| === liushuyu1 is now known as liushuyu | ||
| perkelix | howdy! can anybody think of a POSIX-compliant way of doing the same? https://paste.debian.net/hidden/5f0b0633/ | 06:02 |
|---|---|---|
| RikMills | bdmurray: did I see you say somewhere that you were in charge of MoM? if so, it seems to be down | 06:51 |
| === pushkarnk1 is now known as pushkarnk | ||
| === pushkarnk1 is now known as pushkarnk | ||
| === pushkarnk1 is now known as pushkarnk | ||
| Skia | RikMills: it's back | 08:28 |
| RikMills | \o/ | 08:28 |
| === pushkarnk1 is now known as pushkarnk | ||
| === pushkarnk1 is now known as pushkarnk | ||
| cpete | kanashiro: I'm trying to understand why ruby-sdbm is FTBFS only on i386. It looks like some circular dependency on libruby and ruby-sdbm? Although I don't really see any reason why it would be i386 specific. | 16:34 |
| kanashiro | cpete src:ruby3.3 which provides libruby3.3 does not build on i386, so this dependency is not satisfiable there | 18:59 |
| kanashiro | there are other packages involved in the ruby transition FTBFSing on i386 for the same reason | 19:00 |
| kanashiro | I am not sure what to do there, remove those package from i386 (?) | 19:00 |
| kanashiro | vorlon any thoughts on the above? ^ | 19:11 |
| ginggs | kanashiro: do you have the correct package name there? $ rmadison libruby3.3 | 19:40 |
| ginggs | libruby3.3 | 3.3.4-2ubuntu3 | oracular-proposed | amd64, arm64, armhf, i386, ppc64el, riscv64, s390x | 19:41 |
| ginggs | also $ rmadison libruby | 19:43 |
| ginggs | libruby | 1:3.3~ubuntu3 | oracular-proposed | amd64, arm64, armhf, i386, ppc64el, riscv64, s390x | 19:43 |
| kanashiro | ginggs I thought ruby3.3 was not built on i386, but it seems it did, I need to break a circular dependency on i368 then | 19:51 |
| ginggs | kanashiro: thanks! | 21:48 |
| sergiodj | I need to disable the inclusion of package metadata on some binaries built by a package. is unsetting ELF_PACKAGE_METADATA the recommended way to do this? | 22:26 |
| mwhudson | isn't there a DEB_BUILD_OPTIONS thingie for that | 22:53 |
| sergiodj | that's what I'm trying to find, but in this specific case the package is invoking gcc by hand, and from what I've read ELF_PACKAGE_METADATA is what influences this scenario | 22:54 |
| mwhudson | DEB_BUILD_OPTIONS=qa=-elfpackagemetadata i think | 22:55 |
| sergiodj | yeah, I don't want to mess with DEB_BUILD_OPTIONS inside d/rules | 22:57 |
| sergiodj | DEB_BUILD_MAINT_OPTIONS, though… | 22:57 |
| mwhudson | well DEB_MAINT_BUILD_OPTIONS then :-) | 22:58 |
| mwhudson | although i'm not sure if it works quite the same tbh | 22:58 |
| sergiodj | but let see what this option does under the hood | 22:58 |
| sergiodj | https://git.launchpad.net/ubuntu/+source/dpkg/tree/scripts/Dpkg/Vendor/Ubuntu.pm#n177 | 22:59 |
| sergiodj | I think it does exactly what I intended to do manually | 22:59 |
| sergiodj | which is good. let me try unsetting the variable, then | 23:00 |
| sergiodj | thanks, mwhudson :) | 23:00 |
| cpete | kanashiro: thanks for the info! | 23:52 |
| kanashiro | cpete now ruby-sdbm was built on i386, just need to wait to be published | 23:58 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!