| 23:19 | leonardus | left the channel |
| 23:24 | leonardus | joined the channel |
| 23:24 | leonardus | left the channel |
| 23:24 | leonardus | joined the channel |
| 23:25 | leonardus | left the channel |
| 23:26 | leonardus | joined the channel |
| 01:15 | RexOrCine | changed nick to: RexOrCine|away
|
| 02:16 | intrac | left the channel |
| 02:18 | intrac | joined the channel |
| 07:00 | Bertl_zZ | changed nick to: Bertl
|
| 07:01 | Bertl | morning folks!
|
| 08:47 | sebix | joined the channel |
| 08:47 | sebix | left the channel |
| 08:47 | sebix | joined the channel |
| 10:03 | Bertl | off for now ... bbl
|
| 10:03 | Bertl | changed nick to: Bertl_oO
|
| 12:35 | RexOrCine|away | changed nick to: RexOrCine
|
| 14:53 | BAndiT1983|away | changed nick to: BAndiT1983
|
| 14:57 | sebix | left the channel |
| 14:59 | vup | BAndiT1983: btw i think i found the bug, it seems to be a bug in qemu: https://bugs.launchpad.net/qemu/+bug/1805913
|
| 14:59 | vup | (the bug with the broken cmake build)
|
| 15:00 | BAndiT1983 | hi vup, just skimmed through the thread, is there some solution?
|
| 15:02 | vup | not sure yet
|
| 15:02 | vup | (reading through the thread and the linked linux kernel mailing list thread just this moment)
|
| 15:03 | vup | so the reason this breaks in the CI and not locally for anuejn and me is, that we both use btrfs and it seems this bug is only hit on ext4
|
| 15:05 | vup | i think the "easiest" solution is running the whole thing in a loopback mounted temporary filesystem, that is not ext4
|
| 15:06 | vup | as this is a kernel bug (or atleast something needs to be changed in the linux kernel for qemu to fix the bug) i don't think we can easily work around this
|
| 15:11 | Bertl_oO | sounds interesting ...
|
| 15:14 | vup | an alternative workaround would be to use qemu in system mode and not user mode
|
| 15:14 | vup | or of course cross compile the control daemon
|
| 15:16 | vup | anyways, that was definitely a fun bug to track down ;)
|
| 15:16 | Bertl_oO | well, cross compilation is probably the best option in a long run, but it is interesting to see a (host) filesystem related issue in qemu
|
| 15:16 | Bertl_oO | yep, I'm sure it was ...
|
| 15:18 | vup | yes i actually want to switch to cross compilation anyways in the long run, so maybe that's now the time to do it
|
| 15:19 | Bertl_oO | btw, had a similar 'fun' issue with DNSSEC recently
|
| 15:20 | Bertl_oO | where some urls could be resolved but others not, like for example savannah.gnu.org
|
| 15:20 | vup | not at all, or only sometimes?
|
| 15:21 | Bertl_oO | the reason was a mix of systemd-resolvd and the nameservices for gnu.org
|
| 15:21 | vup | ah yes, i think i actually saw some people complaining about that
|
| 15:22 | Bertl_oO | recent arch update on the camera, after that, pacman fails to retrieve the databases
|
| 15:23 | Bertl_oO | error message: mirror.archlinuxarm.org not found :)
|
| 15:23 | Bertl_oO | go figure :)
|
| 15:24 | vup | fun
|
| 15:26 | vup | but also sometimes the archlinuxarm stuff is just down or really slow, we have been hitting that in the CI a lot
|
| 15:26 | BAndiT1983 | changed nick to: BAndiT1983|away
|
| 15:32 | Bertl_oO | yep, that's what I assumed first ... but it worked just fine on the old image :)
|
| 15:35 | vup | I see
|
| 16:46 | BAndiT1983|away | changed nick to: BAndiT1983
|
| 17:18 | shebin_joseph | joined the channel |
| 17:34 | shebin_joseph | left the channel |
| 17:37 | shebin_joseph | joined the channel |
| 20:32 | BAndiT1983 | changed nick to: BAndiT1983|away
|
| 21:32 | shebin_joseph | left the channel |
| 22:06 | RexOrCine | changed nick to: RexOrCine|away
|