=== wgrant__ is now known as wgrant === g2 is now known as g2[ATL] === G_ is now known as G [03:51] hi LocutusOfBorg [03:53] are you still doing sponsorship for debian? === g2[ATL] is now known as g2 [06:46] jbicha, mozjs24 is a merge or a sync? [08:34] karstensrage, please debdiff the package with the one in unstable [08:37] missing changes, e.g. new identity4c package, wrong version (dpkg --compare-versions helps), target should be sid, not "stable", the first changelog timestamp has been changed, old std-version, vcs fields non canonical (see mentors link), pre-depends should be dropped now [11:23] LocutusOfBorg: I'd prefer it if we could RM mozjs24 … here's what's blocking its final rdepends: https://bugs.debian.org/872040 [11:23] Debian bug 872040 in src:duktape "duktape: please ship a duktape library" [Wishlist,Open] [11:24] and I'd like to drop mozjs38 too but that's blocked on https://github.com/linuxmint/cjs/issues/52 [14:20] LocutusOfBorg, thanks, will do === JanC_ is now known as JanC [14:53] LocutusOfBorg, done [15:05] karstensrage, std-version is 4.1.0, dh-autoreconf is useless in compat level 10, changelog doesn't list changes [15:06] bump debian/compat and debhelper to 10 [15:06] and drop --with-autoreconf from rules [15:06] and update the changelog file [15:07] debian/changelog.j2 <-- what is that? [15:07] is that in there [15:08] yikeks [15:08] LocutusOfBorg, thats a template file since everyone wants different versions for their builds [15:08] it shouldnt be in there... [15:09] and does changelog only reflects package changes right? [15:10] upstream in case they fix a debian bug [15:10] and packaging yes [15:10] so all these changes you're suggesting reflected in changelog? [15:11] hmm have to figure out how to get rid of that .j2 [15:11] do a debdiff of the current unstable package [15:11] and check it [15:21] seems to be a bug in openssh-client-ssh1, it fails if ~/.ssh/config has a stanza with "Protocol 1", same as the regular ssh client. [16:36] is sysvinit-utils still supposed to be essential in artful? [17:40] bdmurray: hello, would you be able to take a look at promoting neutron-lbaas-dashboard to zesty-updates when you get a chance? [17:48] coreycb dosaboy - fyi, promoted neutron to uca mitaka-proposed re: https://bugs.launchpad.net/neutron/+bug/1668410 [17:48] Launchpad bug 1668410 in neutron "[SRU] Infinite loop trying to delete deleted HA router" [Medium,In progress] [17:49] beisner: thanks sir [17:56] coreycb: It doesn't seem to be verified [17:57] coreycb dosaboy - fyi, promoted nova from staging to proposed in uca M, N, O for https://bugs.launchpad.net/neutron/+bug/1668410 [17:57] Launchpad bug 1668410 in neutron "[SRU] Infinite loop trying to delete deleted HA router" [Medium,In progress] [17:57] bdmurray: urg sorry, too many tags [18:19] chrisccoulson: fyi, there is a new cargo in unstable, and a new rustc in experimental [18:36] doko: uploaded tomcat8.0 to debian & artful, assuming it'll take too long to process through debian NEW to reach artful.. [18:38] tjaalton: accepted [18:38] :D [18:38] Hi. [18:38] tjaalton: please keep a bug report open to remove it for 18.04 [18:38] 17.10 crashes every time I fill up the RAM. [18:38] doko: much thanks, I'll push new tomcatjss etc later [18:38] doko: yup [18:38] Anyone else seeing that? [18:40] It's almost as though the new swap (file) is broken in some way. [18:44] does anyone know a way to test if it's the swap file that's crashing Ubuntu 17.10? [18:45] in the log it says /failed to activate swap file [18:45] cajhne_: then you aren't usinng the sawp file and it probably OOM'd [18:46] nacc: thanks. I'm surprised there's no bug report for this. [18:48] cajhne_: what's the bug? swap failed to activate and you used up all your memory [18:48] cajhne_: swap failing to activate isn't itself a bug [18:48] nacc: I don't understand... that's expected behaviour? To crash and not swap anything to disk? [18:48] cajhne_: without understanding *why* [18:49] all I can think of is that I made the horrible error of choosing "encrypt my home folder" [18:49] on install. [18:50] cajhne_: how can it swap when there is no swap? [18:50] Other than that, there's nothing at all unusual about my install. [18:50] cajhne_: did you put your swapfile in your /home? [18:50] rbalint: If you wondered why I did not upload the apt SRU yet: I was waiting to get this ACKed upstream (in Debian) for stretch / 1.4.8, but this did not happen yet. I guess I'll have to rebuild with a different version number. very unfortunate after all the work I put in :( [18:50] nacc: Ubuntu has always set up swap however it likes. Does it not do that anymore? [18:51] If not, yes, I'd consider that a pretty major bug. [18:52] cajhne_: of course it does [18:52] cajhne_: also, i think you might be in the wrong channel, did you want #ubuntu? [18:52] nacc: for the install I chose to wipe the disk and install ubuntu 17.10 [18:52] cajhne_: or #ubuntu+1 [18:53] I'm ready to report a bug, but wanted to see if it's a bug. [18:53] cajhne_: right, so wronng channel [18:54] nacc: usually I work with devs to troubleshoot bugs. This is for the upcoming release. [18:55] cajhne_: this appears to be a support request, and this is not the support channel (as pointed out in /topic) [18:55] cajhne_: hence use #ubuntu or #ubuntu+1 as appropriate (imo) === cajhne_ is now known as CRogers [19:08] juliank: thanks for the reminder [19:09] is #ubuntu broken? [19:09] CRogers: you need to be registered [19:09] CRogers: is that what you mean? [19:09] I am. [19:09] CRogers: what are you seeingn? [19:10] ubuntu channel with 0 people in it. [19:10] juliank: i'm working on issues surfaced in u-u thus i'm not blocked on missing apt but i hope apt gets accepted [19:10] no way to type anything. [19:10] CRogers: well, i'm in #ubuntu now ... are you on the wrong server? [19:10] 1210 -!- Irssi: #ubuntu: Total of 1082 nicks [1 ops, 0 halfops, 0 voices, 1081 normal] [19:11] seems you're missing SOMEthing, CRogers [19:11] nacc: freenode, right? [19:11] CRogers: yeah [19:11] yea. :( [19:11] i'm in #ubuntu and according to whois, you're not, CRogers [19:11] * CRogers grrrrs [19:11] Okay, maybe it's Polari. [19:11] can you invite me to the channel? [19:12] the one that you think is #ubuntu [19:12] wxl no, cant type anything. [19:12] maybe I'll re-start polari. [19:12] i'm not asking you to dialogue, but to send an irc command [19:12] if you can't type anything (versus the server refuses your entry for some reason) that's certainly a client issue [19:12] wxl: I can do nothing in that channel. Polari isn't letting me. [19:13] CRogers: /invite wxl #ubuntu [19:13] sarnold: I typed that here, but I don't think it wil do any good. [19:14] brb === cajhne is now known as CRogers [19:15] Heh [19:15] Nickserv is like: "Identified for CRogers" [19:15] Polari is like... nah. :P [19:16] It's definitely a bug in Polari. [19:16] Sorry for the noise. [19:53] bdmurray: ok i've completed verification for https://bugs.launchpad.net/ubuntu/+source/neutron-lbaas-dashboard/+bug/1709604 now [19:53] Launchpad bug 1709604 in neutron-lbaas-dashboard (Ubuntu Zesty) "package installs and enables two panels" [High,Fix committed] [19:53] bdmurray: i promise :) === klebers_ is now known as klebers [21:12] anyone seen this ? my schroot build fails [21:12] http://paste.ubuntu.com/25568080/ [21:16] smoser: https://launchpad.net/ubuntu/+source/python3-defaults/3.6.2-1ubuntu2 [21:18] but that same problem is causing the fix to ftbfs :( [21:18] doko: ^ [21:22] ok LocutusOfBorg im not sure what you are intending to communicate with "do a debdiff of the current unstable package" i have done some debdiffs of the dsc and the changes look correct [21:23] but if you are seeing something else please let me know [21:25] jbicha: The fix will build once the deletion is published. [21:26] Or unpublished, as it were. [22:18] karstensrage, pull-debian-source packagename [22:18] debdiff it with the one I should sponsor [22:18] and filterdiff for the debian directory [22:48] LocutusOfBorg, like https://paste.ubuntu.com/25568716/ [22:56] karstensrage: Why the new package? [22:57] karstensrage: libufpidentity-dev is for dev, and if something's linked against the library, it will get a dependency on libufpidentity1, so identity4c seems entirely pointless. === Guest41832 is now known as RAOF [23:23] infinity, the upstream package was updated [23:23] karstensrage: Yes... And? [23:23] karstensrage: That doesn't really answer my question. :) [23:23] i thought this is what i had to do [23:23] karstensrage: Why the new *binary*? [23:23] binary? [23:24] someone is reporting (well, technically they are complaining that nont all flavors do this) that mate enables the upstream oracle virtualbox repo by default [23:24] karstensrage: That diff adds a new binary package, identity4c.deb, whose only purpose is to be an empty package that depends on the library. [23:24] that seems ... surprising if so, as it would lead to a rather different UX only on mate? [23:25] nacc: That would be a violation of (admittedly, not well-documented) policy. [23:25] infinity, thats how it was originally packaged [23:25] karstensrage: Except, it isn't? [23:25] infinity: that was my guess too, and wasn't where to find said policy :) (still looking to see if it's true) [23:25] nacc: eww. [23:25] karstensrage: It wouldn't be showing up in the diff if that was how it was previously packaged. [23:25] infinity: sarnold: i'll get a live instance going and see [23:26] im not sure i understand [23:26] im sorry [23:26] im not well versed in packaging [23:26] karstensrage: there's no identity4c binary package in Ubuntu [23:26] karstensrage: and your debdiff implies one is being added [23:26] karstensrage: The current source package produces two binaries: libufpidentity-dev and libufpidentity1. Your updated version also produces a identity4c package that depends on the library. [23:28] hmm ok [23:28] im not sure why that is the case [23:28] it was only my intention to update the upstream [23:28] Also, those Vcs-* links still seem to point nowhere useful. [23:29] yes im not exactly sure what those are supposed to be, i just followed the convention in the maintainers guide [23:29] i would be happy to remove them if thats allowed [23:30] If there's no Vcs, having the links is wrong. [23:30] ok [23:30] But I'm more curious about how you "accidentally" add a new binary package to debian/control. ;) [23:30] (Or is this someone else's changes?) [23:30] karstensrage: how did you do the upstream update? uupdate? [23:31] oh i see what youre saying now... hmm [23:31] wait [23:31] oh oh [23:32] i see what youre saying now [23:32] yes nacc maybe uupdate [23:32] i dont think i "added" that, i think it was added for me by something [23:32] That would be pretty much impossible. [23:32] i can definitely make it look like it looked before, im sorry for the misunderstanding [23:33] No automatic tool would have written that Description. [23:33] uupdate won't do that, afaik (it will only buimp your changelog and create a new dir for the updated version) [23:33] infinity: good point :) [23:33] unless ... skynet? :) [23:33] ah ok [23:34] i see more now [23:34] what a sad (sly?) way to show self-awareness, though! subtly modifying our debdiffs [23:34] Next weekend project (on my 1,000,000,000 item long list of them), build a bot to do exactly that, write debian/control descriptions :P [23:34] karstensrage: Just to deal with my own confusion here a bit, are you 'richardl@ufp.com'? [23:34] yes [23:34] Okay. :) [23:34] so i did this a long time ago [23:34] and its hard to remember what occurred when [23:34] tsimonq2: heh [23:35] so there are three possible things to upload to, debian, launchpad, (ppa's) and ubuntu [23:35] karstensrage: Yeah. Well, Debian and then sync to Ubuntu, since we're already in sync. [23:35] i may have indicated i wanted to do apt-get install identity4c and that is what makes that happen [23:35] karstensrage: But then PPAs if you feel the urge. [23:35] and all three need different forms of versioning [23:36] iirc [23:36] karstensrage: Anyhow, libraries shouldn't have a way to install them that way. That's not particularly reasonable. [23:36] karstensrage: Libraries should be pulled in as dependencies of things that link to them. [23:36] hmm [23:37] karstensrage: For instance, there's no way to "apt-get install glibc", you just get it because, well, everything else depends on it. [23:37] And if you want to develop C, you "apt-get install libc6-dev", or in your case, libufpidentity-dev [23:38] karstensrage: So, I get the intent, just saying it's not Debian library maintenance best practices or policy. [23:39] (But drop that delta, and I'd be happy to sponsor the rest to Debian for you, since I assume the issue here is that you're not currently a DM or DD?) [23:39] thats correct im not a DM or DD [23:40] this update is for another module i plan to package as well [23:40] [23:40] Heh. [23:41] so would you be willing to look over another package and tell me if thats the right track? [23:41] and maybe sponsor that too? [23:41] i will remove that delta and resubmit, i do agree that its odd but i cant recall why or where it got added, i realize now i must have put it in for apt-get install identity4c [23:42] but youre right thats a bit odd [23:42] karstensrage: I'd be willing to look over more bits, sure. Also, yes, you should remove the Vcs-* lines if they point to no VCS. ;) [23:42] karstensrage: Looks like libpam-ufpidentity has the same VCS blackhole bug. [23:43] yeah thats the other one i wanted to know if it was the right track [23:43] there is another library, dependent on this new stuff, libnss_ufp [23:43] youre going to absolutely LOVE that one [23:43] :P [23:43] Of course. Can't have a PAM module without an NSS module. [23:43] right [23:43] I mean, you can, but they do seem to come in pairs. :) [23:44] but i felt it was important to mimic glibc [23:44] so it creates its so just like the others in /lib/ [23:44] Famous last words. [23:44] yeah [23:44] ok let me remove that package and resubmit [23:45] karstensrage: I also note you haven't tagged a 1.1.0 on github yet. [23:46] karstensrage: Were you waiting to do that after the package review, in case the reviewer(s) found something you should fix upstream? [23:46] yes and to make sure all the stuff went together, since this is for another library [23:46] the NSS thing [23:47] its much easier for clients to install from packaging, no one really goes to github [23:47] Check. So this is a bit of a sanity check review, not a "please upload right now" review, since there's no upstream orig tarball yet? [23:47] correct [23:47] Works for me. [23:47] nacc: I assume the 3rd party repositories are facilitated by the ubuntu-mate-welcome app [23:48] jbicha: And such facilitation is entirely okay, if it's opt-in. If the assertion that it's opt-out (or no opting at all) is true, then it's a grave violation of all things Good and Pure and True. [23:48] I also assume the Tech Board is generally aware of that app. It's an old Tech Board policy that requires rare, documented exceptions for enabling 3rd party repos by default [23:48] Well, opt-in with some solid explanation of WTF you're signing up for. [23:49] fwiw, i think jbicha is right [23:49] and ubuntu-mate seeds the ubuntu-mate-welcome package [23:49] so it's opt-out, afaict? [23:49] I believe I've been off and on aware of it over the years. :) [23:49] nacc: ubuntu-mate-welcome enables those repos by default? [23:49] infinity: i'm still downloading the iso, will let you know soon [23:49] infinity: the above was just as stmt on that package, sorry [23:50] Anyhow, if it does it by default, it's a pretty nasty bug that we should have caught earlier. [23:50] nacc: I believe the additional repos aren't enabled by default but it's very easy to do so and may not be obvious to users what they're doing [23:50] If it does it on an opt-in basis, but without sufficient warning, it's a normalish bug that should be addressed. [23:50] If it's opt-in with flashing warnings and sirens, there's nothing to see here. [23:50] jbicha: infinity: ack, will check [23:50] anyway, the Tech Board is aware now ;) [23:50] heh [23:51] btw, Ubuntu Budgie has a welcome app now too [23:51] jbicha: BTW, that python3-defaults thing finally got over the hump, if you had any specific builds that needed retrying. [23:52] infinity: thanks, already retried :) [23:52] jbicha: Though, it might be that time of year where I should just retry all of artful-proposed so people can whine about spam. [23:52] (Note that whining about spam generated by the FTBFS packages YOU uploaded has always seemed a bit silly to me) [23:52] ok uploaded [23:52] I used to get so excited when the Release Team would send me an email, but then disappointment [23:52] i think that might be what you wanted [23:52] jbicha: Hahaha. [23:52] now they email me all the time :| [23:53] karstensrage: No idea where you uploaded to (note I was commenting on your pasted diff before). [23:53] karstensrage: Though, seeing the whole package would indeed be better, if you have a pointer to that. [23:53] infinity: somebody already retried all of artful-proposed last week-ish, this might be why you get complaints about spam ;) [23:54] slangasek: Was that somebody you? [23:54] \no [23:54] I suspect doko [23:54] If it wasn't me, I question if it was "all of", unless someone wrote a very, very, very slow API client to do it. [23:54] infinity, https://mentors.debian.net/package/identity4c [23:54] Since I'm one of the few with access to do it directly from ftpmaster without seven hours of round trip hell. [23:55] karstensrage: Ta. [23:55] infinity: well, I got a /lot/ of random build failure mails, but yeah, not sure [23:55] But also, I figure the solution to "I don't like FTBFS mails on old uploads" is "fix your old upload". [23:56] any AA volunteers to remove the ostree binaries from artful/s390x, general tracker bug is LP: #1712083 [23:56] Launchpad bug 1712083 in gjs (Ubuntu) "Please remove gjs/s390x" [Undecided,Fix released] https://launchpad.net/bugs/1712083 [23:56] I get more than most, since I "own" all the proposed-to-proposed forward-copies, but most others shouldn't have that excuse. [23:56] the Debian ostree package build-depends on s390x already so it currently doesn't build on s390x at all [23:56] *build-depends on gjs [23:57] Ugh. [23:57] WHY DO PEOPLE DO THIS. [23:57] Maybe I should just make the tools not allow it. [23:57] So, the problem with removing existing published binaries is that they'll come back if the source is copied around. [23:58] Like, say, when opening a new series. [23:58] The correct way to do this is to upload a mozjs and gjs that no longer produce s390x binaries, than reupload all the rdeps so they end up dep-waiting.