[01:14] LocutusOfBorg: Thanks, mate! === tacobat` is now known as tacobat [05:19] slangasek: When you have a second, can you poke at https://bugs.launchpad.net/ubuntu/+source/lvm2/+bug/1573982 ? [05:19] Launchpad bug 1573982 in lvm2 (Ubuntu) "LVM boot problem - volumes not activated after upgrade to Xenial" [Undecided,Confirmed] [06:33] Unit193, I don't know why, but you are welcome :) [06:33] Irssi. [06:34] LP never notified me, so didn't notice. [06:34] :) [06:34] thanks tsimonq2 :) === ledeni_ is now known as ledeni === StevenK_ is now known as StevenK [13:36] hi, can someone please accept the bionic task for me for this bug? https://bugs.launchpad.net/debian/+source/postfix/+bug/1753470 [13:36] Launchpad bug 1753470 in postfix (Ubuntu) "Postconf segfaults every 5 minutes" [Low,Fix released] [13:41] ahasenack: Done [13:41] StevenK: thanks [14:32] ahasenack is getting https://lintian.debian.org/tags/hardening-no-bindnow.html for a proposed new package. On review, I asked why, if hardening=+bindnow is needed, why it isn't already default in dpkg-buildflags. What would be best practice here? For all Ubuntu-specific packages to individually maintain a hardening=+bindnow in debian/rules, for Ubuntu to adjust dpkg-buildflags when we want everything [14:33] to do it, or to just ignore the lintian note? [14:34] https://wiki.ubuntu.com/Security/Features#bindnow seems relevant [14:36] That suggests to me that it should be on by default on amd64 already? [14:39] On Bionic, I have to specify hardening=+bindnow to get -Wl,-z,now in LDFLAGS [14:41] rbasak: I use export DEB_BUILD_MAINT_OPTIONS = hardening=+all in many of my packages [14:41] even with that, I get the lintian warning for some of the packages [14:42] I understand that in Debian the general approach is one where maintainers slowly opt in to a new thing. [14:43] In Ubuntu though, I'm not sure that makes sense. [14:43] If project-wide we decide to enable something, we can just do it across some set of packages. [14:44] And that's how hardening flags have been enabled over time in Ubuntu, AIUI. [16:18] rbasak: indeed, our gcc is set to enable bind-now by default if its being built with pie. I have to step into a meeting in a little bit, but can you point me at ahasenack's code? [16:19] sbeattie: https://git.launchpad.net/~ahasenack/ubuntu/+source/ndctl/tree/debian/rules [16:19] and https://git.launchpad.net/~ahasenack/ubuntu/+source/pmdk/tree/debian/rules [16:20] upstream is https://github.com/pmem/pmdk and https://github.com/pmem/ndctl [16:24] sbeattie: it doesn't seem like it works, for instance, lintian says that bindnow isn't used if you build https://salsa.debian.org/gnome-team/gnome-characters after [16:24] commenting out the hardening line in debian/rules [16:27] yeah, same for me [16:27] I removed the line that added hardening=+all and got the lintian warning [16:28] ahasenack: how are you building the package? I'm not that proficient with gpb [16:29] checkout that branch (master), sudo apt-get builddep ./, dpkg-buildpackage -uc -us [16:29] you'll need the orig tarball, it's the upstream release from github [16:30] ahasenack: ah, okay [16:30] sbeattie: if it helps, I think that the build uses dpkg-buildflags, which should show you everything that's going on wrt. flags including interpreting DEB_BUILD_MAINT_OPTIONS=hardening=+foo etc [16:30] If that's true, then you should be able to experiment just with that and not the entire build environments I think. [16:31] sbeattie: pmdk takes a while to build, ndctl is much faster [16:31] ahasenack: thanks [16:31] * sbeattie vanishes into a meeting black hole [16:31] * ahasenack -> lunch === blackboxsw is now known as blackboxs === blackboxs is now known as blackboxsw [17:44] sbeattie: oh, I forgot, there is a ppa which you might find more convenient: ppa:canonical-server/nvdimm it has both [18:08] ahasenack: are you seeing this with bionic builds or cosmic builds? [18:09] bionic, haven't tried cosmic yet, but will [18:12] okay, I'm not able to reproduce in a local bionic schroot, rebuilding with the DEB_BUILD_MAINT_OPTIONS line commented out entirely gives me a daxctl binary that hardening-check tells me is set up with immediate binding. [18:14] how is lintian being invoked? [18:15] (I don't see it in my ndctl buildlog) [18:17] sbeattie: lintian -iI --pedantic $@ [18:17] it complains about the libraries [18:22] ahasenack: ah okay, I see that. [19:08] infinity: could you handle bug 1769694? [19:08] bug 1769694 in ubuntu-community "Please update DMB members" [Undecided,New] https://launchpad.net/bugs/1769694 [19:09] jbicha: Yep, on it. [19:11] jbicha: Hrm, the bug links to the IRC meeting, but I don't actually see the election results anywhere to confirm. :P [19:12] jbicha: And the bug fails to mention what's up with the other two expiring members. Ben and Brian stepping down tomorrow? [19:14] yes [19:14] were you looking for https://civs.cs.cornell.edu/cgi-bin/results.pl?id=E_b99fd5c7c9572192 ? [19:14] jbicha: Exactly that, yes. Thanks. [19:14] jbicha: Oh, I see, 5 slots and 5 candidates. "Democracy!" [19:15] NOTA never wins anything fun :( Maybe next time [19:17] jbicha: I should have gotten around to expressing concern about non-core-devs on the DMB before the election happened. [19:17] jbicha: Oh well, now I just get to pressure tsimonq2 to up his game and get core-deve. [19:17] s/deve/dev/ [19:18] infinity: Man, I've been trying. [19:18] I did send a signed statement to the DMB saying I won't review applications for people seeking upload access I don't have. [19:18] But I totally get your point. [19:19] jbicha: https://launchpad.net/~developer-membership-board/+members Look right to you before I close the bug? [19:19] tsimonq2: Yeah, and I appreciate that you did that, but that also makes it harder to meet quorum for core-dev applications if you have to recuse yourself from those. [19:19] infinity: True. [19:19] infinity: I asked about that issue a year ago https://lists.ubuntu.com/archives/ubuntu-devel-discuss/2017-May/thread.html [19:19] tsimonq2: (plus, you still have the keys to various kingdoms, signed statement or no) [19:21] infinity: yes that looks fine. Feel free to close the bug [19:22] jbicha: You guys might want to discuss voluntarily mangling some of your terms, if the intent is to rotate slots with coninuity (US senate style). [19:22] jbicha: Since there are 5 expiring at the same time, and 2 in the other batch. A bit weird. [19:22] jbicha: And the 2019 batch should, ideally, expire in May, not Sep. [19:23] infinity: Indeed, but by being voted onto the DMB, enough Ubuntu Developers trust me with those keys to not do anything malicious; I appreciate that. But I'll keep my word and not press buttons I shouldn't be pressing. Feel free to remove me from whatever team you want if I break that. [19:24] infinity: And yes, pressure me to become a Core Developer. ;) [19:24] tsimonq2: Yeah, I'm not actually concerned, it's just a principle thing. So, let's get you on your way to core-dev, so when I suggest that DMB members must be core-devs, the current DMB already conforms. :P [19:26] infinity: we'll just nominate you to sit on the DMB then if we can't find enough enthusiastic core dev volunteers to fill the board :) [19:27] jbicha: Tried that. I sucked at caring enough to attend meetings and stepped down. [19:28] Though, a big part of that was the calendar being perpetually wrong during my entire tenure. [19:28] infinity: Alright. I did do an sbuild merge, slang2 merge, brltty merge, and a transmission sync in the past day or so. If you have more things to do, feel free to throw them at me, otherwise I'll stick to the strategy of stealing merges (with consent) by scanning merges.u.c for good candidates. [19:28] And, hilariously, fixed shortly after, but without removing me from the invite list. [19:29] tsimonq2: Talk to your fellow DMB members for their perspective, but I prefer to see bugfixes over merges. Merges don't often demonstrate any real understanding of complexity (I mean, except when they do, but the complex ones are rare, most are so mechanically simple that MoM can spit out the result) [19:29] infinity: Good point. [19:30] infinity: Can you please press buttons so I can be subscribed to the DMB mailing list? [19:31] (I submitted a subscription request.) [19:31] jbicha: In fact, I appear to still be on the DMB meeting invite list. Please clean that up. ;) [19:31] infinity: more details please :) [19:32] jbicha: Fridge calendar DMB meeting. List of invites does not match current board. [19:33] tsimonq2: I don't moderate that list. At least, I don't think I do. If I do, I do it really poorly. [19:34] infinity: The TB moderates it. :P [19:34] Yeah, that's not how mailman works. [19:34] infinity: sil2100 owns that calendar appt [19:34] ¯\_(ツ)_/¯ [19:35] infinity: The footer of lists.u.c/developer-membership-board disagrees. [19:35] tsimonq2: Yes, that just means that's who admin notifications get sent to. [19:35] tsimonq2: Still need the admin password to actually get in and do stuff. And I don't think I have that one. That's all I'm saying. :P [19:35] infinity: Ah, right. [19:43] infinity: are you able to moderate my u-devel-announce email re: DMB Election Results ? [19:43] jbicha: That, I can do. [19:44] jbicha: And done. === Elimin8r is now known as Elimin8er [23:28] #ubuntu-flavors is now a thing, fyi. [23:30] To what end? [23:30] Unit193: Hm? [23:30] It was in response to Kev's ML thread. [23:32] to what end? we need moar *buntu chans, obv [23:32] hehe [23:32] more channels to <3 each other in :) [23:32] ^ [23:32] And really, that's the point of this. [23:32] So flavors can really start to communicate better. [23:33] So, no purpose..? [23:33] Unit193: :P [23:33] what's the point of you? [23:33] the '3' looks pretty pointy... [23:33] what's the point of any of this? [23:33] * wxl cries [23:33] oh [23:33] hah [23:40] Also what tsimonq2 said a second before I did was more the answer I was looking for, "A place to check on issues that may affect multiple flavors" what the goal of the channel was. Or, if it's some cross flavor effort for the new ubiquity re-write, or calamares, or what. [23:41] Just in general, a place for coordinated albeit informal discussion among flavors that doesn't fit here or in -release, I think. [23:42] Unit193: Anyway, can I consider this checking the "let the IRCC know" box? :) [23:45] ACL looks about right, no logbot so no entry message. So I'd say good enough. [23:46] Alright.