[00:38] seb128: OK. Let me check if I can borrow the problematic hardware to see what is going on. [02:45] root@disco:~# for x in `seq 1000`; do ./int64test nan; done | sort | uniq -c [02:45] 504 result -9223372036854775808 status 0 [02:45] 496 result 0 status 1 [02:45] icu in disco-proposed :( [02:46] mind you icu in disco release does the same with NaN [02:48] and of course it's completely consistent if you turn aslr off :( [08:42] tkamppeter: https://bugs.launchpad.net/ubuntu/+source/hplip/+bug/1804772 isn't that because hplip is installing stuff from the web? [08:42] Launchpad bug 1804772 in hplip (Ubuntu) "package python3-gdbm:amd64 3.6.7-1~18.04 failed to install/upgrade: afhængighedsproblemer - efterlader den ukonfigureret" [Undecided,New] [09:41] Hi [09:43] dpkg -i | grep linux-image [09:43] returns many kernel version, but I need to test 4.4 version and booting with Shift doesn't list it. How can I make kernel 4.4 available for ubuntu advanced options during boot? [09:44] according dpkg -i I've 4.4 already installed [09:44] `sudo update-grub` (relatively harmless) is a good way to see what will be on the boot menu. [09:47] Faux: thank you [09:48] There is not 4.4 version of kernel listed from update-grub [09:48] How can I add it? [09:49] I wouldn't try and do it by hand. Find a repo that has a 4.4 kernel in that's compatible with your release, and install it from there. (This isn't really a good channel.) [12:25] jbicha: around? [12:25] yes [12:25] jbicha: any particular reason you synced ndctl ffrom debian? [12:25] because I didn't realize how broken ndctl was :/ [12:25] it introduced some packaging regressions compared to what we had in bionic and cosmic [12:26] yes LP: #1804725 [12:26] Launchpad bug 1804725 in ndctl (Ubuntu) "Fix up ndctl upgrade from cosmic to disco" [High,New] https://launchpad.net/bugs/1804725 [12:27] we already had to NMU it twice in Debian [12:27] I was kind of in touch with the debian maintainer(s) about it, they pinged me a few weeks ago asking if they could upload ubuntu's package into debian [12:27] Adam Borowski [12:27] that's the guy that did the NMU [12:28] https://tracker.debian.org/pkg/ndctl [12:28] if I (or someone) is going to change the binary package names, I want to confirm with the maintainer (Breno) first [12:28] I asked to co-maintain, he said great, then disappeared, and I didn't ping him again either [12:29] I jsut emailed him today asking to at least create a salsa repo [12:29] Adam is not the maintainer. You should ask those questions to Breno [12:30] he later found breno's package, I think it was in their NEW queue [12:30] anyway, the ubuntu changes were lost, we can start the work to re-add them, and if debian gets the package in salsa, we can coordinate better [12:30] Adam & I have sort of been hijacking ndctl since the package was basically unusable as originally uploaded [12:32] I need to rejoin, 1s, xchat is behaving erratically [12:32] better [12:33] jbicha: I'm having some trouble following why we aren't just uploading a +really back into Ubuntu to revert your sync. [12:33] rbasak: we don't need a really. We just need to rebase on top [12:34] ahasenack: do you want to fix up the disco package or do you want me to do it? [12:34] Are you planning on doing that soon? [12:34] jbicha: how would you do it? [12:35] I would go through the issues I listed in bug 1804725. Some of the Debian changes are probably ok [12:35] bug 1804725 in ndctl (Ubuntu) "Fix up ndctl upgrade from cosmic to disco" [High,New] https://launchpad.net/bugs/1804725 [12:36] go for it then [12:36] ok [12:36] also note that in this new upstream version, that new service, monitor-something, won't start, but at least dpkg doesn't fail [12:38] I added that and other info to the bug [12:40] thank you. I can't do anything about the monitor service but I can fix up the packaging [12:40] it's better already, I remember when I first tried to update to this version (upstream requested), I noted that it would fail hard [12:41] that bit they fixed at least (upstream) [12:42] https://github.com/pmem/ndctl/commit/52a4d38e7b5f56d85712f6bb29758521604a2b3e [12:42] ok [12:43] jbicha: some suggestions on detecting this: [12:45] Before sync, check the Ubuntu changelog for the common ancestor. If there is none, then it needs further investigation. [12:46] And speak to the TIL? [13:41] jbicha: libqalculate ftbfs on arm64 === karimsye is now known as karimsye-afk === karimsye-afk is now known as karimsye === JanC_ is now known as JanC === blahdeblah_ is now known as blahdeblah