[01:37] ddstreet: Hrm, does `backportpackage` work for you still? For me, I get: Invalid URL 'https:/incoming.debian.org/debian-buildd/pool/main/l/limnoria/limnoria_2021.11.9-1.dsc': No host supplied (if I forcefully downgrade just python3-ubuntutools it "fixes" the problem.) [01:43] I guess the single / there is the issue [01:44] That's what it echos back, not my input. [01:47] ew [01:49] I just rebuilt ubuntu-dev-tools to not have a versioned depends and held back python3-ubuntutools, so far so "good". :3 [01:58] crimes, crimes everywhere :) === genii is now known as genii-core === sem2peie- is now known as sem2peie [09:12] * ginggs pokes scipy and will look at the python-skbio FTBFS [09:51] Anyone here has any experience bootstrapping a language in a "new" architecture? picolisp FTBFS on s390x because it needs itself, and I'm idly wondering how to solve the problem [10:13] schopin: use debian's build to build ubuntu package; vendor it in a source package; upload to bileto ppa; rebuild again afresh; publish to ubuntu archive; rebuild in ubuntu archive proper again. [10:13] schopin: alternative is to cross-build the first one. [10:13] schopin: sometimes, debian/* has instructions for bootstrapping. as the first upload of the thing had to be bootstrapped by the person === cpaelzer_ is now known as cpaelzer [10:32] xnox: thanks! This seems too involved for me to get into it rn, but I put it all on an LP bug for future reference. [10:48] schopin: also for example it might be buildable with any other lisp compiler? like emacs or clisp? [10:48] schopin: but yeah, it takes time and gymnastics [10:48] I wouldn't know. I just stumbled onto the package during my openssl work, and noticed the s390x situation. [10:49] schopin: but in that case you can just carry on, it is not a regression. [10:49] schopin: it's in build-depwait state in both jammy-release & jammy-proposed, so your openssl transition will migrate regardless. [10:51] i think it might cross-compile just fine. [11:33] Question for archive gurus: we currently have vde2 and vdeplug4 that both build the package libvdeplug-dev (amongst other packages). The latter package has been added back in hirsute, and has a version number greater than the former. [11:34] Now, I tried to upload vde2 as no-change to rebuild against openssl3, but obviously I got errors because of said packages. Will the non-conflicting binaries be published anyway? [11:36] (the previous build has been carried over from groovy, AFAICT) === sem2peie- is now known as sem2peie [13:50] bryceh: are you talking to anyone specific about your apache2 SRU reverts? I see them in the queue, but they don't have bug references so need special handling/planning if that's intentional. [13:52] Normally we have a bug to track the fix for each regression (one bug for all series) so I'm not sure if it's intentional or not. [15:54] rbasak, I've worked with bdmurray on that, although he wasn't on irc when I uploaded so hadn't checked in with him [15:56] the bug # for the original issue, mentioned in the changelog entry, has been the main point of discussion. Let me know if I should file an additional bug on the revert itself. === genii-core is now known as genii === klebers_ is now known as klebers [20:52] schopin, are you looking to fix libnet-ssleay-perl installability issue? [20:52] this is blocking some packages builds for me [22:41] ffs what is going on here https://launchpad.net/~mwhudson/+archive/ubuntu/devirt/+build/22448228 [22:46] mwhudson: Would it surprise you if I told you that isn't even the weirdest thing I've seen today? [22:46] not really [22:46] Didn't think so. [22:46] but the same source built fine last week [22:46] smh [23:07] uh and now i can't debootstrap jammy? [23:11] ah maybe debootstrap works but the other parts of a buildd chroot don't install