mwhudson | another mystery https://autopkgtest.ubuntu.com/results/autopkgtest-jammy/jammy/armhf/d/dolfin/20211108_223753_d075a@/log.gz | 00:38 |
---|---|---|
mwhudson | (that one went away on retry) | 01:58 |
mwhudson | another another wtf https://autopkgtest.ubuntu.com/results/autopkgtest-jammy/jammy/amd64/p/pythonmagick/20211108_193907_5e147@/log.gz | 01:58 |
sarnold | TypeError: 'Boost.Python.class' object is not iterable | 01:59 |
sarnold | boost .. python? wat? | 01:59 |
mwhudson | yeah and the error is coming from the stdlib enum.py?? | 02:00 |
mwhudson | https://paste.ubuntu.com/p/SNpMtzgdnc/ <- something be busted bad | 02:04 |
sarnold | oh my | 02:05 |
mwhudson | can we apply the infinity gauntlet to the packages in ubuntu perhaps | 02:11 |
mwhudson | checking whether boost_python310 is the correct library... yes | 02:14 |
mwhudson | i would like to offer a second opinion | 02:14 |
mwhudson | ginggs: so pythonmagick linked its python3.9 extension against the python3.10 boost_python library | 02:16 |
mwhudson | ginggs: i think thanks to more autoconf nonsense | 02:17 |
mwhudson | also in general --with-python-min-version={version.major}.{version.minor} seems a risky way to select a python version, i hope there aren't too many packages using it | 02:17 |
mwhudson | i think a build of pythonmagick where python3.10 was default but python3.9 was still supported would build a python3.10 extension twice | 02:18 |
mwhudson | probably better to set PYTHON in the environment | 02:20 |
mwhudson | ginggs: what do you think about https://launchpad.net/~mwhudson/+archive/ubuntu/devirt/+packages?field.name_filter=sphinxbase ? | 02:50 |
=== genii is now known as genii-core | ||
ginggs | mwhudson: pythonmagick seems to have had issues building for two supported python versions, see debian #954812 | 04:29 |
ubottu | Debian bug 954812 in src:pythonmagick "pythonmagick: autopkgtest regression: cannot import name '_PythonMagick'" [Important, Open] https://bugs.debian.org/954812 | 04:29 |
ginggs | mwhudson: sphinxbase looks good to me, shipit! | 04:37 |
mwhudson | ginggs: re pythonmagick: hmm | 07:10 |
mwhudson | ginggs: sphinxbase uploaded | 07:11 |
ginggs | mwhudson: thanks! | 07:12 |
* mwhudson stares at reverse-depends src:pythonmagick | 07:13 | |
mwhudson | maybe we can just bodge it to build for the default python? | 07:13 |
mwhudson | hmm not sure that will help actually | 07:13 |
mwhudson | ginggs: do you know who is looking after the r stuff? | 07:34 |
mwhudson | i find working on r packages pretty incomprehensible so i'm happy to look into specific things but not sure i can find generally useful things to do myself | 07:35 |
ginggs | mwhudson: it's been a group effort; vorlon, xnox and me | 07:35 |
ginggs | i think it's just about ready to migrate, just needs some retrying of autopkgtests now | 07:36 |
ginggs | mwhudson: building pythonmagick for only the default python sounds reasonable to me. the autopkgtest will need to be adjusted as well | 07:44 |
cpaelzer | Hi, I'm unsure if something is down or if my memory is playing tricks on me | 08:40 |
cpaelzer | https://wiki.ubuntu.com/ProposedMigration points to https://git.launchpad.net/~ubuntu-release/britney/+git/hints-ubuntu | 08:40 |
cpaelzer | the latter seems down from Web and git fetch POV | 08:40 |
cpaelzer | I vaguely remember that the location of some repos got changed to be more accessible and wonder if that might be one of them | 08:41 |
cpaelzer | but then OTOH https://code.launchpad.net/~ubuntu-release/britney/+git/hints-ubuntu/+ref/devel is still up | 08:41 |
cpaelzer | so I wonder if this is a technical issue of LP-git or if it really moved and I (and the wiki page) just need a hint to the new place? | 08:42 |
gjolly_ | cpaelzer: LP-git is under maintenance apparently, even https://git.launchpad.net is down | 08:46 |
=== gjolly_ is now known as gjolly | ||
cpaelzer | thanks gjolly - that explains | 08:48 |
cjwatson | It was also announced on #launchpad and https://twitter.com/launchpadstatus | 08:51 |
nibbon_ | o/ | 09:41 |
nibbon_ | is there a special reason why golang-1.14 installs the binary in /usr/lib/go-1.14/bin instead of /usr/bin like the golang package does? | 09:42 |
cjwatson | nibbon_: the point of the golang-$version packages is so that you can have multiple versions installed at the same time, and that wouldn't work if they all tried to install to the same path | 09:54 |
cjwatson | nibbon_: and the point of the golang (and golang-go, etc.) metapackages is so that you don't have to worry about that, so if it bothers you, just install golang and be done with it? | 09:54 |
nibbon_ | I would like to build a .deb with golang 1.14 but the meta package fetches only golang 1.13 | 09:56 |
nibbon_ | so I forced golang-1.14 but the build fails because cannot find the golang binary | 09:56 |
cjwatson | golang-go basically just ships symlinks | 09:56 |
cjwatson | nibbon_: set PATH then? | 09:56 |
nibbon_ | I'm trying but I'm failing in doing so in debian/rules :/ | 09:57 |
cjwatson | "export PATH := /usr/lib/go-1.14/bin:$(PATH)" should do it | 09:57 |
cjwatson | (untested) | 09:58 |
* nibbon_ tries | 09:59 | |
nibbon_ | cjwatson: it works :) | 10:01 |
nibbon_ | cjwatson: my mistake was to export GOPATH | 10:01 |
nibbon_ | cjwatson: thx | 10:01 |
cjwatson | np | 10:05 |
=== Fallen[m] is now known as Fallen|pto[m] | ||
slyon | ddstreet: wrt lp:#1949089 Thanks for your comment on the MP! Apparently they had some issues with the testing and it turned out that SRU isn't actually the root-cause of this failure, so no need to revert it anyway. It's right AFAIU that the snap mount unit is being restarted during a daemon-reload. | 11:57 |
ubottu | Launchpad bug 1949089 in systemd (Ubuntu Bionic) "systemd randomly fails to activate mount units in Ubuntu Core 18" [Undecided, New] https://launchpad.net/bugs/1949089 | 11:57 |
seb128 | RikMills, hey, is there any chance you could try to handle some of the kde items on https://people.canonical.com/~ubuntu-archive/transitions/html/auto-poppler.html ? unsure if they need fixes or if no change rebuilds would be enough | 12:19 |
RikMills | seb128: sure, I will get on that in a short while | 12:21 |
seb128 | RikMills, thanks! | 12:23 |
seb128 | Is there anyone who could tell me how are ubuntu-manual-tests items associated with images on the ISO tracker? | 13:08 |
seb128 | http://iso.qa.ubuntu.com/qatracker/milestones/429/builds/239456/testcases for example (ubuntu desktop arm64) has no test case listed where it should probably have the same than the amd64 iso | 13:08 |
seb128 | similarly canary has no tests atm | 13:09 |
ddstreet | slyon ok thanks, i'll let you handle it then, feel free to engage sustaining eng if needed | 13:22 |
slyon | ddstreet: ACK! | 13:25 |
seb128 | bdmurray, ^ maybe you know about the iso tracker question? | 14:16 |
Orz | So, any development of ubuntu is discussed here before being published? | 14:33 |
seb128 | we discuss desktop work on daily basis here if that's what you are interested in | 14:37 |
seb128 | reading https://discourse.ubuntu.com/ is also useful if you want to follow what's going on | 14:37 |
Orz | awesome! | 14:39 |
=== genii-core is now known as genii | ||
tsimonq21 | <seb128> "http://iso.qa.ubuntu.com/..." <- It's been a while, but if you're logged in and have the right perms you should be able to assign testcases to that test suite + series. It's done manually iirc. | 15:06 |
seb128 | tsimonq21, either I don't have the right or I'm not finding the right screen to use | 15:14 |
Orz | one last question: How do you guys know when a topic is being discussed? I mean being online 24*7 is not possible and the chat gets reset everytime you join... | 15:20 |
tsimonq21 | <seb128> "Simon Quigley, either I don't..." <- Don't quote me on it, but this looks right: https://launchpad.net/~ubuntu-testcase | 15:24 |
tsimonq21 | seb128: Can you see this page? http://iso.qa.ubuntu.com/admin/config/services/qatracker/testsuites/266/edit | 15:25 |
tsimonq21 | <Orz> "one last question: How do you..." <- Being online 24/7 is possible, just use a bouncer or something that works in a similar way. | 15:28 |
Orz | bouncer? | 15:33 |
bdmurray | seb128: What tsimonq21 said is correct and you need to edit the relationship on the ISO tracker. | 15:33 |
seb128 | tsimonq21, no, I don't have access to that one | 15:34 |
seb128 | bdmurray, do you know if it's restricted to release team or if I could get added and by who? | 15:34 |
tsimonq21 | seb128: So from here iirc, you can either provide one of us with a list of testcases you like added, or file a bug here for an audit log: https://launchpad.net/ubuntu-manual-tests | 15:35 |
bdmurray | seb128: You should have access are your logged in? | 15:35 |
tsimonq21 | (But I do think you should have access.) | 15:35 |
seb128 | bdmurray, I'm logged in but http://iso.qa.ubuntu.com/admin/ gives me an oops page | 15:38 |
bdmurray | seb128: what about the url tsimonq gave earlier? | 15:38 |
seb128 | same, I started from there and went down by deleting parts | 15:39 |
bdmurray | seb128: hrmm, there are two seb128's in the access list seb128 and seb128_ | 15:43 |
seb128 | I'm logged in as seb128 | 15:44 |
bdmurray | I'm in a meeting at the moment but I'm trying and will try and figure out how to manage permissions here. | 15:49 |
Orz | tsimonq21 so you all host a bouncer server in order to remain online? Why didn't you consider discord? This is such a hassle to setup | 15:53 |
tsimonq21 | Orz: Nope, we all do it a little bit differently. Discord came after IRC and is proprietary with much less flexibility. You don't have to set up a bouncer, just use something like Matrix to bridge it. | 15:55 |
seb128 | bdmurray, thanks! | 16:01 |
usvsnsp | Hey! Thanks again for the advices. Finally got Element, a matrix bridging app to work for me. Can't wait to be a part of this awesome community! | 16:50 |
usvsnsp | (I was nicked as Orz before) | 16:51 |
bdmurray | seb128: I updated your role on the site - let's see if that helps. | 20:07 |
* mwhudson eyeballs libseccomp build failure | 20:34 | |
mwhudson | ah looks like a good lump of r stuff migrated | 21:36 |
mwhudson | hm | 22:20 |
mwhudson | does someone have a thing that can do "retry autopkgtests of package foo with an additional trigger"? | 22:20 |
mwhudson | mind you just all-proposed retries of r stuff seems to be doing the trick currently | 22:20 |
mwhudson | and i guess i can do the first thing i asked for with sed | 22:22 |
=== genii is now known as genii-core | ||
mwhudson | *something* in the R blob has regressed on s390x | 23:33 |
mwhudson | probably https://launchpad.net/ubuntu/+source/r-cran-s2 | 23:34 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!