| xnox | juliank, postgresql-10-debversion sounds like a broken update, where did you see that? | 00:12 |
|---|---|---|
| juliank | xnox: postgresql-debversion/1.0.8-3 autopkgtest on !amd64 triggered by apt/1.6~alpha3 | 00:13 |
| juliank | in failure to setup testbed, of course | 00:13 |
| nacc | there are a bunch of those pg10 packages that are migrating right now | 00:14 |
| nacc | so it could be related to all of that | 00:14 |
| nacc | and the builders catching up, etc. | 00:14 |
| juliank | It's a bit strange that the autopkgtest failed because the package to be tested was not available, but I guess it can happen | 00:15 |
| nacc | juliank: rmadison says it should be available, but only in bionic-proposed | 00:15 |
| nacc | so it might be stuck | 00:15 |
| juliank | Right, the amd64 retried with all packages from proposed | 00:15 |
| nacc | juliank: http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#postgresql-10 is probably the blocker :/ | 00:16 |
| nacc | juliank: it's on my todo (and cpaelzer's) | 00:16 |
| juliank | But armhf also retried using all packages from proposed, but could not find it | 00:16 |
| juliank | Anyway, I saw that with the last transition IIRC, so I'm not worrying too much | 00:17 |
| nacc | yeah | 00:17 |
| cpaelzer | juliank: thanks, the upload looks good, I'll push it into the git on Monday to track it properly | 05:20 |
| cpaelzer | juliank: I'll do some regression checks along the way, but the change seems small and safe | 05:21 |
| cpaelzer | juliank: also while not upstream yet, the mailing list response so far seems good | 05:21 |
| cpaelzer | lets see what the weekend can do to these long queues ... | 05:29 |
| === infinity0 is now known as Guest39948 | ||
| === infinity0_ is now known as infinity0 | ||
| erle- | Why are the commands in /bin now shared objects instead of executable files? | 15:31 |
| === WieWoWasWozu is now known as hggdh | ||
| rbasak | erle-: I'm not sure what you mean. They are still executable files. | 15:33 |
| erle- | rbasak, but they are of time ELF shared object instead of ELF executable | 15:34 |
| erle- | *of type | 15:34 |
| rbasak | I'm pretty sure that's just a consequence of the tooling you're using to report on it. | 15:34 |
| erle- | rbasak, use «file /bin/cp» | 15:35 |
| erle- | you will see it is shared object in 17.10 | 15:35 |
| erle- | it was executable in 16.04 | 15:35 |
| rbasak | Yes. Like I said it's a consequence of the tooling you're using to report on it. | 15:35 |
| rbasak | Try copying an executable from 16.04 into a 17.10 system and then see what file says. | 15:36 |
| erle- | rbasak, I got an answer somewhere else | 15:39 |
| erle- | they are actually shared objects | 15:39 |
| erle- | point is that the linker can put them to random locations im memory | 15:39 |
| erle- | it's about address randomization | 15:39 |
| jbicha | erle-: https://codywu2010.wordpress.com/2014/11/29/about-elf-pie-pic-and-else/ | 15:40 |
| jbicha | https://wiki.ubuntu.com/Security/Features#pie | 15:40 |
| erle- | jbicha, thanks | 15:42 |
| === hggdh is now known as WieWoWasWozu | ||
| cjwatson | Some were PIE in 16.04 too | 15:53 |
| cjwatson | $ lsb_release -sr; file /usr/sbin/sshd | 15:53 |
| cjwatson | 16.04 | 15:53 |
| cjwatson | /usr/sbin/sshd: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=cf2da4087de886767f688f645786229131e4af0c, stripped | 15:53 |
| cjwatson | It's just got a bit more default in the toolchain | 15:54 |
| rbasak | erle-: oh, I see. I didn't realise that change would cause a different output from file. Thanks. | 16:16 |
| infinity | rbasak: PIE binaries are literally type DYN, so yeah, file(1) doesn't know what to do with them. | 16:17 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!