=== ben_r_ is now known as ben_r === popey8 is now known as popey === Eleventh_Doctor is now known as Pharaoh_Atem [08:14] vorlon, hey, I probably already asked that, but is there a manual process for adding source to build on i386? gedit picked up some new depends, amtk & tepl ... will those be pulled on the whitelist automagically by depends chain or is there any manual work involved? [08:15] vorlon, other i386 question, libsoup2.4 fails to build now due apache/php i386 installability problem, is that a known issue? [08:20] vorlon, oh and mozjs68 is about to replace mozjs60 for GNOME so we also need to get it built on i386, again unsure if I should do some manual work or if that's going to automatically happen due to depends? [09:42] seb128: Just saw your i386 comments. For bootstrapping you need to add them to the code of update-i386-whitelist in lp:ubuntu-archive-tools, re-run that to update the packageset, get the build record re-created (copy the package over itself), and then once the leaf package has picked up the dependencies on i386 germinate should keep it in the seed after that, so it can be dropped from that script. [09:44] Laney, thx [09:44] Laney, vorlon, is that process documented somewhere? [09:45] I'm probably not the only one who is going to hit that kind of question [09:46] dunno [09:46] is there a place for archive admin docs? [09:46] Laney, https://wiki.ubuntu.com/ArchiveAdministration [09:47] looks like a good place [09:53] vorlon, could you write a i386/whitelist on there ^? [09:54] +section [10:06] !dmb-ping is tsimonq2, rafaeldtinoco, slashd, teward, sil2100, ddstreet, rbasak: DMB ping [10:26] !no dmb-ping is ddstreet, rafaeldtinoco, rbasak, sil2100, slashd, teward, tsimonq2: DMB ping [10:26] I'll remember that JackFrost [10:29] sounds threatening [10:30] JackFrost: thanks. Was my command correct? It's from our checklist. [10:31] rbasak: It's close enough, you have to be an editor in order to edit factoids too though. [10:32] Thanks - yeah I had a private message back saying it was being moderated [11:41] FYI: On the Ubuntu 20.04 docker image, command-not-found is not working: /usr/lib/cnf-update-db , KeyError: 'suite' [15:06] if a package needs root for the build time tests triggered by dh_auto_test what would be the best way to provide it that? [15:07] Fake it somehow or move them to autopkgtest [15:07] You can't have root in dh_auto_test [15:07] (real root that is) [15:07] Maybe fakeroot would do though? [15:09] I need to check if I can fake it, it seems most parts work already ~140 of 160 tests [15:09] what fails right now is that it wants to read /boot/vmlinuz-... [15:09] which is "-rw------- 1 root root" [15:10] In that case your choices are to install some kind of mock version (if appropriate/possible), to disable those particular tests, or to run the test suite in autopkgtest instead [15:10] Or some combination [15:10] yeah, I guess I skip them and move to autopkgtest [15:12] cjwatson: might I ask - is that different in Debian? as there the test works [15:12] No [15:12] Something else must be different [15:12] Probably /boot/vmlinuz-* is readable in Debian's kernel packages [15:13] checking ... [15:13] I vaguely remember that's something Kees got changed in Ubuntu because it impeded automatic operation of some rootkits (even though the actual contents of the kernel image are of course public knowledge) [15:13] yep - -rw-r--r-- [15:14] ok, now things fit together and on that I can at least add my changes "knowing why" [15:14] thanks cjwatson [16:17] seb128: I wasn't aware of the libsoup2.4/apache issue, no [16:17] seb128: re: docs, sure but I'm not sure when I'll get to it [16:17] vorlon, hey, k, I will try to poke at it === alan_g_ is now known as alan_g [16:29] Hi, can anyone here restart building of netboot image for focal ? It is not there in cdimage... [16:38] http://cdimage.ubuntu.com/netboot/focal returns error [16:38] vorlon, do you know how close we are from icu migrating? we started some other transitions (poppler/gnome-desktop) which have some common package, we didn't rebuild those yet since I think it's better to wait for icu to migrate first before risking complicating things [16:39] seb128: it is better to wait, thanks. I don't know exactly how close it is, I would've hoped it to migrate today during Europe's day. xnox are you working on this today? [16:39] seb128: I am going to be working on it today and I think the only remaining blockers are all test failures that need to be worked through [16:40] vorlon, k, anything desktopish/we could help with? [16:40] seb128: find packages in update_excuses that are listed as depending on icu and not candidates due to autopkgtest failures, and help resolve those [16:40] looks like rdkit/s390x still needs resolved, I'll do that now [16:41] vorlon, ack [17:14] vorlon: previously some TB members have had the developer-membership-board@ admin password. [17:15] vorlon: what do you think I should ask IS to do? Reset it and give it to you? [17:15] rbasak: I was going to say you should have them update the list membership [17:15] I dunno, stgraber might have the password [17:15] but the list management stuff is pretty lame overall in terms of credentials sharing, so I'd rather just make it IS's problem ;) [17:16] Previously the TB and the DMB both had the password [17:16] I guess it's good practice to rotate it when either group's membership changes ... [17:16] I may still have it in my listadmin config [17:17] I have the moderator password [17:17] But I can't admin from there AFAICT. [17:17] hi, which package should I use for a bug report about notifications? [17:17] just "mutter", and let them reassign as needed? [17:18] I have *a* password too [17:18] gnome-shell [17:18] thanks [17:18] ideally report upstream too :-) [17:24] Found a password [17:26] It works [17:26] I might as well just make the changes [17:26] And then someone can tell me who should have it later, and I can do that. [17:35] vorlon: rdkit needs rebuild hit [17:35] xnox: it did. it's been done [17:35] vorlon: i was waiting for pandas removal to be published [17:36] vorlon: from my point of view icu is done; if you want me to fix anything else let me know. [17:36] vorlon: i was working on completing ocaml [17:36] xnox: it's not done until it's in the release pocket [17:40] vorlon: it's a valid candidate.... or are you expecting me to fix all of nodejs & R too? [17:41] xnox: you said you were managing the *transition*. It's not done until all the reverse-dependencies can go through together [17:42] xnox: r-base has two autopkgtest failures blocking it, I don't see that anyone has checked yet if those are present in the release pocket (I've done this now) [17:42] vorlon: i didn't start icu. Somebody accepted icu NEW without coordination. I only stepped in to clean up fallouts when it was half way through. [17:43] vorlon: when i saw icu in NEW i asked AA to reject it, to avoid it entagling with everything. And instead AA accepted it. [17:43] hence we are here now. === ben_r_ is now known as ben_r [21:38] oh huh go 1.14 release [21:39] And already in Debian.