:~mterry/ubuntu-core-launcher/tmpdir-15.04 needs to land in the PPA as well (its not yet, correct?) [21:21] mvo: correct [21:21] mvo, I'd like it to land yeah [21:21] cool, shall I do the merge/upload-to-ppa dance or is someone already on it? [21:22] (sorry, I'm a bit out of the loop) [21:22] I've +1'ed and top-approved [21:22] one other we need review https://code.launchpad.net/~mterry/ubuntu-core-launcher/tmpdir-15.04/+merge/261104 [21:22] and then upload as well [21:22] rsalveti: that's the one mvo just asked about [21:22] rsalveti: and the review is done [21:22] indeed, sorry, was going over my tabs [21:22] (as i'd reviewed the original one, it was quick) [21:23] mvo: I can do as well, you should not even be here :-) [21:24] the workaround for the apparmor mr should be automatically merged and uploaded, so that is fine [21:24] guess we just need to create the backporting branch once that lands in trunk [21:26] Chipaca: lol, that article would be a good one for the onion [21:26] rsalveti: :-/ [21:27] rsalveti: I have a backport branch pushed as well :) [21:27] * rsalveti hugs mvo [21:28] rsalveti: I can do the merge/upload, thats fine (unless you already started that or I duplicate the work of someone else) [21:28] didn't yet start, unless Chipaca is on it, I don't think anyone is on it yet [21:29] i can't upload, i don't think [21:29] you should be able to [21:29] you're part of ~snappy-dev [21:30] Excellent. [21:30] * Chipaca did not just do his mr burns impression [21:30] I'll trigger another image once everything is there [21:30] and hopefully we should have our first RC [21:32] mvo, I saw your comment about helpers/, which I also helped make worse with my EnvVar stuff [21:32] mvo, maybe we need a second "general" package [21:32] mterry: what was the comment? [21:32] mterry: yeah! or maybe I'm overthinking this, I'm not sure yet :) [21:32] on lp:~sergiusens/snappy/YouShallNotPass merge, just saying a func maybe doesn't belong in helpers/ [21:32] Chipaca, ^ [21:33] but I would love to have a tiny bit more structure in helpers as its a bit of a misc/ dir right now and at some point it was meant to be stuff-missing-in-stdlib or something like that [21:33] Chipaca, so wait. About the go1.4 bug. no workaround yet? Should I just go back to 15.04 for testing? [21:34] mterry: given the bug, yes, go back to 15.04 [21:34] especially as that's what we're releasing rsn :) [21:35] mterry: or, compile snappy with 1.3 for W [21:35] mterry: until we've sorted that one out [21:35] rsalveti: once the workaround branch lands (assuming it gets a ok fromthe reviewers) we need to trigger the daily build for that branch, let me quickly search the url for that [21:35] Chipaca, pfft, 15.04 is old hat, even if we are shipping a point release :) [21:35] go's "compatibility promise", on which they base not supporting releases for more than ~6 months, has bit us before now, so [21:35] rsalveti: https://code.launchpad.net/~snappy-dev/+recipe/snappy-daily/ <- thats the page [21:36] * mvo wishes we had git-dch for bzr or git so that it would auto-create the debian/changelog from the vcs comments [21:41] mvo: awesome, thanks [21:42] yeah [21:45] Chipaca: can I just cross lp:~jamesodhunt/ubuntu/vivid/ubuntu-core-upgrader/add-sync-calls and lp:~jamesodhunt/ubuntu/vivid/ubuntu-core-upgrader/use-os-sync ? [21:46] rsalveti: AIUI if you're asking, no [21:46] rsalveti: mvo said he could do it, you told him you could and that he shouldn't be here [21:46] rsalveti: or maybe i misunderstood [21:47] * mvo is confused now ;) [21:47] Chipaca: sorry, you said earlier that you tried merging jhunt's branches and got a 0 diff [21:47] * rsalveti reads backlog [21:48] yeah, the use-os-sync was merged but not uploaded, I uploaded that now to the ppa and to wily [21:48] yeah, then all good [21:49] crossing then [21:49] https://launchpad.net/~snappy-dev/+archive/ubuntu/image/+packages should have the info looks like its indeed there [21:50] awesome [21:50] Chipaca: I need to buy you tea (or beer) for all the silly mistakes you need to endure in my branches, thanks a lot for spotting these issues [21:51] * Chipaca blushes [21:51] mvo: you've put up with a lot of nonsense in my branches, glad to be finally paying it back :D [21:51] lol [21:52] still have to learn to do reviews like yours though; almost makes me want to put more mps up just to get them :) [21:53] ANYway, i should put down the computer and go wash the dishes and get the house ready for the night [21:53] Chipaca: a more serious question about the oauth signing branch, you point out (correctly) that buf.Grow(len(inputString)*3) may be too small and I need to count the bytes. should I simply use len(s)*3 (quotes) *4 (utf-8 len) - or convert the str to bytes before checking the len? [21:53] mvo: you're converting it to bytes anyway [21:54] mvo: might as well do it earlier [21:54] Chipaca: do that, I will call it a day soon too [21:54] Chipaca: yeah, sounds sensible [21:54] Chipaca: thanks a bunch [22:17] rsalveti: its hopefully all prepared for the release-candidate, just ping me again if there is anything that needs work. I will keep a eye on my mail to see if my MPs get approved [22:17] mvo: sure, thanks so much [22:19] your welcome! [22:19] * mvo & [23:46] sergiusens: https://code.launchpad.net/~mvo/snappy/snappy-lp1460152-workaround/+merge/261144 would only work if the base image (stable) already had such changes, right? [23:47] it seems our problem is that we'll be updating from stable to edge using the tools that are currently available at stable [23:47] so it would only work after doing another update [23:48] guess that's the usual problem of updating using the tools available from the older image (like we had with touch) [23:53] added a comment to the MR (for lp:snappy)