=== mborzeck1 is now known as mborzeck === mborzeck is now known as mborzecki [08:24] sergiodj: Hey! I see you updated the s390-tools package, do you plan to update s390-tools-signed with the same version as well, to resolve the cross-dependency? [12:10] sergiodj: the net-snmp empty archives, is that really just a packaging issue? [12:11] tkamppeter: could you have a look at the failing autopkg tests, triggered by cups? [14:03] Unpacking dash (0.5.11+git20200708+dd9ef66+really0.5.11+git20200708+dd9ef66-5ubuntu1) over (0.5.11+git20200708+dd9ef66+really0.5.10.2-0ubuntu1) ... [14:03] lol [14:13] 0_o [14:13] ouch [14:33] que [15:52] slyon: ah, I wasn't aware of s390-tools-signed! thanks for the heads up, I can certainly upload it with the same version :-) [15:52] sergiodj: thanks! [15:54] slyon: done; thanks :) [17:43] doko: I'm going to dig deeper into net-snmp's archive files & empty .text segments. it felt like we shouldn't be shipping the .a files anyway, but there's probably something else at play here since the net-snmp's .a files on (e.g.) focal have non-empty .text [17:46] slyon: it seems s390-tools-signed have to be adjusted to depend specifically on s390-tools-zkey with the same version. blah... if you're fine with it, I'll reupload s390-tools (version 2.15.1-0ubuntu5), and then reupload s390-tools-signed and adjust its B-D to s390-tools-zkey (= 2.15.1-0ubuntu5), that way everything will have the same version again [18:35] I'm looking at libgit2 that's blocked because of golang-gopkg-libgit2-git2go.v28 autopkgtest failures − that's a libgit2 version incompatibility that's solved by golang-gopkg-libgit2-git2go.v31, but I'm not sure of the correct way to tell autopkgtests to run the tests for v31 instead of v28, help appreciated [18:43] oSoMoN, golang-gopkg-libgit2-git2go.v28 should be removed? [18:46] that makes sense, can an archive admin help here, please? === ijohnson is now known as ijohnson|lunch === ijohnson|lunch is now known as ijohnson [23:49] doko: arguably we could install all of the other archive files from net-snmp but leave libnetsnmphelpers.a out, since that's the only .a that doesn't have a .text segment. I'm still unsure whether it makes sense to install them, though. [23:49] anyway, just wanted to follow up on the conversation [23:49] FWIW, libnetsnmphelpers.a has always had an empty .text, it seems