/srv/irclogs.ubuntu.com/2018/07/20/#snappy.txt

=== jero is now known as Guest35472
=== chihchun_afk is now known as chihchun
palassoHello, is there an RSS/Atom feed for this blog: https://snapcraft.io/blog ?06:14
=== pstolowski|afk is now known as pstolowski
pstolowskimornings06:56
mvopstolowski: hey, good morning07:00
mvopedronis: can I merge 5474 (the single task-runner PR) or does someone else should look at it first? it got two +1 now07:41
pstolowskiyay!07:41
pstolowskimvo, would you have a moment to take a look at https://github.com/snapcore/snapd/pull/5433 (it's simple)?07:42
mupPR #5433: interfaces/repo: added AllHotplugInterfaces helper <Simple> <Created by stolowski> <https://github.com/snapcore/snapd/pull/5433>07:42
mvopstolowski: sure, let me do this now07:45
pstolowskimvo: thanks!07:56
pedronismvo: yes, can be merged, yes I will do a couple follow ups to see how it feelds to make some of those now often empty methods optional07:57
pstolowskimerged08:03
mvopedronis: \o/08:04
mvohm, I want mup back08:05
pstolowskihmm indeed what happened to it?08:07
Chipacamo'in08:10
mvopstolowski: it got disabled when github was acting up some days ago08:16
pstolowskii see08:17
mvoa second review for 5542 would be great08:18
Chipacamvo: +108:25
pedronisgithub is a bit unresponsive for me today08:26
Chipacapedronis: it's the universe way of telling you to head to hr.canonical.com and ask for the day off08:27
pedronisChipaca: I have things to do also in gdocs, which work, too bad08:27
pedronis(and the forum as well)08:27
Chipacapedronis: *that*'s the universe's way of saying https://www.youtube.com/watch?v=kdOPBP9vuZA08:27
Chipacapedronis: it's the week after next that you're away for two weeks, right?08:30
pedronisyes08:31
Chipacapedronis: ok08:31
pedronisfrom the 31st to be precise08:31
pedronisI'll be around Monday 3008:31
Chipacapedronis: ack08:34
Chipacafiled go-flags#264 fwiw08:36
* Chipaca hoped mup would point that to https://github.com/jessevdk/go-flags/issues/264 but, alas08:36
kjackalHello snappy people! I facing a problem with building snaps from launchpad (LP) builders. I have a single snap, named microk8s, and I would like to release to edge from the code's master branch and to track 1.10/edge  from the code's  1.10 branch. I first created a snap for the master->edge path, but then when I select the 1.10 code branch and click "Create a new snap package" in the snap "Name: " I have to use the name "microk8s"10:05
kjackalagain and this errors with "There is already a snap package owned by microk8s developers with this name."10:05
kjackalSo, same snap release with LP builders from different code branches to different tracks is where I am loosing it.10:07
kjackalShould I go to the forum or is there something obvious I am missing?10:08
kjackalLet me ask on #launchpad as well10:10
cjwatsonI've answered on #launchpad.10:18
kjackalthanks cjwatson10:19
pedronisChipaca: mvo: I think I never noticed but it seems you cannot edit very old posts (that are not wikis) even as original author.10:29
mvopedronis: interessting, I did not notice that yet10:35
Chipacapedronis: do you have an example?10:38
pedronisChipaca: yes, I was trying to update: https://forum.snapcraft.io/t/transactionality-locking-and-other-concurrency-coordination/5010:39
pedronisno edit button for me10:39
Chipacapedronis: hmm, I see an edit button10:40
pedronisChipaca: remember you have more powers than me10:40
Chipacapedronis: but I need to expand the options to see it10:40
Chipacapedronis: that is: to the left of Reply I see like, link, and ellipsis10:41
Chipacapedronis: click the ellipsis, and in there i have edit10:41
pedronisno ellispsis for me10:41
pedronisI'm talking about the top post to be clear10:41
Chipacayep yep10:41
Chipacapedronis: and if i make it a wiki? would that help?10:42
pedronisChipaca: not sure,   does the top post and the last makes sense?  do we fear people only reading the top one and being confused because SetBlocked is going away?10:42
Chipacapedronis: if this is meant to document the topic, having half in the top and half at the bottom doesn't work, if that's what you're asking10:43
pedronisChipaca: yea, that's what I fear,  the top post sounds very descriptive/prescriptive, not narrative of a moment in time10:44
Chipacapedronis: if you reload, does it now let you edit?10:44
pedronisno10:45
Chipacapedronis: and now10:45
pedronisyes10:46
pedronisyou turned it into a wiki?10:46
Chipacapedronis: ok. let me know if you want me to dewoke it10:46
Chipacapedronis: yes. It's a reversible operation, and i can't make you admin, so ¯\_(ツ)_/¯10:46
pedronisChipaca: I have an errand very soon but I can tweak it after10:46
pedronisand see from there10:46
Chipacapedronis: ok10:47
Chipacapedronis: I've got to go to the boys' school in a bit, and i might not be back online before your eod (but i'll be on later)10:47
Chipacain fact i'm late!10:47
* Chipaca runs10:47
pstolowskimvo: hey, i'm hitting core18 failure in my disconnect-hooks branch, and i'm most likely doing something wrong in the new world. the problem i've is disconnect failure - "Disconnect core:core-support-plug from snapd:core-support (internal error: connection "core:core-support-plug snapd:core-support" not found in state)". i'm probably missing some sort of remapping in the new code; any hints?10:49
mvopstolowski: hm, this has the latest master?10:51
mvozyga: -^10:51
pstolowskimvo: well, yes, but it's my branch with master changes merged10:51
mvopstolowski: hm, hm, I need to look at this10:51
pstolowskimvo: so very likely something i'm not doinf or doing wrong10:51
pstolowskimvo: cause i've changed a lot wrt disconnects10:52
mvopstolowski: or its a bug in the remapper10:52
mvopstolowski: not unlikely as well10:52
pstolowskimvo: this is in code triggered on snap removals - i run "disconnect" tasks for all connections, and the handler expects connection in 'conns' in the state10:53
pedronispstolowski: in theory getConns and setConns should do the job for you10:53
pedronisbut I don't know if you have code by-passing them10:53
* pedronis has some relatively quick errands now10:55
pstolowskipedronis: right; no, i'm not bypassing them10:55
pstolowskialthough, maybe my lookups on conns are invalid when it comes to snap names now? i've connection refs coming from repo; i look them up in conns retrieved via getConns()11:00
pstolowskiok, i suspect i need RemapSnapToState() as I look up manually constructed conn IDs .. testing11:28
=== pstolowski is now known as pstolowski|lunch
=== pstolowski|lunch is now known as pstolowski
zygaHey12:53
zygaSorry for not finishing the PR work last night mvo12:53
mvozyga: good morning - now worries13:00
threshdoes anyone know if mozilla plans to release asan-enabled firefox via snaps?13:33
zygathresh: I don't know anything about that, perhaps popey knows14:11
popeyZyga. Sorry on vacation14:50
popeyKenvandine and willcooke  know about Mozilla things though14:50
zygathanks15:13
mbenetohi, has anybody seen kyrofa around lately?15:34
mbenetoor can anybody enable again the experimental channel "edge/classic-trusty"?15:39
phoceanhi15:40
phoceanany idea about this error when trying to launch a classic snap : "snap-confine has elevated permissions and is not confined but should be. Refusing to continue to avoid permission escalation attacks"15:40
phoceanmy guess is that it is an apparmor issue, but how to fix it ?15:40
mbenetomore details regarding the "edge/classic-trusty" channel here: https://forum.snapcraft.io/t/building-classic-snap-on-trusty-fails/5472/1315:41
phoceanit happened after I uninstalled snapd by accident because of a dependencie, and then I reinstalled it15:41
pstolowskizyga ^15:48
pedronispstolowski: the fix abour remapping still looks strange to me16:32
pedroniswe are missing something16:32
pedronismvo: we'll need to look into that ^16:42
pstolowskipedronis: thanks for keeping an eye on this16:45
pstolowskii think i will leave it at this point and revisit after i'm back from vacation16:48
mvopedronis: ok16:52
pstolowskisee you o/17:02
=== pstolowski is now known as pstolowski|off
=== mup_ is now known as mup
zygawhat's the issue about the remapping you were talking about?17:14
pedroniszyga: pawel is hitting strange bugs in his new code in disconnect-hooks, he added some strange remapping that doesn't make much sense on its own, either there's a bug in the remapping or the new code, either way we need to understand17:17
pedroniswe don't have a unit test, only failing spread tests, that the remapping that I don't understand seeems to fix17:17
zygapedronis: I see, ok17:17
zygathanks, I'll talk to pawel next week17:18
pedronispawel is off17:18
pedronisfor two weeks17:18
pedroniswe'll need to dig but not today17:18
zygaoooh17:22
zygapedronis: yeah, I see17:23
zygaI will be off on Monday17:23
zygabut then I need to task switch to fedora17:23
zygaand then I plan a week of holidays17:23
zygaso my best bet is to just work with you next week but not in full capacity17:23
zygaso that this part is fully bullet-proof17:23
pstolowski|offpedronis, zyga, mvo now that i think of it, i suspect this bug would be hit with existing discard-conns, if we had proper test. discard-conns is pretty liberal right now. with disconnect-hooks i'm essentially doing what discard-conns does, only more strict17:26
zygapstolowski|off: are we hitting the state manually there17:27
zygabecause all of connection load/store code applies mappings17:28
pstolowski|offzyga: i get conns via getConns in doDisconnect() - something we didn't do before. and i do this: https://pastebin.ubuntu.com/p/TYDZc3v8dC/ based on plug/slot refs from disconnect command17:30
pstolowski|off(the remapping part is the part that pedronis commented on above - the suspicious bit)17:31
zygainteresting17:31
zygaI don't think you need that17:31
zygayou should not need that17:31
zygaI'll look next week17:31
zygaI'm a bit tired despite it being 1:30PM17:31
pstolowski|offzyga: without that it fails17:31
zygaI see17:32
zygaI'll look next week17:32
pstolowski|offwith that - it passes17:32
zygawe have another meeting coming up now17:32
pstolowski|offat least passes the spread test i iterated on17:32
pstolowski|offzyga: sure, i'm really of now as well.17:32
pstolowski|offzyga: safe trips!17:32
zygathank you :)17:32
mvopstolowski|off: enjoy your time off!17:33
pstolowski|offthanks!17:33
mvozyga: and safe travels for you17:33
zygathank you guys17:33
zygaenjoy your weekends with your families17:33
zygaI'll check if my legs fit :)17:33
thirashello18:18
thirasgnome's default calculator doesn't start18:18
thirashttps://paste.ubuntu.com/p/TXBY2nskGz/18:19
thirashere is the error log18:19
thirasi'm really get sick of this permission denied errors (i've experienced few more of them and solved with re-installing the snap packages)18:19
thirasit's 18.0418:20
=== gurmble is now known as grumble
=== devil is now known as Guest41148
bmathHi all, having some trouble with snaps after a kernel upgrade.  Getting an error about how ptrace was denied in syslog.  I checked the apparmor config file and it looks like that capability should be allowed.  Is there a kernel setting I need to enable?23:34
=== gurmble is now known as grumble
zygabmath: hey, can you please provide the output of "snap version"23:49
zygabmath: since I'm boarding soon I won't be able to assist you please open a forum topic at forum.snapcraft.io and reference me there with @zyga syntax23:49
zygabmath: ideally please also include the name of the snap that causes the issue and the output of `dmesg | grep DENIED`23:50
zygabmath: I'll be here for a few more moments but expect any answers from me on Tuesday, I have a long way home and I will rest a little23:50

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!