| -queuebot:#ubuntu-release- New: accepted llvm-toolchain-12 [amd64] (impish-proposed) [1:12.0.0-3ubuntu1] | 06:54 | |
| -queuebot:#ubuntu-release- Unapproved: accepted nvidia-graphics-drivers-460 [source] (groovy-proposed) [460.80-0ubuntu0.20.10.1] | 08:01 | |
| -queuebot:#ubuntu-release- Unapproved: accepted nvidia-graphics-drivers-460 [source] (focal-proposed) [460.80-0ubuntu0.20.04.1] | 08:02 | |
| Laney | juliank: fyi, I have cowboyed worker-bos01-arm64.conf to not have bionic | 08:52 |
|---|---|---|
| Laney | if this carries on then we'll need to fix the charm to support this better | 08:52 |
| Laney | (have prioritised the ticket too though) | 08:53 |
| Laney | but (some) charm config updates will cause that to be lost | 08:53 |
| juliank | ok | 08:54 |
| juliank | I'm out starting tomorrow, hopefully we can get this resolved by the time I'm back Monday,or well before I break it :D | 08:55 |
| -queuebot:#ubuntu-release- Unapproved: accepted nvidia-graphics-drivers-460 [source] (bionic-proposed) [460.80-0ubuntu0.18.04.1] | 09:01 | |
| -queuebot:#ubuntu-release- Unapproved: accepted nvidia-graphics-drivers-465 [source] (groovy-proposed) [465.27-0ubuntu0.20.10.2] | 09:08 | |
| -queuebot:#ubuntu-release- Unapproved: accepted nvidia-graphics-drivers-465 [source] (hirsute-proposed) [465.27-0ubuntu0.21.04.2] | 09:09 | |
| -queuebot:#ubuntu-release- Unapproved: accepted nvidia-graphics-drivers-465 [source] (focal-proposed) [465.27-0ubuntu0.20.04.2] | 09:12 | |
| juliank | waveform: So https://autopkgtest.ubuntu.com/packages/apt is slow; I can improve the query time from 0.8s to 0.008s on my laptop by creating an indedx on test(package) - which is char[120] | 09:12 |
| juliank | waveform: The query is SELECT MAX(run_id), exitcode, release, arch FROM test, result WHERE package = 'apt' AND test.id = result.test_id GROUP BY release, arch ; | 09:12 |
| juliank | And other page queries on package, release, arch all together | 09:12 |
| juliank | so I'm wondering if I should create index on all three fields (package, release, arch) | 09:12 |
| juliank | arch names are like 5 bytes and release names like 10, so performance does not matter much | 09:13 |
| juliank | index on all three fields improves runtime by another 50% | 09:13 |
| juliank | Index on all three certainly looks nice https://paste.ubuntu.com/p/qppVsJbWNP/ | 09:15 |
| juliank | Everything querying package also seems to query release and arch or group by them | 09:20 |
| juliank | I think let's do all three, even after vacuuming i don't see a size increase, and the queries look a lot nicer | 09:22 |
| juliank | Laney, waveform https://code.launchpad.net/~juliank/autopkgtest-cloud/+git/autopkgtest-cloud/+merge/402630 | 09:37 |
| juliank | Don't know performance impact IRL | 09:40 |
| juliank | The query only takes 0.8s here, but on the cloud the page takes 16s to render | 09:40 |
| juliank | but cloud is much slower than my i5-8250u | 09:40 |
| juliank | like 6 times slower or so at least | 09:41 |
| juliank | The rest is just template rendering, and I'd hope that's not super slow | 09:42 |
| juliank | so I expect huge gains to be made there | 09:42 |
| -queuebot:#ubuntu-release- Unapproved: nvidia-graphics-drivers-465 (bionic-proposed/restricted) [465.27-0ubuntu0.18.04.1 => 465.27-0ubuntu0.18.04.2] (no packageset) | 09:45 | |
| -queuebot:#ubuntu-release- Unapproved: rejected nvidia-graphics-drivers-465 [source] (bionic-proposed) [465.27-0ubuntu0.18.04.2] | 09:47 | |
| -queuebot:#ubuntu-release- Unapproved: accepted nvidia-graphics-drivers-465 [source] (bionic-proposed) [465.27-0ubuntu0.18.04.2] | 09:49 | |
| Laney | juliank: nice, now do /statistics :D | 10:01 |
| Laney | (but I'd like to wait for w_aveform's review) | 10:01 |
| juliank | Laney: wave form is working on that.one | 10:02 |
| Laney | cool | 10:03 |
| Laney | private chats? | 10:03 |
| * Laney :( | 10:03 | |
| waveform | juliank, on the index for test -- yes in my current design I've actually got a unique key covering package, release, arch (as it appears that table is basically an optimization converting that "real" key into a surrogate integer for the purposes of making result smaller) | 10:23 |
| juliank | Laney: Last Friday, at ~0900 UTC | 10:23 |
| juliank | waveform: sounds better :D | 10:24 |
| waveform | incidentally, I'm still working on /statistics -- I can make it *much* faster (and tidy up several issues with the schema, including that several package+versions are duplicated in the triggers columns which also affects the reports) -- unfortunately my current design also turns the database from a slightly chubby 1GB into a truly lumbering 4GB which is ... not ideal | 10:24 |
| doko | rbalint: how did you make the opensaml test pass? ;p | 10:33 |
| juliank | waveform: Well, I for one would be happy with just faster /packages/<name> for now :D | 10:34 |
| juliank | waveform: so the unique key/index; because that's a common workflow :D | 10:35 |
| waveform | yeah -- I'll review your MP in a mo | 10:35 |
| juliank | waveform: I can add UNIQUE to the CREATE INDEX, too :D | 10:35 |
| waveform | juliank, actually -- I'd just use ALTER TABLE test ADD CONSTRAINT test_package_uk UNIQUE KEY (package, release, arch) because firstly a unique key always implies the create of a unique index (in practice anyway) and secondly that permits us a foreign key to that unqiue key should we happen to need it in future (I don't seem to recall there's any further differences between unique keys and indexes in SQLite -- there are in some other engines | 10:36 |
| waveform | though) | 10:36 |
| juliank | waveform: sqlite does not support ADD CONSTRAINT, though, so ... | 10:38 |
| juliank | So create unique index it'd have to be, or create new table and copy over madness :D | 10:39 |
| waveform | argh -- that's annoying (almost as annoying as the continuing lack of RIGHT+FULL JOIN) | 10:40 |
| juliank | But adding the unique constraint on the index seems sensible | 10:40 |
| waveform | yes - that'll do for now | 10:40 |
| waveform | juliank, MP reviewed | 10:42 |
| juliank | waveform: thanks | 10:46 |
| juliank | Laney, waveform Published, check this speed out: https://autopkgtest.ubuntu.com/packages/apt | 10:49 |
| juliank | 0.3s | 10:50 |
| juliank | down from 14s or so | 10:50 |
| Laney | :D | 10:50 |
| Laney | ...databases, eh... | 11:01 |
| juliank | crazy world | 11:07 |
| doko | bryce, cpaelzer: please subscribe server to the php8.0 package. I've promoted it now, to resolve the components mismatch after the php-defaults upload | 11:34 |
| cpaelzer | doko: bryce: done | 12:07 |
| -queuebot:#ubuntu-release- Unapproved: accepted nfs-utils [source] (bionic-proposed) [1:1.3.4-2.1ubuntu5.4] | 12:47 | |
| rbalint | doko, i've added it to the tests requiring a big vm | 13:59 |
| rbalint | doko, at least filed an mp for that, i can't add packages on my own for the record | 14:00 |
| rbalint | sil2100, could you please process LP: #1928202 to let glibc migrate? | 14:07 |
| ubot3 | Launchpad bug 1928202 in r-cran-treespace (Ubuntu) "Please remove1.1.4.0+dfsg-1 from impish" [Undecided, New] https://launchpad.net/bugs/1928202 | 14:07 |
| sil2100 | o/ | 14:07 |
| sil2100 | rbalint: I'm always a bit weary about package removals - does this mean the package as-is is completely broken and unusable? | 14:09 |
| rbalint | sil2100, yes | 14:10 |
| rbalint | there may be working bits in it, but unit tests failing does not promise a log | 14:11 |
| rbalint | sil2100, it has a low popcon as well | 14:11 |
| sil2100 | Okay o/ | 14:15 |
| sil2100 | Should be done | 14:18 |
| -queuebot:#ubuntu-release- Unapproved: accepted gtk+3.0 [source] (hirsute-proposed) [3.24.25-1ubuntu4.1] | 15:06 | |
| -queuebot:#ubuntu-release- Unapproved: accepted livecd-rootfs [source] (bionic-proposed) [2.525.53] | 15:26 | |
| bryce | cpaelzer, doko thanks | 15:49 |
| cpaelzer | yw | 16:03 |
| -queuebot:#ubuntu-release- Unapproved: rejected ubuntu-advantage-tools [source] (groovy-proposed) [27.0.2~20.10.1] | 17:18 | |
| -queuebot:#ubuntu-release- Unapproved: rejected ubuntu-advantage-tools [source] (hirsute-proposed) [27.0.2~21.04.1] | 17:18 | |
| -queuebot:#ubuntu-release- Unapproved: rejected ubuntu-advantage-tools [source] (bionic-proposed) [27.0.2~18.04.1] | 17:19 | |
| -queuebot:#ubuntu-release- Unapproved: rejected ubuntu-advantage-tools [source] (focal-proposed) [27.0.2~20.04.1] | 17:19 | |
| -queuebot:#ubuntu-release- Unapproved: rejected ubuntu-advantage-tools [source] (xenial-proposed) [27.0.2~16.04.1] | 17:20 | |
| -queuebot:#ubuntu-release- Unapproved: ubuntu-advantage-tools (bionic-proposed/main) [27.0~18.04.1 => 27.0.2~18.04.1] (core) | 17:27 | |
| -queuebot:#ubuntu-release- Unapproved: ubuntu-advantage-tools (groovy-proposed/main) [27.0~20.10.1 => 27.0.2~20.10.1] (core) | 17:27 | |
| -queuebot:#ubuntu-release- Unapproved: ubuntu-advantage-tools (focal-proposed/main) [27.0~20.04.1 => 27.0.2~20.04.1] (core) | 17:27 | |
| -queuebot:#ubuntu-release- Unapproved: ubuntu-advantage-tools (hirsute-proposed/main) [27.0~21.04.1 => 27.0.2~21.04.1] (core) | 17:28 | |
| -queuebot:#ubuntu-release- Unapproved: accepted ubuntu-advantage-tools [source] (hirsute-proposed) [27.0.2~21.04.1] | 17:30 | |
| -queuebot:#ubuntu-release- Unapproved: accepted ubuntu-advantage-tools [source] (groovy-proposed) [27.0.2~20.10.1] | 17:31 | |
| -queuebot:#ubuntu-release- Unapproved: accepted ubuntu-advantage-tools [source] (focal-proposed) [27.0.2~20.04.1] | 17:31 | |
| -queuebot:#ubuntu-release- Unapproved: accepted ubuntu-advantage-tools [source] (bionic-proposed) [27.0.2~18.04.1] | 17:32 | |
| -queuebot:#ubuntu-release- Unapproved: accepted evolution-data-server [source] (focal-proposed) [3.36.5-0ubuntu1] | 18:12 | |
| -queuebot:#ubuntu-release- Unapproved: cloud-init (groovy-proposed/main) [21.1-19-gbad84ad4-0ubuntu1~20.10.2 => 21.2-3-g899bfaa9-0ubuntu2~20.10.1] (core, ubuntu-cloud) | 18:24 | |
| -queuebot:#ubuntu-release- Unapproved: accepted libsecret [source] (focal-proposed) [0.20.4-0ubuntu1] | 18:45 | |
| -queuebot:#ubuntu-release- Unapproved: cloud-init (focal-proposed/main) [21.1-19-gbad84ad4-0ubuntu1~20.04.2 => 21.2-3-g899bfaa9-0ubuntu2~20.04.1] (core, edubuntu, ubuntu-cloud) | 19:47 | |
| -queuebot:#ubuntu-release- Unapproved: nvidia-modprobe (hirsute-backports/multiverse) [460.32.03-1 => 465.24.02-1~ubuntu21.04.1] (no packageset) | 20:15 | |
| -queuebot:#ubuntu-release- Unapproved: nvidia-modprobe (groovy-backports/multiverse) [460.32.03-0ubuntu20.10.1 => 465.24.02-1~ubuntu20.10.1] (no packageset) | 20:15 | |
| -queuebot:#ubuntu-release- Unapproved: nvidia-modprobe (focal-proposed/multiverse) [460.32.03-0ubuntu20.04.1 => 465.24.02-1~ubuntu20.04.1] (no packageset) | 20:19 | |
| Eickmeyer[m] | ^ nvidia-modprobe: SRU bug 1928146 | 20:20 |
| ubot3 | Bug 1928146 in nvidia-settings (Ubuntu Hirsute) "Please SRU nvidia-modprobe 465.24 and nvidia-settings 460.73" [High, New] https://launchpad.net/bugs/1928146 | 20:20 |
| -queuebot:#ubuntu-release- Unapproved: cloud-init (bionic-proposed/main) [21.1-19-gbad84ad4-0ubuntu1~18.04.2 => 21.2-3-g899bfaa9-0ubuntu2~18.04.1] (edubuntu, ubuntu-cloud, ubuntu-server) | 20:47 | |
| blackboxsw | bdmurray: if there is time for an SRU review of ubuntu-advantage-tools 27.0.2 for Bionic -> HIrsute it would be much appreciated. We are holding on Xenial SRU of 27.0.2 until we sort whether it is ok to depend on golang-1.10-go/universe on Xenial (instead of golang-1.6-go which is in main). The previous SRU of 27.0.1 was held up on bionic due to the same changeset causing FTBFS on riscv64 focal. | 21:38 |
| * blackboxsw rbasak: uploaded this morning and I just got through verification now | 21:38 | |
| * blackboxsw our hope with 27.0.2 is to avoid aging since the changes 27.0(.1) changes were aging for 11 days until we rejected them in order to provide 27.0.2 package build fixes | 21:41 | |
| * blackboxsw I will now search for why that /me prefix keeps getting injected by my client | 21:42 | |
| -queuebot:#ubuntu-release- Unapproved: rejected ubuntu-release-upgrader [source] (hirsute-proposed) [1:21.04.12] | 21:52 | |
| mwhudson | vorlon: https://bugs.launchpad.net/ubuntu-cdimage/+bug/1928264 | 22:14 |
| ubot3 | Launchpad bug 1928264 in Ubuntu CD Images "release ISOs can contain phasing" [Undecided, New] | 22:14 |
| vorlon | mwhudson: ta | 22:14 |
| bdmurray | blackboxsw: I was in a meeting let me see about this. | 22:28 |
| * blackboxsw thanks bdmurray. | 22:28 | |
| * blackboxsw darnit | 22:28 | |
| bdmurray | blackboxsw: 27.02 is stuck in -proposed for impish due to an autopkgtest failure. Do you know anything about this? | 22:30 |
| * blackboxsw bdmurray: if it was i386 robie tried to put a hint to ignore | 22:31 | |
| bdmurray | no its armhf | 22:32 |
| blackboxsw | ahh we looked at that too. let me recall | 22:32 |
| * blackboxsw we did see something earlier, I thought robie re-kicked the job as it looked like it may have been related to old update_motd which I think changed this week per a fix from rb_alint . Right now I see success on https://autopkgtest.ubuntu.com/packages/ubuntu-advantage-tools | 22:33 | |
| blackboxsw | ahh it's <ctrl> enter adds the /me. | 22:33 |
| blackboxsw | MM vs IRC fail | 22:34 |
| bdmurray | https://autopkgtest.ubuntu.com/packages/u/update-motd/impish/armhf | 22:34 |
| blackboxsw | ahh yes his comment was, looks unrelated to any of our changes in 27.0.2 and we also know that update-motd was just updated a couple days ago | 22:35 |
| blackboxsw | I'm looking over update-motd changes now to see what recently landed | 22:36 |
| bdmurray | update-motd 3.8 passed w/ u-a-t/27.0.1 though | 22:36 |
| blackboxsw | our changes 27.0.1 impish vs 27.0.2 would just be package build diffs that include or exclude our apt hook on powerpc + riscv64. hrm. I'll try grabbing the arm debs build in ppa:ua-client/staging | 22:37 |
| * blackboxsw not certain as the apt-esm-json-hook was properly built/tested and included in https://pastebin.ubuntu.com/p/qzJ23cGDp3/ | 22:40 | |
| blackboxsw | trying to see if I can get an autopkgtest env together for arm | 22:40 |
| bdmurray | blackboxsw: I have something I can try | 22:44 |
| blackboxsw | I'm reading through update-motd/debian/tests/wait-for-apt-update-to-finish now to see what's up | 22:48 |
| blackboxsw | able to run on clean impish machine root@dev-i:~# /wait-for-apt-update-to-finish | 22:52 |
| blackboxsw | Attempting to lock /var/lib/apt/lists/lock to wait for any currently running apt updates to finish... locked, exiting. | 22:52 |
| * blackboxsw also if no apt lock is held that test gives a traceback https://pastebin.ubuntu.com/p/zQtRFvbCf4/ | 22:54 | |
| bdmurray | blackboxsw: when you said "sru review" I interpreted that to mean a review of the package in the unapproved queue not release of a package in -proposed. | 23:04 |
| * blackboxsw bdmurray: probably needs a more full day do you think? I know sil_2100 looked intially over 27.0.1 and raised the build failures, maybe it's worth pinging for review start of sil's day do you think? | 23:06 | |
| blackboxsw | if I cut off my left pinky I may avoid the /me | 23:06 |
| bdmurray | blackboxsw: also bug 1927795 isn't tagged verification-done-groovy | 23:07 |
| ubot3 | Bug 1927795 in ubuntu-advantage-tools (Ubuntu Xenial) "27.0 package build failures focal riscv64 architecture" [Undecided, New] https://launchpad.net/bugs/1927795 | 23:07 |
| blackboxsw | bdmurray: I wasn't sure why verification-needed-groovy wasn't an included tag on the bug. I did confirm package builds succeed both in the ppa, plus confirmation that sbuild locally on hirsute riscv64. I can run through sbuild groovy too if that further confirms no pkg build failures. | 23:10 |
| blackboxsw | per this comment https://bugs.launchpad.net/ubuntu/+source/ubuntu-advantage-tools/+bug/1927795/comments/7 | 23:11 |
| ubot3 | Launchpad bug 1927795 in ubuntu-advantage-tools (Ubuntu Xenial) "27.0 package build failures focal riscv64 architecture" [Undecided, New] | 23:11 |
| bdmurray | blackboxsw: Is it building in the archive not enough? https://launchpad.net/ubuntu/+source/ubuntu-advantage-tools/27.0.2~20.10.1/+build/21540085 | 23:11 |
| bdmurray | There was probably an LP timeout so the groovy verification needed comment and tag were not added. | 23:11 |
| blackboxsw | right that is enough. was just doubling the effort. | 23:11 |
| blackboxsw | and doubling the logs to read | 23:12 |
| bdmurray | but not doublemint gum | 23:12 |
| blackboxsw | hah! | 23:12 |
| blackboxsw | I am still trying to reproduce the lock failure in update-motd impish | 23:13 |
| bdmurray | IDK that would have been a blocker / concern for me accepting it into -proposed, less for accepting it into -updates. | 23:14 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!