=== med_ is now known as Guest82506 === Guest82506 is now known as medberry === medberry is now known as med_ [19:03] Good day. I recently approached completion of retiring my Ubuntu membership, because I haven't been active in Ubuntu. As it turns out, I end up using some of the rights and privileges associated with membership for lots of things, so rather than completing the process of going away, I should probably fix the issue of not being involved. [19:04] Could someone point me at any recent documentation on becoming MOTU, some of the tasks that are done at this point in the cycle these days? [19:59] persia: This channel tends to be a bit quiet, but have you seen the wiki on it, mailing list, or link in the topic? [20:01] Unit193: I wrote https://wiki.ubuntu.com/MOTU/Contributing : my concern is that I did that so long ago that things have changed. I remember many discussions about the right way to do things as I left, and it seemed that my opinion was not still consensus then. If those procedures still work, it is relatively easy for me to chase some merges, NBS, and FTBFS this weekend. [20:02] Haha, well then. Yep you win. :) [20:03] My main goal of checking in here is to make sure that submitting debdiff patches is a sensible path to regaining membership. [20:05] As far as I know debdiffs and LP test builds are both useful, depends on the sponsor. [20:05] * Unit193 is not a MOTU. [20:05] Unit193: Also, thanks a lot for prodding me last week: I hadn't realised my half-and-half status, and I would have continued to not take a decision to either be involved or not be involved in Ubuntu probably for several years without your help. [20:06] persia: Heh, no problem at all. Thanks for taking it well and am very glad it's gotten you off the fence. [20:08] My only complaint is how long I was in limbo: it's my own fault, and I presume you're busy, but faster prods when people change status is likely to cause more useful results: my first reaction was simply that I didn't understand how I got there or what to do. No idea if you can fix that, or what tools help make the decision to prod, but I figure I'll share my feedback. [20:35] persia: In this case you were in the cloaked group but no longer in ubuntumembers. In my defence, I was just elected last month or so, but will certainly keep that in mind. [20:35] Unit193: Yes, that is the half-life I meant. Consider yourself entirely defended: as mentioned above, I appreciated the prod: I just wanted to give feedback in hopes of helping you prod even better in the future :) [20:36] \o/ [20:39] I'm currently a packageset uploader, and thinking about going for MOTU this cycle. [20:41] I don't know if it still applies, but near the end of my most active time, a packageset uploader was usually a flavour developer, and tended to be more recognised than MOTU (who mostly clean up everything not owned by someone else) [20:42] My interest in MOTU is mostly because I don't have attachment to a packageset or flavour, and so would just be exchanging validation/testing for membership. [20:42] That fits, though MOTU gives more access since my packageset is all in universe. [20:42] "universe" was really supposed to have gone away by now. Oh well [21:07] persia: getting debdiffs for FTBFS and merges into the sponsorship queue is still the right way to do things. [21:08] To my knowledge not much has changed from that perspective, though I've only been around since 2011. [21:15] teward, yt? [21:16] for testing my stuff for the backports... is it enough to make sure the sbuilds work for both the library and the module [21:16] because actually testing the pam module is quite a bit of work on a chroot [21:16] but its been tested extensively on both precise and trusty actual [21:17] * karstensrage asks teward to channel the backporter mindset [21:22] * teward was pinged [21:22] * teward checks scrollback [21:22] karstensrage: (1) verify the packages build [21:22] (2) verify the packages install [21:22] (3) verify the packages actually *run* as one would expect [21:23] (4) make sure that any interim supported releases (if X -> T, then W also needs the backport too) have steps 1, 2, and 3 work for them [21:23] karstensrage: this is why I have a bunch of VMs [21:23] karstensrage: I actually have a "Testing" VM for each supported release - Trusty, Wily, and Xenial currently [21:24] not precise obviously but meh [21:24] teward, i also do that [21:24] karstensrage: there are cases, as micahg had told me in the past, where a package that doesn't ship any executable binaries and just libraries (swig3.0 for example), just gets an install test and not a run - though I ran a test with the 'znc' source to make sure swig3.0 worked heh [21:25] and for at least trusty ive done 1, 2, and 3 [21:25] but as I'm not on the backports team, I can only say what I've seen in the past with my backport reqa. [21:25] reqs. [21:27] karstensrage: there may be more evil things for pam tests - that's something MOTUs might be able to comment on - I usually don't touch pam if i can avoid it ;) [21:27] what do you mean evil? [21:37] karstensrage: they may want other tests [21:49] well if there was something evil, that would apply to everything i assume [21:49] not backports specifically [21:50] * teward shrugs [21:51] I think pam plugins can just be a little weird is all [21:51] rbasak: thanks. We overlapped, so if you report no real changes, then I will believe no real changes since I left. Thank you. [21:53] ohi persia :) [21:56] well id love a security review [21:57] but thats outside the purview of backports [21:57] yep [21:57] oops that's to another channel sorry :) [21:57] * teward is everywhere [21:57] and i feel like the cognitive load for pam is too high to be coupled with backports or packaging [21:58] i liked how xenial did it just pulling from debian/unstable no questions asked [21:58] but i understand the problem [21:58] * Unit193 watches a tumbleweed blow past. [21:58] how are you supposed to trust me [21:59] but also you cant scale testing every single little thing that comes along esp. with volunteers that are overworked already [21:59] so between scilla and charybdis [22:05] hi Unit193 [22:06] Hello good sir. [22:11] Unit193: you wouldn't happen to be an expert in bash completion scripts, right? [22:58] Logan: No such luck. [22:59] damn [22:59] was worth a shot :P