| nacc | xnox: i'll see if that's doable and if not just disable the test for now so it can propogate | 00:00 |
|---|---|---|
| xnox | yeah, and there is reboot command support in dep8 as well, if you want to like 1) [ -z "$ADT_REBOOT_MARK" ] && setup disks, and then later [ -n "$ADT_REBOOT_MARK" ] && check that daemon is autostarted and running. | 00:00 |
| xnox | with reboot between the two. | 00:00 |
| nacc | xnox: thanks, that's handy | 00:01 |
| xnox | nacc, we can override the test results for open-iscsi to propagate, but that's best to have a fixed test.... | 00:01 |
| nacc | xnox: ack, i'll try and get it fixed by EOW | 00:04 |
| xnox | mwhudson, there are now smart ExecStartPre= shell script in the systemd unit to bail out when not-needed / setup | 00:12 |
| xnox | slangasek, https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1708313 | 00:12 |
| ubot5 | Ubuntu bug 1708313 in systemd (Ubuntu) "PrivateNetwork=yes fails in lxc1 armhf on arm64 kernel" [Undecided,New] | 00:12 |
| xnox | not I ?! =) | 00:12 |
| sergiusens | slangasek: long shot, but can I get snapcraft 2.33 into xenial-updates and zesty-updates? the armhf test failure for snapcraft proper is just timeouts (ran twice already, squid.internal timeout in different tests) and the ubuntu-image ones are test errors in ubuntu-image proper (sil mentioned those are fixed in 1.1) | 00:24 |
| slangasek | sergiusens: xenial-updates is currently frozen for point release; snapcraft seems safe since it shouldn't impact images but I'll just highlight infinity to double-check | 00:43 |
| slangasek | checking ubuntu-image; I'm going to re-test with -proposed u-i + snapcraft so it's clear from the log | 00:44 |
| slangasek | xnox: why should PrivateNetwork fail there? it's just containers | 00:44 |
| sergiusens | thanks slangasek | 00:49 |
| xnox | slangasek, it is a good point. but also i don't know how our armhf infra is setup. | 00:59 |
| xnox | slangasek, is it priviledged or unpriviledged lxc1? what is the host? are there any config overrides that punch things through? | 00:59 |
| slangasek | hmm | 01:00 |
| xnox | at the moment i setup straight up lxc1 artufl armhf container on an artful arm64 host | 01:00 |
| slangasek | all good questions | 01:00 |
| xnox | but i suspect the host may not be xenial, and i'm debugging obsolete software not matching production. | 01:00 |
| slangasek | "may not be xenial" - I would expect it is. don't the logs report the kernel version? | 01:00 |
| xnox | the logs report armhf kernel.... | 01:01 |
| slangasek | the armhfness is a lie | 01:01 |
| xnox | testing everything in lxc1 on an old kernel/host with new binaries inside makes a lot of sense when one is developing that for the arm64 android phones. | 01:01 |
| slangasek | I don't know if it fakes the version? | 01:01 |
| * xnox wants armhf testing to move to lxd | 01:01 | |
| * xnox wants arm64 lxd testing added, given we do have arm64 hosts hooked up in adt | 01:02 | |
| * xnox wants arm64 kvm testing added - but this might be trickier to do | 01:02 | |
| xnox | slangasek, given we are not doing as much phone testing, i am suspecting our armhf and arm64 testing went downhill. | 01:03 |
| slangasek | I haven't seen evidence of this in general | 01:04 |
| slangasek | arm64 will be VMs, not containers, when the RTs finish | 01:04 |
| xnox | ok. | 01:04 |
| xnox | moving armhf from lxc to lxd? | 01:05 |
| xnox | that should be easy, given how reliable s390x lxd is. | 01:05 |
| xnox | no? | 01:05 |
| slangasek | that part I don't know | 01:06 |
| slangasek | xnox: on the autopkgtest side, the infra believes it is using lxd for armhf, not lxc. kernel is 4.4 on xenial. | 01:12 |
| xnox | slangasek, how old/new lxd? from release and/or backports? | 01:14 |
| slangasek | xnox: 2.0.9 | 01:15 |
| slangasek | which I guess means we're a little behind | 01:16 |
| xnox | slangasek, so trusty? | 01:16 |
| slangasek | xnox: no, xenial-updates-1 | 01:16 |
| xnox | ah, ok. | 01:17 |
| -queuebot:#ubuntu-release- New: accepted gcc-7-cross-ports [amd64] (artful-proposed) [3ubuntu2] | 01:23 | |
| -queuebot:#ubuntu-release- New: accepted node-libs-browser [amd64] (artful-proposed) [2.0.0-1] | 01:23 | |
| -queuebot:#ubuntu-release- New: accepted syslog-ng [arm64] (artful-proposed) [3.10.1-3ubuntu1] | 01:23 | |
| -queuebot:#ubuntu-release- New: accepted syslog-ng [i386] (artful-proposed) [3.10.1-3ubuntu1] | 01:23 | |
| -queuebot:#ubuntu-release- New: accepted syslog-ng [s390x] (artful-proposed) [3.10.1-3ubuntu1] | 01:23 | |
| -queuebot:#ubuntu-release- New: accepted gcc-7-cross-ports [i386] (artful-proposed) [3ubuntu2] | 01:23 | |
| -queuebot:#ubuntu-release- New: accepted syslog-ng [armhf] (artful-proposed) [3.10.1-3ubuntu1] | 01:23 | |
| -queuebot:#ubuntu-release- New: accepted syslog-ng [amd64] (artful-proposed) [3.10.1-3ubuntu1] | 01:23 | |
| -queuebot:#ubuntu-release- New: accepted syslog-ng [ppc64el] (artful-proposed) [3.10.1-3ubuntu1] | 01:23 | |
| xnox | slangasek, on artful host, launching artful armhf lxd container says armv8l in uname -a | 01:28 |
| xnox | slangasek, about stalled amd64/i386 systemd adt tests - can i get a juju log? or like nova console log of the instances running the test? | 01:36 |
| xnox | or ssh backdoor into them? i think pitti did open up a connection into those machines by hand before. | 01:36 |
| xnox | running adt tests here on xenial host, with qemu runner, using artful cloud-image as base - passes. | 01:36 |
| xnox | did that with amd64, todo with i386. | 01:36 |
| xnox | .. | 01:38 |
| xnox | i just got hashsum missmatch in ftpmaster.internal | 01:39 |
| xnox | http://paste.ubuntu.com/25230244/ | 01:39 |
| xnox | this is wow. | 01:39 |
| slangasek | xnox: sorry, it's dinner time here. I can look at this later in the evening | 01:41 |
| xnox | i should sleep too | 01:41 |
| * xnox goes to drink some tea to calm down | 01:41 | |
| infinity | xnox: Our buildds and autopkgtest hosts are booted with 'compat_uts_machine=armv7l' on the cmdline. | 02:35 |
| * apw had forgotten that thing exists | 02:37 | |
| infinity | apw: If we could come up with a sane way to boot armv7 images in scalingstack, we could ditch the silly hack. | 02:38 |
| infinity | Maybe I should find some "free time" (ha ha ha) for that. | 02:38 |
| * apw giggles at infinity | 02:38 | |
| infinity | jbicha: ubuntu-gnome/xenial seems to have 0 testing so far. | 02:41 |
| stgraber | xnox: hmm, uname -m not showing the right personality is weird, we don't see that behavior on x86 | 02:44 |
| stgraber | ah, just found apw's kernel patch. Feels pretty weird that linux32 on aarch64 otherwise gives you the exact same as linux64... | 02:51 |
| jbicha | infinity: sorry, I'll take a look in the morning | 02:51 |
| apw | stgraber, i thought it gave you something else, just the 32bit form of the hardware, not the form our 32bit images are optimised for | 02:52 |
| stgraber | oh, it does indeed | 02:53 |
| stgraber | you get aarch64 for 64bit and armv8l for 32bit | 02:53 |
| stgraber | root@1ss-arm64:~# uname -a | 02:53 |
| stgraber | Linux 1ss-arm64 4.4.0-75-generic #96-Ubuntu SMP Thu Apr 20 09:56:48 UTC 2017 aarch64 aarch64 aarch64 GNU/Linux | 02:53 |
| stgraber | root@1ss-arm64:~# setarch linux32 -- uname -a | 02:53 |
| stgraber | Linux 1ss-arm64 4.4.0-75-generic #96-Ubuntu SMP Thu Apr 20 09:56:48 UTC 2017 armv8l armv8l armv8l GNU/Linux | 02:53 |
| apw | right, just unfortuantly we are armv7 optimised | 02:54 |
| infinity | Has less to do with optimisation. | 02:54 |
| apw | s/optimised/targetted/ | 02:54 |
| infinity | And more to do with Stupid Software doing Stupid Things when building and seeing a uname it doesn't know. | 02:55 |
| stgraber | it can get even more confusing though :) | 02:55 |
| infinity | This is why i686 "stalled" on i686 too. | 02:55 |
| infinity | Despite us now being up to i786 (or i868 for Core, I've lost track) according to Intel. | 02:55 |
| stgraber | root@blah:~# uname -a | 02:55 |
| stgraber | Linux blah 4.11.0-10-generic #15-Ubuntu SMP Thu Jun 29 15:07:09 UTC 2017 aarch64 aarch64 aarch64 GNU/Linux | 02:55 |
| infinity | 886... | 02:55 |
| stgraber | root@blah:~# setarch linux32 -- uname -a | 02:55 |
| stgraber | Linux blah 4.11.0-10-generic #15-Ubuntu SMP Thu Jun 29 15:07:09 UTC 2017 aarch64 aarch64 aarch64 GNU/Linux | 02:55 |
| stgraber | that's on an arm64 system that doesn't support the 32bit personality | 02:55 |
| infinity | stgraber: Right, and that's correct. | 02:56 |
| infinity | Well, more correct would be setarch bubbling up a "that personality no exist" error, but whatever. | 02:56 |
| stgraber | yeah, I'd expect an error :) | 02:56 |
| stgraber | oh, the kernel does do the right thing, it's just setarch being useless :) | 02:57 |
| stgraber | personality(PER_LINUX32) = -1 EINVAL (Invalid argument) | 02:57 |
| infinity | Right. | 02:57 |
| infinity | I still think it was a mistake for ARM to not require v7 compat in v8, but oh well. | 02:58 |
| infinity | It's an even larger headache than v7 not requiring NEON. | 02:59 |
| infinity | Mistakes, though, why learn from them? | 02:59 |
| slangasek | sergiusens: ubuntu-image+snapcraft looking a bit better now in zesty, two passes, two xfails, and one fail (ppc64el) | 03:20 |
| slangasek | infinity: how frozen is xenial-updates, could snapcraft be released or no? | 03:21 |
| slangasek | (non-image-affecting) | 03:21 |
| infinity | slangasek: non-image-affecting is fine. | 03:21 |
| xnox | infinity, about that. to get "armhf" images booting in scaling stack, all we need is a regular arm64-uefi-firmware image with arm64 kernel and armhf userspace and compat_uts_machine=armv7l. slap together and done. we will still not be able to test armhf kernels in adt, but at least we will gain all the isolation-machine tests | 04:16 |
| * xnox assumes arm64 images are uefi images in scaling stack. if not mimic whatever arm64 images have for a bootloader. | 04:17 | |
| xnox | basically use multiarch technology and tools | 04:17 |
| wgrant | xnox, infinity: You can also construct an armhf image that boots an armhf kernel directly, without using UEFI, but it's a bit messy. | 04:19 |
| slangasek | yes, arm64 images are uefi | 04:20 |
| wgrant | Last time I tried it it required using images that masqueraded as AMIs and AKIs, so I got ovmf working instead, but direct kernel boot does work on bos01. | 04:20 |
| infinity | xnox: My point wasn't to boot armhf-on-arm64, I want armhf-on-armhf. But I'd rather drop armhf entirely, given a perfect world. | 04:25 |
| xnox | infinity, sure. armhf-on-arm64 kvm is better than lxd for adt testing. armhf-on-armhf does sound backwards. | 04:48 |
| xnox | to kill armhf, do we need an armv8l port then? or whatever the tag is? | 04:48 |
| xnox | also what is the point of i386 if joules edison and galileo got axed, and even that were 64-bit capable. | 04:58 |
| -queuebot:#ubuntu-release- New binary: behave [amd64] (artful-proposed/universe) [1.2.5-1ubuntu1] (no packageset) | 04:59 | |
| -queuebot:#ubuntu-release- Builds: Xubuntu Desktop amd64 [Xenial 16.04.3] has been marked as ready | 06:16 | |
| -queuebot:#ubuntu-release- Builds: Xubuntu Desktop i386 [Xenial 16.04.3] has been marked as ready | 06:16 | |
| -queuebot:#ubuntu-release- New binary: te923con [i386] (artful-proposed/universe) [0.6.1-1ubuntu1] (no packageset) | 06:23 | |
| -queuebot:#ubuntu-release- New binary: te923con [s390x] (artful-proposed/universe) [0.6.1-1ubuntu1] (no packageset) | 06:23 | |
| -queuebot:#ubuntu-release- New binary: te923con [ppc64el] (artful-proposed/universe) [0.6.1-1ubuntu1] (no packageset) | 06:23 | |
| -queuebot:#ubuntu-release- New binary: te923con [amd64] (artful-proposed/universe) [0.6.1-1ubuntu1] (no packageset) | 06:24 | |
| -queuebot:#ubuntu-release- New binary: te923con [arm64] (artful-proposed/universe) [0.6.1-1ubuntu1] (no packageset) | 06:29 | |
| -queuebot:#ubuntu-release- New binary: te923con [armhf] (artful-proposed/universe) [0.6.1-1ubuntu1] (no packageset) | 06:29 | |
| sil2100 | infinity: let me pick up netboot amd64 images now for some testing | 07:57 |
| infinity | sil2100: Not super concerned about netboot testing, it's not really coupled to the ISOs and point release (despite evidence to the contrary). | 07:59 |
| sil2100 | infinity: ok, I'll move to -GNOME then | 08:01 |
| -queuebot:#ubuntu-release- Builds: Ubuntu MATE Desktop amd64 [Xenial 16.04.3] has been marked as ready | 08:23 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu MATE Desktop i386 [Xenial 16.04.3] has been marked as ready | 08:23 | |
| sil2100 | Ubuntu GNOME looking fine on amd64, giving a quick spin of i386 now | 08:51 |
| === klebers_ is now known as klebers | ||
| LocutusOfBorg | hello, not sure if tsimonq2 already requested that, but please force badtest | 10:15 |
| LocutusOfBorg | fpc/3.0.2+dfsg-2 on amd64 (zesty) | 10:15 |
| LocutusOfBorg | libreoffice/1:5.3.1-0ubuntu2 i386 (zesty= | 10:16 |
| LocutusOfBorg | we tried them with themself as trigger | 10:16 |
| LocutusOfBorg | and the failed | 10:16 |
| tsimonq2 | LocutusOfBorg: I didn't say something yet, thank you :) | 10:18 |
| infinity | LocutusOfBorg: libreoffice failing might be a kernel issue. | 10:25 |
| infinity | apw: There was some talk that those fixes for the fixes for the fixes still didn't fix libreoffice, right? | 10:26 |
| tsimonq2 | infinity: But it's not a gtk+2.0 issue, correct? (that's what triggered the test) | 10:27 |
| infinity | tsimonq2: Sure, but force-badtest would also be a lie. | 10:27 |
| infinity | tsimonq2: You can say "this shouldn't hold up gtk" without saying "please ignore the test forever". | 10:27 |
| tsimonq2 | infinity: "this shouldn't hold up gtk" :P | 10:27 |
| LocutusOfBorg | infinity, yes, I agree, I'm following the Debian bug too | 10:27 |
| tsimonq2 | infinity: Regardless, the point in asking is to make sure that the SRU can migrate properly. | 10:28 |
| LocutusOfBorg | ok, well, so ignore it only once is the correct request? I agree this shouldn't be forced forever | 10:28 |
| LocutusOfBorg | I see libreoffice blacklisted on s390x not sure what does it mean | 10:29 |
| tsimonq2 | infinity (cc apw): This isn't just a Zesty issue from what I can tell, it seems to be a Xenial issue as well. | 10:29 |
| infinity | LocutusOfBorg: Nah, just informing an SRU team member that it's not a gtk regression should be fine. I mean, we'll mentally ignore it, but no need to commit anything to infra, since britney doesn't drive migrations in SRUs... yet. | 10:29 |
| infinity | tsimonq2: And yes, we know. | 10:29 |
| LocutusOfBorg | oh ok, so they are not blocking? really nice! | 10:29 |
| tsimonq2 | infinity: Alright, just covering my bases. :) | 10:30 |
| tsimonq2 | +1 LocutusOfBorg | 10:30 |
| LocutusOfBorg | I mean, not so nice but meh | 10:30 |
| tsimonq2 | Yeah, I don't know about nice but maybe convenient is the better word. :P | 10:30 |
| -queuebot:#ubuntu-release- Builds: Ubuntu Base amd64 [Xenial 16.04.3] has been marked as ready | 10:43 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Base arm64 [Xenial 16.04.3] has been marked as ready | 10:43 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Base armhf [Xenial 16.04.3] has been marked as ready | 10:43 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Base i386 [Xenial 16.04.3] has been marked as ready | 10:43 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Base powerpc [Xenial 16.04.3] has been marked as ready | 10:43 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Base ppc64el [Xenial 16.04.3] has been marked as ready | 10:43 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Base s390x [Xenial 16.04.3] has been marked as ready | 10:43 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Desktop amd64 [Xenial 16.04.3] has been marked as ready | 10:44 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Desktop i386 [Xenial 16.04.3] has been marked as ready | 10:44 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Server amd64 [Xenial 16.04.3] has been marked as ready | 10:45 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Server arm64 [Xenial 16.04.3] has been marked as ready | 10:45 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Server armhf+raspi2 [Xenial 16.04.3] has been marked as ready | 10:45 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Server i386 [Xenial 16.04.3] has been marked as ready | 10:45 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Server powerpc [Xenial 16.04.3] has been marked as ready | 10:45 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Server ppc64el [Xenial 16.04.3] has been marked as ready | 10:45 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Server s390x [Xenial 16.04.3] has been marked as ready | 10:45 | |
| sil2100 | infinity: GNOME i386 looks good as well, marked as so some minutes ago | 10:49 |
| infinity | sil2100: Shiny. | 10:49 |
| sil2100 | infinity: anything else needs testing? I don't see anything that's not-touched besides netboot | 10:49 |
| infinity | sil2100: I think we're getting down to just paperwork and me asking computers to do things that take a very long time. | 10:52 |
| -queuebot:#ubuntu-release- New binary: budgie-wallpapers [amd64] (artful-proposed/universe) [17.10] (no packageset) | 10:53 | |
| sil2100 | infinity: sweet, the changes for .3 look ok or should I work some more on those? | 11:03 |
| infinity | sil2100: I'll look in a bit. | 11:04 |
| infinity | sil2100: But if it looks kinda like a list of bugs, I'll probably not have complaints. | 11:04 |
| infinity | sil2100: Looks reasonable to me. | 11:06 |
| sil2100 | infinity: if there's anything I can help, just poke! We'll be going out for lunch in a moment but I'll be back soonish | 11:07 |
| -queuebot:#ubuntu-release- New binary: ganeti [i386] (artful-proposed/universe) [2.15.2-10] (no packageset) | 11:13 | |
| -queuebot:#ubuntu-release- New binary: ganeti [ppc64el] (artful-proposed/universe) [2.15.2-10] (no packageset) | 11:15 | |
| -queuebot:#ubuntu-release- New binary: ganeti [amd64] (artful-proposed/universe) [2.15.2-10] (no packageset) | 11:16 | |
| -queuebot:#ubuntu-release- New binary: ganeti [s390x] (artful-proposed/universe) [2.15.2-10] (no packageset) | 11:22 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Studio DVD amd64 [Xenial 16.04.3] has been marked as ready | 11:24 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu Studio DVD i386 [Xenial 16.04.3] has been marked as ready | 11:24 | |
| apw | sil2100, do you want these ubuntu-image things in the queue accpted over what is in -proposed ? | 11:30 |
| -queuebot:#ubuntu-release- New binary: ganeti [arm64] (artful-proposed/universe) [2.15.2-10] (no packageset) | 11:57 | |
| -queuebot:#ubuntu-release- New: accepted behave [amd64] (artful-proposed) [1.2.5-1ubuntu1] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted ganeti [amd64] (artful-proposed) [2.15.2-10] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted ganeti [i386] (artful-proposed) [2.15.2-10] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted ganeti [s390x] (artful-proposed) [2.15.2-10] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted te923con [arm64] (artful-proposed) [0.6.1-1ubuntu1] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted te923con [i386] (artful-proposed) [0.6.1-1ubuntu1] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted te923con [s390x] (artful-proposed) [0.6.1-1ubuntu1] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted budgie-wallpapers [amd64] (artful-proposed) [17.10] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted ganeti [ppc64el] (artful-proposed) [2.15.2-10] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted te923con [armhf] (artful-proposed) [0.6.1-1ubuntu1] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted ganeti [arm64] (artful-proposed) [2.15.2-10] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted te923con [ppc64el] (artful-proposed) [0.6.1-1ubuntu1] | 12:19 | |
| -queuebot:#ubuntu-release- New: accepted te923con [amd64] (artful-proposed) [0.6.1-1ubuntu1] | 12:19 | |
| doko | looking at gcc-7-cross-ports in update_excuses.html: why is the ppc64el build listed as missing | 13:17 |
| doko | and the cross compilers seem to be installable as well | 13:17 |
| infinity | doko: The uninstallable stuff might be component mismatches. Not sure about the ppc64el thing. | 13:32 |
| doko | ahh, yes, demoting | 13:33 |
| doko | gcc-7-cross-ports is kept in main because of the need of powerpc on ppc64el | 13:33 |
| infinity | doko: Erm, the ppc64el thing is because you didn't bump versions. | 13:34 |
| infinity | Why the second one didn't fail to upload is a mystery I'd rather not think about. | 13:35 |
| doko | hmm, I removed the binaries before uploading the second one ... | 13:35 |
| infinity | That would be why it didn't fail to upload. | 13:35 |
| infinity | Also, ick. | 13:35 |
| doko | anyway, doing the 7.2 release candidate today | 13:35 |
| infinity | doko: Binaries (and sources) should never, ever, ever re-use versions in the archive. Things go nutty. | 13:35 |
| infinity | Totally an LP bug that it even let that happen. | 13:36 |
| infinity | cjwatson: ^ | 13:37 |
| cjwatson | I'm pretty surprised that was allowed, certainly. Please file. | 13:40 |
| cjwatson | And don't pull that sort of binary-removal stunt again. | 13:40 |
| infinity | doko: Yes. Pretty please don't remove a file because it conflicts with another you want with different contents. :P | 13:40 |
| infinity | doko: Files in the history of an archive must be unique. | 13:41 |
| cjwatson | I guess it's possible we only explicitly check that for sources. | 13:42 |
| doko | infinity: ack | 13:43 |
| infinity | cjwatson: Things fail to upload all the time for that reason. Or does it maybe only check published binaries, rather than all of history? | 13:45 |
| cjwatson | infinity: Do you happen to remember an example? | 13:45 |
| infinity | I suspect doko's own package is an example (though, not sure how far back I'd have to go to find it). | 13:46 |
| infinity | But yes, we get builds failing to upload due to binary version conflicts (and also version downgrades). | 13:46 |
| cjwatson | Version conflicts are different. | 13:46 |
| infinity | Different than... A version conflict? | 13:47 |
| cjwatson | I'm not saying there are no checks on binary uploads, just that I suspect the same-version-different-contents check is absent. | 13:47 |
| infinity | We may be saying the same thing, sort of? | 13:47 |
| infinity | I mean, a build with a version already published will reject. Unless it's a copy (thus same contents). But a build with a version that's historically been used but not currently published will, apparently, accept. | 13:48 |
| infinity | Or, that seems to describe what I've seen. | 13:48 |
| cjwatson | Does it actually reject, or just crash during publication? | 13:49 |
| cjwatson | I see plenty of "Version older than that in the archive" rejections in recent history, but nothing that's obviously a version-conflict rejection. | 13:49 |
| infinity | I think doko's gcc-cross packages have rejected in the past for this same reason. | 13:50 |
| infinity | Cause they really should encode more version info than they do. | 13:50 |
| infinity | It's entirely possible this one rejected and he did the removal and then retried the build. :P | 13:50 |
| cjwatson | Ah, https://launchpad.net/ubuntu/+source/gcc-6-cross-ports/20ubuntu4/+build/12555755 is an example. | 13:51 |
| infinity | Yup. | 13:52 |
| cjwatson | Missed it because it's done in lp.soyuz.model.queue:PackageUpload rather than in archiveuploader. | 13:53 |
| cjwatson | So right, that's basically a last-ditch check against stuff that hasn't been removed. | 13:53 |
| cjwatson | Might be interesting to see what would break if we just removed the "AND bpph.dateremoved IS NULL" condition from that query. | 13:54 |
| cjwatson | Possibly a performance disaster, not sure. | 13:54 |
| cjwatson | So yeah, removal successfully hammered the upload through to the point where it failed later. And it might have failed even more disastrously depending on how it interacted with https://bugs.launchpad.net/launchpad/+bug/238826 .. | 13:56 |
| ubot5 | Ubuntu bug 238826 in Launchpad itself "Death-row misbehave on binary collision" [Low,Triaged] | 13:56 |
| infinity | cjwatson: It's almost certainly failed pretty hard already, because a-f cache. | 13:57 |
| infinity | (as in, the sums in Packages are probably for the old files) | 13:57 |
| cjwatson | Quite possibly. | 13:57 |
| infinity | Because a-f makes the entirely reasonable assumption that archive members are unique. :P | 13:57 |
| cjwatson | doko: Is this a thing you've done in the past? Because if so we may need to audit the archive. | 13:58 |
| doko | cjwatson: no, afaicr. in the past we had failing builds on one or two archs, whith other archs succeeding, and then the version skew for the binaries | 13:59 |
| doko | for the next upload | 13:59 |
| doko | the skew goes away with the next gcc-7 upload, and all cross builds succeeding | 14:00 |
| cjwatson | I don't understand why that would result in version skew. Wouldn't you bump all the versions in sync? | 14:00 |
| doko | no, it's looking at the binaries found in the archive | 14:01 |
| doko | if I hardcode that, then I usually forget to bump the number for a new upload :/ | 14:01 |
| infinity | doko: It really needs to tack on the version of the gcc-cross package as well, so it remains unique per build. | 14:01 |
| cjwatson | Ah, so retrying -cross-ports with a different base toolchain or something | 14:01 |
| infinity | doko: As in, the "cross1" at the end could be "cross20ubuntu5" or something for gcc-6-cross-ports_20ubuntu5 | 14:02 |
| doko | infinity: yeah, but then you had a version number like -12ubuntu1cross5ubuntu2 ... | 14:02 |
| doko | ugly | 14:02 |
| infinity | doko: Shorten "Ubuntu" to "u" and it's not the worst. | 14:02 |
| infinity | I mean, they're ugly regardless. :P | 14:02 |
| infinity | And shorten "cross" to "c" or "x" | 14:03 |
| infinity | -12ubuntu1x5u2 | 14:03 |
| apw | and it looks like a disk name in solaris | 14:03 |
| infinity | In Debian, it's be a nice short -12x5 | 14:03 |
| infinity | Hahaha. | 14:03 |
| sil2100 | apw: yes :) | 14:11 |
| xnox | what is solaris? | 14:14 |
| -queuebot:#ubuntu-release- Builds: Ubuntu GNOME Desktop amd64 [Xenial 16.04.3] has been marked as ready | 14:14 | |
| -queuebot:#ubuntu-release- Builds: Ubuntu GNOME Desktop i386 [Xenial 16.04.3] has been marked as ready | 14:14 | |
| doko | the kiosk seed is pulling in qtubuntu. is this expected? | 14:21 |
| doko | http://people.canonical.com/~ubuntu-archive/component-mismatches-proposed.svg | 14:21 |
| jbicha | doko: looks like yes: https://code.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/ubuntu.artful | 14:23 |
| doko | jbicha: but all the -cpp packages are unmaintained, aren't they? | 14:24 |
| infinity | I was under the impression pretty much that whole dep tree is unmaintained. | 14:24 |
| doko | xnox: ^^^ | 14:24 |
| xnox | doko, yes kiosk seed pulling in qtubuntu is expected. | 14:25 |
| xnox | and alan_g is on the hook to minimise deps. | 14:25 |
| xnox | (alan_g and co) | 14:25 |
| xnox | they have requested that in a bug report, mentioned in the seed text. | 14:26 |
| alan_g | Saviq: FYI ^^ | 14:27 |
| xnox | the graph is nice: http://people.canonical.com/~ubuntu-archive/component-mismatches-proposed.svg this may be a useful guide on as to what to drop. | 14:28 |
| Saviq | we're cutting off at platform-api | 14:28 |
| xnox | whoop whoop =) | 14:28 |
| Saviq | as in, we won't depend on that, soon | 14:28 |
| Saviq | so that graph will look much nicer | 14:28 |
| xnox | doko, infinity ^ | 14:28 |
| Saviq | the other fruit is content-hub, which we need to cut off in qtubuntu, too, and are already on it | 14:29 |
| doko | looks like the debian perl team is busy updating everything ... | 14:48 |
| xnox | doko, helpful. stop the importer?! =) | 14:53 |
| -queuebot:#ubuntu-release- Builds: Kubuntu Desktop amd64 [Xenial 16.04.3] has been marked as ready | 15:09 | |
| -queuebot:#ubuntu-release- Builds: Mythbuntu Desktop amd64 [Xenial 16.04.3] has been marked as ready | 15:09 | |
| -queuebot:#ubuntu-release- Builds: Mythbuntu Desktop i386 [Xenial 16.04.3] has been marked as ready | 15:09 | |
| -queuebot:#ubuntu-release- Builds: Kubuntu Desktop i386 [Xenial 16.04.3] has been marked as ready | 15:09 | |
| -queuebot:#ubuntu-release- Builds: 39 entries have been added, updated or disabled | 15:16 | |
| -queuebot:#ubuntu-release- New binary: fwupd [ppc64el] (artful-proposed/main) [0.9.6-0ubuntu1] (desktop-core) | 15:35 | |
| -queuebot:#ubuntu-release- New binary: fwupd [amd64] (artful-proposed/main) [0.9.6-0ubuntu1] (desktop-core) | 15:37 | |
| -queuebot:#ubuntu-release- New binary: fwupd [s390x] (artful-proposed/main) [0.9.6-0ubuntu1] (desktop-core) | 15:38 | |
| -queuebot:#ubuntu-release- New binary: fwupd [i386] (artful-proposed/main) [0.9.6-0ubuntu1] (desktop-core) | 15:38 | |
| -queuebot:#ubuntu-release- New binary: fwupd [arm64] (artful-proposed/main) [0.9.6-0ubuntu1] (desktop-core) | 15:41 | |
| -queuebot:#ubuntu-release- New binary: fwupd [armhf] (artful-proposed/main) [0.9.6-0ubuntu1] (desktop-core) | 15:43 | |
| -queuebot:#ubuntu-release- New binary: linux [s390x] (artful-proposed/main) [4.11.0-13.19] (core, kernel) | 15:50 | |
| -queuebot:#ubuntu-release- New binary: linux-signed [amd64] (vivid-proposed/main) [3.19.0-90.98] (core, kernel) | 15:57 | |
| -queuebot:#ubuntu-release- New binary: linux-signed [amd64] (zesty-proposed/main) [4.10.0-30.34] (core, kernel) | 15:57 | |
| -queuebot:#ubuntu-release- New: accepted linux-signed [amd64] (zesty-proposed) [4.10.0-30.34] | 15:57 | |
| -queuebot:#ubuntu-release- New: accepted linux-signed [amd64] (vivid-proposed) [3.19.0-90.98] | 15:58 | |
| -queuebot:#ubuntu-release- New binary: linux-signed-hwe-edge [amd64] (xenial-proposed/main) [4.11.0-13.19~16.04.1] (kernel) | 16:00 | |
| -queuebot:#ubuntu-release- New binary: linux-signed [amd64] (xenial-proposed/main) [4.4.0-89.112] (core, kernel) | 16:00 | |
| -queuebot:#ubuntu-release- New binary: linux-signed-lts-xenial [amd64] (trusty-proposed/main) [4.4.0-89.112~14.04.1] (kernel) | 16:01 | |
| -queuebot:#ubuntu-release- New binary: linux-signed-hwe [amd64] (xenial-proposed/main) [4.10.0-30.34~16.04.1] (kernel) | 16:01 | |
| -queuebot:#ubuntu-release- New: accepted linux-signed-hwe-edge [amd64] (xenial-proposed) [4.11.0-13.19~16.04.1] | 16:02 | |
| -queuebot:#ubuntu-release- New: accepted linux-signed [amd64] (xenial-proposed) [4.4.0-89.112] | 16:02 | |
| -queuebot:#ubuntu-release- New: accepted linux-signed-hwe [amd64] (xenial-proposed) [4.10.0-30.34~16.04.1] | 16:02 | |
| -queuebot:#ubuntu-release- New: accepted linux-signed-lts-xenial [amd64] (trusty-proposed) [4.4.0-89.112~14.04.1] | 16:02 | |
| === infinity changed the topic of #ubuntu-release to: Released: Xenial 16.04.3, Zesty 17.04 | Archive: open | Artful Release Coordination | Please don't upload things during freezes where you shouldn't, or be prepared to apologise to the release team | We accept payment in cash, check or beer | melior malum quod cognoscis | ||
| -queuebot:#ubuntu-release- New binary: linux [ppc64el] (artful-proposed/main) [4.11.0-13.19] (core, kernel) | 16:17 | |
| -queuebot:#ubuntu-release- New source: mozjs52 (artful-proposed/primary) [52.2.1-1~git1] | 17:27 | |
| teward | is there a reason https://bugs.launchpad.net/ubuntu/+source/postfix/+bug/1595096 has just sat on the sponsoring list for eternity and never been looked at? | 19:11 |
| ubot5 | Ubuntu bug 1595096 in postfix (Ubuntu Xenial) "cannot create multi postfix instance by postmulti command" [Medium,In progress] | 19:11 |
| teward | in queue as of 3/25/2017 | 19:11 |
| teward | jgrimm: ^ cc | 19:11 |
| nacc | teward: not especially | 19:14 |
| nacc | teward: i think the sponsorship queue is a bit behind for server | 19:14 |
| nacc | i believe i have a pilot session next week | 19:14 |
| teward | nacc: "behind" I think is an understatement - this is 5 months since last activity... just saying. | 19:14 |
| teward | (and this is breaking my experimenting with postfix heh) | 19:15 |
| teward | i'll do a power user hack for now and compile an updated package in a PPA, but... | 19:15 |
| -queuebot:#ubuntu-release- New binary: linux [i386] (artful-proposed/main) [4.11.0-13.19] (core, kernel) | 19:15 | |
| teward | (at least NGINX updates get handled real quickly... oh wait, that's my uploads... nevermind :P) | 19:15 |
| nacc | teward: feel free to bump in the bug and we should see it in the triage rota | 19:15 |
| teward | bumped :p | 19:16 |
| nacc | teward: thanks | 19:20 |
| nacc | teward: i'm guessing hte debdiff probably needs updating too, but that's ok | 19:20 |
| -queuebot:#ubuntu-release- New binary: linux [arm64] (artful-proposed/main) [4.11.0-13.19] (core, kernel) | 19:33 | |
| teward | nacc: indeed, but I'm happy to do that if needed. | 19:35 |
| teward | because i want this working :P | 19:35 |
| nacc | teward: :) | 19:35 |
| jgrimm | teward, nacc. as far i know its ready to go, had just been waiting for sponsors to get around to it | 19:46 |
| nacc | jgrimm: +1 agreed | 19:46 |
| teward | which is "never" if you don't prod them every so often :P | 19:46 |
| jgrimm | template including how to test for sponsor so should be easy enough nacc. let me know if you have questions | 19:47 |
| teward | 5 months is infinite lagtime. | 19:47 |
| jgrimm | teward, :) | 19:47 |
| teward | (in the IT world anyways) | 19:47 |
| nacc | jgrimm: yep, i saw that | 19:47 |
| teward | jgrimm: I had a habit before I got PPU rights - I always bothered sponsors every week until they uploaded my things, esp. if it's critical. Now for nginx, I just upload and wait for the SRU teams :P | 19:47 |
| -queuebot:#ubuntu-release- New binary: linux [amd64] (artful-proposed/main) [4.11.0-13.19] (core, kernel) | 20:55 | |
| -queuebot:#ubuntu-release- Unapproved: nvme-cli (xenial-proposed/universe) [0.5-1ubuntu0.1 => 0.5-1ubuntu0.2] (no packageset) | 21:42 | |
| -queuebot:#ubuntu-release- New binary: linux [armhf] (artful-proposed/main) [4.11.0-13.19] (core, kernel) | 22:12 | |
| === nacc_ is now known as nacc | ||
| -queuebot:#ubuntu-release- New binary: git-lfs [ppc64el] (artful-proposed/none) [2.2.1-1] (no packageset) | 23:10 | |
| -queuebot:#ubuntu-release- New binary: golang-github-approvals-go-approval-tests [amd64] (artful-proposed/none) [0.0~git20170712.0.c1e747e-1] (no packageset) | 23:10 | |
| -queuebot:#ubuntu-release- New binary: golang-github-chzyer-readline [amd64] (artful-proposed/none) [1.4-1] (no packageset) | 23:10 | |
| -queuebot:#ubuntu-release- New binary: lua-mode [amd64] (artful-proposed/universe) [20151025-2] (no packageset) | 23:10 | |
| -queuebot:#ubuntu-release- New binary: golang-github-aliyun-aliyun-oss-go-sdk [amd64] (artful-proposed/none) [1.5.0+dfsg1-1] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: golang-github-hashicorp-go-rootcerts [amd64] (artful-proposed/none) [0.0~git20160503.0.6bb64b3-1] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: golang-github-cheekybits-is [amd64] (artful-proposed/none) [0.0~git20150225.0.68e9c06-1] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: git-lfs [amd64] (artful-proposed/none) [2.2.1-1] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: git-lfs [s390x] (artful-proposed/none) [2.2.1-1] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: inflection [amd64] (artful-proposed/none) [0.3.1-1] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: org-mode [amd64] (artful-proposed/universe) [9.0.9+dfsg-3] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: git-lfs [i386] (artful-proposed/none) [2.2.1-1] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: more-itertools [amd64] (artful-proposed/none) [3.2.0-2] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: golang-github-glendc-gopher-json [amd64] (artful-proposed/none) [0.1.0+git20170414.0.dc47430-1] (no packageset) | 23:11 | |
| -queuebot:#ubuntu-release- New binary: backdoor-factory [amd64] (artful-proposed/none) [3.4.2+dfsg-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: golang-github-gophercloud-gophercloud [amd64] (artful-proposed/none) [0.0~git20170728.0.0118feec-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: golang-github-tdewolff-test [amd64] (artful-proposed/none) [0.0~git20170115.0.a7cf99a-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: ruby-declarative-option [amd64] (artful-proposed/none) [0.1.0-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: golang-github-denverdino-aliyungo [amd64] (artful-proposed/none) [0.0~git20170721.0.80ceb80-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: golang-gopkg-go-playground-validator.v8 [amd64] (artful-proposed/none) [8.18.1-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: golang-github-opentracing-opentracing-go [amd64] (artful-proposed/none) [1.0.2-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: ruby-rash-alt [amd64] (artful-proposed/none) [0.4.3-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: astral [amd64] (artful-proposed/none) [1.4-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: ruby-uber [amd64] (artful-proposed/none) [0.1.0-1] (no packageset) | 23:12 | |
| -queuebot:#ubuntu-release- New binary: ruby-errbase [amd64] (artful-proposed/none) [0.0.3-1] (no packageset) | 23:13 | |
| -queuebot:#ubuntu-release- New binary: ruby-test-xml [amd64] (artful-proposed/none) [0.1.8-1] (no packageset) | 23:13 | |
| -queuebot:#ubuntu-release- New binary: ruby-declarative [amd64] (artful-proposed/universe) [0.0.9-1] (no packageset) | 23:14 | |
| -queuebot:#ubuntu-release- New: accepted astral [amd64] (artful-proposed) [1.4-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-denverdino-aliyungo [amd64] (artful-proposed) [0.0~git20170721.0.80ceb80-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-tdewolff-test [amd64] (artful-proposed) [0.0~git20170115.0.a7cf99a-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted ruby-declarative [amd64] (artful-proposed) [0.0.9-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted ruby-rash-alt [amd64] (artful-proposed) [0.4.3-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted ruby-uber [amd64] (artful-proposed) [0.1.0-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted backdoor-factory [amd64] (artful-proposed) [3.4.2+dfsg-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted golang-gopkg-go-playground-validator.v8 [amd64] (artful-proposed) [8.18.1-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted ruby-test-xml [amd64] (artful-proposed) [0.1.8-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-opentracing-opentracing-go [amd64] (artful-proposed) [1.0.2-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted ruby-errbase [amd64] (artful-proposed) [0.0.3-1] | 23:45 | |
| -queuebot:#ubuntu-release- New: accepted git-lfs [amd64] (artful-proposed) [2.2.1-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted git-lfs [ppc64el] (artful-proposed) [2.2.1-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-aliyun-aliyun-oss-go-sdk [amd64] (artful-proposed) [1.5.0+dfsg1-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-cheekybits-is [amd64] (artful-proposed) [0.0~git20150225.0.68e9c06-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-glendc-gopher-json [amd64] (artful-proposed) [0.1.0+git20170414.0.dc47430-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-hashicorp-go-rootcerts [amd64] (artful-proposed) [0.0~git20160503.0.6bb64b3-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted lua-mode [amd64] (artful-proposed) [20151025-2] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted org-mode [amd64] (artful-proposed) [9.0.9+dfsg-3] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted git-lfs [i386] (artful-proposed) [2.2.1-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-approvals-go-approval-tests [amd64] (artful-proposed) [0.0~git20170712.0.c1e747e-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-gophercloud-gophercloud [amd64] (artful-proposed) [0.0~git20170728.0.0118feec-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted more-itertools [amd64] (artful-proposed) [3.2.0-2] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted git-lfs [s390x] (artful-proposed) [2.2.1-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted inflection [amd64] (artful-proposed) [0.3.1-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted golang-github-chzyer-readline [amd64] (artful-proposed) [1.4-1] | 23:46 | |
| -queuebot:#ubuntu-release- New: accepted ruby-declarative-option [amd64] (artful-proposed) [0.1.0-1] | 23:46 | |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
