| mwhudson | i presume it's more or less standard policy that tzdata changes end up getting SRUed? | 01:38 |
|---|---|---|
| Unit193 | Pretty normal. | 01:38 |
| === JanC is now known as Guest16348 | ||
| === JanC_ is now known as JanC | ||
| === infinity1 is now known as infinity | ||
| Laney | zul: Your neutron is still failing tests. | 10:38 |
| === _salem is now known as salem_ | ||
| === marcusto_ is now known as marcustomlinson | ||
| === Son_Goku is now known as Conan_Kudo | ||
| === Conan_Kudo is now known as Son_Goku | ||
| === boiko_ is now known as boiko | ||
| === jgrimm-away is now known as jgrimm | ||
| tsdgeos | was a pleasure working for Canonical! /me waves | 13:32 |
| === boiko_ is now known as boiko | ||
| bdmurray | barry: What did you say about the pyclean package install failure issue? | 17:55 |
| barry | bdmurray: it just looks like pyclean is being run by python3 which makes no sense given that py3clean should be run by python3 | 17:56 |
| bdmurray | barry: so presumably a local configuration issue? | 17:57 |
| barry | bdmurray: my guess is that this is another case of someone pointing /usr/bin/python to /usr/bin/python3 | 17:58 |
| barry | which no one should do | 17:58 |
| dobey | or if you're going to do it, you should at least remove all the python2 stuff, and pin those packages so they can't be installed again | 18:00 |
| bdmurray | barry: Maybe we should check for that in apport? | 18:01 |
| barry | bdmurray: that would actually be very helpful for debugging these cases because it seems like it might not be an uncommon thing for people to do, even though they have to manually toggle the symlink | 18:02 |
| bdmurray | barry: bug 1681528 if you are interested | 18:09 |
| ubottu | bug 1681528 in apport (Ubuntu) "Include information about python versions" [Undecided,New] https://launchpad.net/bugs/1681528 | 18:09 |
| barry | bdmurray: thanks | 18:10 |
| === bregma__ is now known as bregma | ||
| * xnox feels kind of bad for flying united | 20:30 | |
| * xnox hopes the fares will go down even further | 20:30 | |
| ogra_ | wel, according to some news sites you might need to use ships to leave your country after brexit :) | 20:30 |
| jackpot51 | https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/1681566 | 20:38 |
| ubottu | Launchpad bug 1681566 in update-manager (Ubuntu) "nvidia-375 DKMS module not recompiled on upgrade to 17.04" [Undecided,New] | 20:38 |
| jackpot51 | Updating from 16.10 to 17.04 is broken on NVIDIA systems | 20:38 |
| bdmurray | could you add your upgrade log files found in /var/log/dist-upgrade? | 20:39 |
| jackpot51 | Will do | 20:39 |
| jackpot51 | https://bugs.launchpad.net/ubuntu/+source/ubuntu-release-upgrader/+bug/1681566/+attachment/4859865/+files/dist-upgrade.tar.gz | 20:42 |
| ubottu | Launchpad bug 1681566 in ubuntu-release-upgrader (Ubuntu) "nvidia-375 DKMS module not recompiled on upgrade to 17.04" [Undecided,Confirmed] | 20:42 |
| jackpot51 | bdmurray: Maybe it should be moved to nvidia-graphics-drivers-375? | 20:52 |
| bdmurray | jackpot51: I found something else suspicious | 20:54 |
| jackpot51 | bdmurray: What is it? | 20:59 |
| === acheronuk is now known as acheronUK | ||
| bdmurray | jackpot51: Do you know what the system76-driver-nvidia package does? | 21:02 |
| dmj_s76 | Why yes! | 21:02 |
| dmj_s76 | bdmurray: It just depends on nvidia-375 and ubuntu-driver-common. It also installs some xorg quirks to handle a few specific gpu/display combinations. | 21:05 |
| bdmurray | So in apt-term.log we can see dkms building the driver for the 4.8 kernel - around line 4100 | 21:07 |
| jackpot51 | Right. It does not trigger for 4.10 | 21:07 |
| jackpot51 | On line 4928, it should have run | 21:08 |
| jackpot51 | And 5256 | 21:09 |
| bdmurray | 4928? Mine only goes to 10 er 4823 | 21:09 |
| dmj_s76 | Line 5450 shows it loading nvidia dkms modules for 4.8 | 21:10 |
| jackpot51 | Oh, sorry. There are \r's that are counted as newlines | 21:10 |
| dmj_s76 | Line 5450 -> 4141 | 21:13 |
| bdmurray | The suspicious thing I saw was in main.log re nvidiaUpdate() and "no old nvidia driver installed" but that probably isn't causing this | 21:14 |
| jackpot51 | I wonder where that comes from | 21:18 |
| bdmurray | ubuntu-release-upgrader's DistUpgradeCache.py | 21:19 |
| bdmurray | It looks like its for transitioning from different driver version numbers, but I'm guessing a bit. | 21:20 |
| bdmurray | Regardless ubuntu-drivers-obsolete.pkgs seems a bit old to me | 21:21 |
| jackpot51 | bdmurray: Any ideas on how to fix this? | 21:26 |
| bdmurray | jackpot51: Could you test w/o the system76 ppa? | 21:29 |
| jackpot51 | Sure thing bdmurray | 21:30 |
| bdmurray | jackpot51: thanks! | 21:32 |
| Zta | I'm trying to build a package from some random app. I want to properly fill in the Depends: section of my DEBIAN/control file. Is there a tool for this? I mean something that analyses (ldd) the binary and looks up the epoch version? | 21:43 |
| Zta | So that given a binary I get "Depends: liballegro4.4, libc6 (>= 2.14), libgcc1 (>= 1:3.0), libgif7 (>= 5.1), libjpeg8 (>= 8c)" | 21:43 |
| Zta | I've scripted it in bash for now, but it's not pretty | 21:46 |
| tarpman | Zta: sounds like you might be working one or two levels too low. the stuff that goes in the binary package (DEBIAN/ and such) is normally auto-generated by the dpkg-dev tools (dpkg-gencontrol and friends). the specific tool you're looking for is called dpkg-shlibdeps. at a higher level, most packages use debhelper to orchestrate the build. start here: https://www.debian.org/doc/manuals/maint-guide/ | 21:55 |
| Zta | Thanks! | 21:56 |
| Zta | I'll look into this tomorrow. | 21:57 |
| cardboard64 | my touchpad stops working after some time on 17.04 | 22:10 |
| cardboard64 | how can I troubleshoot that? | 22:11 |
| === salem_ is now known as _salem | ||
| === coreycb is now known as coreycb_vaca | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!