| xnox | jodh: there is a typo in the failed case. | 10:00 |
|---|---|---|
| xnox | jodh: the filename will not be defined :P | 10:00 |
| jodh | xnox: need a few more bytes of context - we're talking about lp:~xnox/upstart/fix-qemu-test presumably? | 10:02 |
| xnox | jodh: yes, mere re-reproposal is on the way. | 10:02 |
| xnox | jodh: https://code.launchpad.net/~xnox/upstart/fix-qemu-test/+merge/137517 | 10:02 |
| xnox | s/mere/the/ | 10:03 |
| jodh | xnox: ah! I'd missed that. Could you also therefore wrap the unlink in assert0() like: 'assert0 (unlink (filename))' | 10:04 |
| xnox | jodh: no, since it may not exist. | 10:05 |
| xnox | cause it's in the maybe-failed code path. | 10:05 |
| jodh | xnox: gah - agreed :) | 10:08 |
| jodh | xnox: btw - I raised bug 1085836 just now. It would be great to have this facility for Raring, particularly for the DEP-8 tests. We can discuss with stgraber later. | 10:09 |
| xnox | jodh: interesting. /me ponders if we can re-exec ourselfs under sudo for a test wrapped in NEEDS_ROOT | 10:13 |
| stgraber | xnox, jodh: one option would be to have a separate "root-check" target or similar which requires root permissions, those wouldn't be called at build time but could be called under sudo on the autopackagetest machine | 14:09 |
| jodh | stgraber: right. | 14:28 |
| xnox | jodh: https://launchpad.net/~xnox/+archive/scratch/+build/4031890 | 15:12 |
| xnox | jodh: stgraber: https://buildd.debian.org/status/fetch.php?pkg=upstart&arch=amd64&ver=1.6-3&stamp=1353265111 | 15:13 |
| xnox | Kernel: Linux 2.6.32-5-amd64 amd64 (x86_64) | 15:14 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!