| mborzecki | morning | 05:12 |
|---|---|---|
| mbeneto | morning | 05:26 |
| mborzecki | mvo: morning | 06:28 |
| mvo | mborzecki: good morning! | 06:31 |
| === abeato_ is now known as abeato | ||
| mborzecki | mvo: shall we land #5122 ? | 06:40 |
| mup | PR #5122: snap: add support for `snap advise-snap --from-apt` <Created by mvo5> <https://github.com/snapcore/snapd/pull/5122> | 06:40 |
| mvo | mborzecki: let me quickly double check | 06:45 |
| mvo | mborzecki: does it have two +1 already? | 06:45 |
| mborzecki | mvo: yup | 06:47 |
| mborzecki | mvo: 2.34.2 is tagged but it's not 'released' on github | 07:46 |
| mvo | mborzecki: yeah, let me fix that, its just test fixes but still let me fix it | 07:47 |
| mborzecki | google:ubuntu-core-16-64:tests/main/unhandled-task failed on master and journal is basically Jul 23 07:56:56 jul230742-893813 audit[8830]: AVC apparmor="DENIED" operation="accept" profile="snap.network-bind-consumer.network-consumer" pid=8830 comm="python3" laddr=127.0.0.1 lport=8081 family="inet" sock_type="stream" protocol=6 requested_mask="accept" denied_mask="accept" repeated over and over again | 08:18 |
| mborzecki | thought that the patches cleaning up snap.network-bind-consumer.network-consumer were landed already? | 08:18 |
| mborzecki | or not, merged #5477 just now | 08:20 |
| mup | PR #5477: tests: change the service snap used instead of network-bind-consumer <Created by sergiocazzolato> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/5477> | 08:20 |
| mvo | mborzecki: thank you for this | 08:26 |
| Chipaca | moin moin | 08:34 |
| Chipaca | mborzecki: welcome back! | 08:34 |
| mborzecki | Chipaca: hey | 08:35 |
| Chipaca | mborzecki: are you appropriately revitalized | 08:35 |
| mborzecki | Chipaca: i still think that vacation snould be at least 3 weeks long | 08:35 |
| Chipaca | mborzecki: I see no problem with that | 08:36 |
| Chipaca | mborzecki: (that's the way nearly everybody does it in Argentina) | 08:36 |
| mborzecki | Chipaca: here it's usually 2 weeks :( | 08:38 |
| Chipaca | mborzecki: ah, duration varies, i meant most everybody takes it in a single uninterrupted block | 08:38 |
| Chipaca | indeed if memory serves the first full year you're working somewhere you get about a week | 08:39 |
| mvo | hey Chipaca ! good morning | 08:41 |
| Chipaca | on a completely unrelated note, the person behind https://play.jsgo.io/ just let me know they bumped the limit on git objects, so you can now import snapd in there | 08:41 |
| mborzecki | Chipaca: the whole yearly allowance here is 26 days, but people usually take at least one 2 weeks block during summer (probablyl because that's also a minimum required by law when you're on permanent contract) | 08:41 |
| Chipaca | https://play.jsgo.io/375878919304e8143970831774c69ee44f2b204a | 08:41 |
| Chipaca | ^ that's snapd running in the browser :) | 08:42 |
| Chipaca | allow me to find the biggest quotes in the uiverse to put around the 'running' word, there | 08:42 |
| mborzecki | next step, snapd in wasm | 08:42 |
| mvo | Chipaca: woah, fun | 08:43 |
| Chipaca | mborzecki: snapd for android watches | 08:43 |
| mborzecki | on android wear | 08:43 |
| Chipaca | yus | 08:44 |
| Chipaca | pedronis: are you around? | 08:46 |
| pedronis | Chipaca: yes | 08:50 |
| Chipaca | pedronis: hiya :-) | 08:50 |
| mborzecki | pedronis: hi | 08:51 |
| Chipaca | pedronis: I anwered myself already, but just to make sure: managers should no longer have a KnownTaskKinds, and their Ensure/Wait/Stop should no longer call the same thing on the runner, right? (and they shouldn't really be holding on to the runner anyway) | 08:51 |
| pedronis | Chipaca: yes, no local runner, they should take one in | 08:52 |
| pedronis | just look at any one | 08:52 |
| pedronis | Chipaca: also I might do a follow up today that makes Wait/Stop on them optional | 08:52 |
| Chipaca | yes, I was confused by what's probably a bad merge, but looked at master for a clear picture | 08:52 |
| Chipaca | pedronis: thanks | 08:53 |
| mvo | sil2100: mup is currently not working, so I play mup :) https://github.com/snapcore/core18/pull/47 is new and pretty trivial | 09:51 |
| mup | PR core18#47: hooks: preserve /usr/local/share/fonts <Created by mvo5> <https://github.com/snapcore/core18/pull/47> | 09:51 |
| * Chipaca wanders off to ponder lunch | 11:04 | |
| pedronis | mvo: Chipaca: some recent change is making the tests print out squashfs output: Parallel unsquashfs: Using 4 processors ... | 11:14 |
| pedronis | that's a bit annoying | 11:14 |
| pedronis | mvo: Chipaca: try for example the overlord tetss | 11:15 |
| pedronis | tests | 11:15 |
| === chrisccoulson_ is now known as chrisccoulson | ||
| mvo | pedronis: oh, interessting. I saw this as well but thought it was because I had a locally modified version of squashfs | 11:22 |
| pedronis | Chipaca: is that an affect of passing -no-progress ? | 11:28 |
| Chipaca | hmmmm | 12:02 |
| Chipaca | pedronis: mvo: I'll take a look | 12:02 |
| Chipaca | it might be we're setting os.Stdout where before we weren't? | 12:03 |
| Chipaca | I thought I'd undone those for-testing changes but maybe I missed something | 12:03 |
| Chipaca | +cmd.Stdout = os.Stderr | 12:03 |
| Chipaca | darn | 12:03 |
| Chipaca | mvo: pedronis: my bad: https://github.com/snapcore/snapd/pull/5494/files#diff-f440ed51e6e7c93f4920731345213192R147 | 12:04 |
| mup | PR #5494: snap/squashfs, tests: pass -n[o-progress] to {mk,un}squashfs <Created by chipaca> <Merged by stolowski> <https://github.com/snapcore/snapd/pull/5494> | 12:04 |
| mborzecki | Chipaca: some conflicts in #5506 | 12:09 |
| mup | PR #5506: cmd/snap: add a green check mark to verified publishers <Created by chipaca> <https://github.com/snapcore/snapd/pull/5506> | 12:09 |
| Chipaca | mborzecki: yes. Was hoping to get a second review before deconflicting yet again | 12:09 |
| Chipaca | mborzecki: HINT HITN NUDGE NUDGE | 12:10 |
| mborzecki | understood :) | 12:10 |
| mvo | does "configure: error: C compiler cannot create executables | 12:21 |
| mvo | See `config.log' for more details" ring any bells? release/2.34 is currently failing with that it seems | 12:21 |
| mborzecki | mvo: did it come up again? | 12:22 |
| mvo | mborzecki: yes, in my PR from this morning | 12:22 |
| mborzecki | mvo: builfing on fedora now | 12:30 |
| mborzecki | maybe something changes with the deps | 12:30 |
| mvo | mborzecki: you mean you are trying the build? what is also odd is that master seems to be ok | 12:31 |
| mborzecki | configure: error: in `/root/rpmbuild/BUILD/snapd-1337.2.34~pre1/cmd': | 12:31 |
| mborzecki | configure: error: C compiler cannot create executables | 12:31 |
| mborzecki | heh | 12:31 |
| mvo | mborzecki: yeah, thats the error I get (my version number is slightly different though) | 12:32 |
| mborzecki | mvo: running cmd/autogen.sh works though | 12:33 |
| mborzecki | configure:2740: gcc -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -fexceptions -fstack-protector-st | 12:40 |
| mborzecki | rong -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -m64 -mtune=generic -fasyn | 12:40 |
| mborzecki | chronous-unwind-tables -fstack-clash-protection -mcet -fcf-protection -Wl,-z,relro -z now conftest.c >&5 | 12:40 |
| mborzecki | annobin: conftest.c: Error: plugin built for compiler version (8.0.1) but run with compiler version (8.1.1) | 12:40 |
| mborzecki | cc1: error: fail to initialize plugin /usr/lib/gcc/x86_64-redhat-linux/8/plugin/annobin.so | 12:40 |
| mborzecki | configure:2744: $? = 1 | 12:40 |
| mborzecki | mvo: ^^ | 12:40 |
| mvo | mborzecki: woah, what is annonbin | 12:42 |
| mborzecki | cachio: can you update fedora 28 image? | 12:42 |
| mborzecki | mvo: no clue, Son_Goku do you have an idea what that might be? | 12:43 |
| Son_Goku | is this rawhide? | 12:44 |
| Son_Goku | also, god damn it annobin | 12:44 |
| mborzecki | Son_Goku: 28 | 12:44 |
| * Son_Goku growls in anger | 12:45 | |
| Son_Goku | mborzecki, is the f28 image fully up to date? | 12:45 |
| mborzecki | Son_Goku: i'm afraid not, just pulled in >200MB of updates and retrying the build | 12:45 |
| Son_Goku | you might be out of sync with gcc + annobin | 12:45 |
| Son_Goku | mvo, mborzecki: https://fedoraproject.org/wiki/Changes/Annobin | 12:46 |
| Chipaca | mvo, pedronis: #5547 | 12:47 |
| mup | PR #5547: snap/squashfs: stop printing unsquashfs info to stderr <Created by chipaca> <https://github.com/snapcore/snapd/pull/5547> | 12:47 |
| Chipaca | heh | 12:47 |
| Chipaca | pedronis: fastest +1 in the west | 12:47 |
| Son_Goku | the compiler team has developed a number of plugins to enforce security hardening, expand debuginfo, and include extra information for auditing purposes | 12:47 |
| Son_Goku | annobin is one of those plugins | 12:47 |
| mborzecki | Son_Goku: interesting | 12:48 |
| pedronis | Chipaca: mvo: #5546 | 12:48 |
| mup | PR #5546: many: make Wait/Stop optional on StateManagers <Created by pedronis> <https://github.com/snapcore/snapd/pull/5546> | 12:48 |
| Chipaca | :-D | 12:48 |
| mborzecki | mvo: the package builds after dnf upgrade | 12:48 |
| mvo | pedronis: cool, I check in a bit | 12:48 |
| mvo | mborzecki: ok, so cachio just needs to update the image? | 12:48 |
| Son_Goku | annobin has the crappy requirement of being ABI sensitive to gcc | 12:48 |
| Son_Goku | it has to use the same gcc it was built for | 12:48 |
| mborzecki | mvo: yes, given what Son_Goku wrote we need to update it each time gcc is updated | 12:49 |
| cachio | mvo, mborzecki Son_Goku I'll update images today | 12:49 |
| mvo | cachio: thank you | 12:50 |
| mborzecki | Son_Goku: hm another intersting thing, annonbin does not seem to depend on gcc | 12:52 |
| Son_Goku | mborzecki, no, strictly speaking, it's a requirement from redhat-rpm-config | 12:52 |
| sil2100 | mvo: looking! :) | 12:54 |
| Son_Goku | redhat-rpm-config has the following: | 12:54 |
| Son_Goku | Requires: (annobin if gcc) | 12:54 |
| Son_Goku | which is a rich dep that imposes annobin to be installed if gcc is | 12:54 |
| mborzecki | Son_Goku: but does not seem to enforce any particular version | 12:55 |
| Son_Goku | mborzecki, yeah, annobin should have a requirement to be run with the gcc it was built from, but it doesn't :/ | 13:02 |
| Son_Goku | hmm, it does | 13:03 |
| mborzecki | Son_Goku: hm maybe it's implicit when using rich dependency https://paste.fedoraproject.org/paste/2Pdj7r~fbdHIj4jzae6kCA/raw | 13:07 |
| * zyga waves | 13:07 | |
| Son_Goku | hey zyga | 13:08 |
| zyga | trying to catch up with my sleep credit | 13:08 |
| mborzecki | zyga: hey | 13:08 |
| zyga | anything on fire or can I get back to sleep? | 13:08 |
| Son_Goku | zyga, you know we only have two weeks until flock, right? | 13:09 |
| zyga | yes | 13:09 |
| Son_Goku | I still need to make the presentation and you need to at least make a demo, if not prepare some basic code for plugging base snap building into fedora tools | 13:09 |
| zyga | I will work on that this week | 13:10 |
| zyga | next week I'd like to take a week off as that's the last chance for spending any time with my wife without a laptop | 13:10 |
| Son_Goku | mborzecki, it looks like an error in the annobin packaging means that the version dep is not on annobin itself | 13:11 |
| Son_Goku | it's on annobin-tests | 13:11 |
| * zyga returns to slumber | 13:13 | |
| mborzecki | Son_Goku: can you file a bug? | 13:19 |
| Son_Goku | yep | 13:20 |
| mborzecki | Son_Goku: thank you | 13:20 |
| Chipaca | what's up with travis eating logs? | 14:33 |
| Chipaca | hmm | 15:14 |
| * Chipaca hmm'ing | 15:14 | |
| Chipaca | master seems broken: if I run the unit tests of cmd/snap, there's a good chance it breaks in the aliases tests | 15:16 |
| Chipaca | with a connection error | 15:16 |
| Chipaca | (?!?) | 15:16 |
| Chipaca | cachio: is this one of the ones you are chasing? | 15:17 |
| cachio | Chipaca, no | 15:19 |
| cachio | Chipaca, working with an opensuse repo error | 15:19 |
| Chipaca | ok | 15:19 |
| cachio | I see unit tests failing on master now | 15:20 |
| cachio | Chipaca, https://travis-ci.org/snapcore/snapd/builds/407066445#L2043 | 15:21 |
| cachio | this is the one | 15:22 |
| cachio | I already updated opensuse image, so it should be gone | 15:22 |
| mvo | cachio: looking | 15:26 |
| mvo | cachio: the "E: Could not read response to hello message from hook [ ! -f /usr/bin/snap ] || /usr/bin/snap advise-snap --from-apt || true: Connection reset by peer" bug has an open PR, one sec | 15:27 |
| cachio | mvo, I saw this one https://travis-ci.org/snapcore/snapd/builds/407177600#L3637 | 15:28 |
| mvo | cachio: the apt error should be fixed with https://github.com/snapcore/snapd/pull/5548#partial-pull-merging | 15:29 |
| mup | PR #5548: debian: ensure dependency on fixed apt on 18.04 <Created by mvo5> <https://github.com/snapcore/snapd/pull/5548> | 15:29 |
| mvo | niemeyer: btw, could you please re-enable mup for our github repos? AIUI github issues are fixed by now | 15:29 |
| mvo | cachio: hm "... value client.ConnectionError = client.ConnectionError{error:(*url.Error)(0xc820192030)} ("cannot communicate with server: Post http://127.0.0.1:39530/v2/aliases: read tcp 127.0.0.1:44800->127.0.0.1:39530: read: connection reset by peer")" is strange, I don't think this has changed in a long time | 15:31 |
| mvo | cachio: could you please ping me once the f28 images are good again? then I will trigger #5545 again :) | 15:32 |
| mup | PR #5545: snapstate: allow setting "refresh.timer=managed" (2.34) <Created by mvo5> <https://github.com/snapcore/snapd/pull/5545> | 15:32 |
| cachio | mvo, I am gonna update them now | 15:32 |
| hunterk | mborzecki: Hi there, I was told you may be interested in vulkan+snap testing/examples | 15:36 |
| mvo | cachio: thank you! | 15:37 |
| cachio | mvo, fedora 28 is updated | 15:56 |
| cachio | mvo, running tests now | 15:56 |
| cachio | I'll check after lunch | 15:56 |
| cachio | mvo, so far so good | 15:56 |
| === chihchun is now known as chihchun_afk | ||
| mvo | looks like we need to put the seed.yaml reodering higher on our todo :/ germintae reorder things so our code needs to ensure the right order | 16:07 |
| cachio | mvo, do you want to check if it is everything ok? | 16:12 |
| mvo | cachio: yeah, looking | 16:13 |
| cachio | tx | 16:13 |
| * Chipaca runs to the shops before they close | 16:20 | |
| Pharaoh_Atem | mborzecki: https://bugzilla.redhat.com/show_bug.cgi?id=1607430 | 16:20 |
| cachio | mvo, tests passed sucessfully on fedora-28 | 16:20 |
| mvo | cachio: nice | 16:22 |
| mvo | kyrofa: do you know about the current rust snapcraft testsuite failures? | 16:22 |
| mvo | kyrofa: there is a discussion in #ubuntu-devel right now and whether or not snapd has anything to do with those | 16:22 |
| === chihchun_afk is now known as chihchun | ||
| pedronis | mvo: ? for classic or in general? at the moment prepare-image will not even grab bases or default-provider | 17:26 |
| niemeyer | mvo: Good point, thanks for the reminder | 17:31 |
| mborzecki | Pharaoh_Atem: thanks, already subscribed to the bug :) | 18:18 |
| === chihchun is now known as chihchun_afk | ||
| Chipaca | ok, giving up for today | 20:02 |
| Chipaca | but something is still broken :-( | 20:02 |
| Chipaca | several somethings actually because the more i try to pin it down the more other things break | 20:02 |
| Chipaca | grr | 20:03 |
| Chipaca | anyhow, tomorrow. | 20:03 |
| shantanoo | hi all, is there a latest clamav snap? | 20:07 |
| shantanoo | tried building one, but after install got '/snap/clamav/current/usr/local/bin/freshclam: error while loading shared libraries: libclamav.so.7: cannot open shared object file: No such file or directory' | 20:08 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!