| === JanC is now known as Guest50625 | ||
| === JanC_ is now known as JanC | ||
| mpt | Spam comments in merge proposals now? Haven’t seen that before. | 08:06 |
|---|---|---|
| mpt | And bug commenters who don’t realize that quoting an entire spam message is exactly as bad as sending the original spam \o/ | 08:12 |
| pipedream | I thought this was our corrupt debmirror cache on campus: E: Failed to fetch http://ppa.launchpad.net/aims/sagemath/ubuntu/pool/main/s/sagemath-upstream-binary/sagemath-upstream-binary_7.2~aimsppa1_amd64.deb Size mismatch | 11:40 |
| pipedream | However, I am getting independent reports from other users in other countries of the same | 11:41 |
| pipedream | the deb should be 600M, the downloaded deb is something like 200M | 11:41 |
| pipedream | https://launchpad.net/~aims/+archive/ubuntu/sagemath/+packages says 600M | 11:41 |
| pipedream | (trusty 64bit) | 11:42 |
| pipedream | the package was copied (use existing binaries) from ppa:aims/sagemath-dev (where it did not exhibit this problem) | 11:42 |
| pipedream | any idea? | 11:42 |
| pipedream | hmm, repeated debmirror downloads and apt installs got bad package. But firefox coudl download the deb manually from direct link | 12:03 |
| pipedream | that dpkg -i OK | 12:03 |
| pipedream | am deleting deb, and later copying in again | 12:04 |
| pipedream | on launchpad | 12:04 |
| wgrant | pipedream: Hm, there was a one-off issue last week which could have explained the file being truncated on disk, but if firefox was able to download it from ppa.launchpad.net successfully then that's not it. | 12:05 |
| wgrant | Can you confirm it was a successful full download from ppa.launchpad.net, not launchpadlibrarian.net? | 12:05 |
| pipedream | yea | 12:05 |
| pipedream | uhm | 12:05 |
| pipedream | https://launchpadlibrarian.net/261359749/sagemath-upstream-binary_7.2~aimsppa1_amd64.deb <- from firefox download history | 12:06 |
| wgrant | Ah | 12:06 |
| pipedream | looks like launchpadlibrarian | 12:06 |
| wgrant | So it may be the issue we saw last week. | 12:06 |
| wgrant | Yeah | 12:06 |
| wgrant | Delete + copy won't directly fix, unless you wait at least six hours between the delete and copy. | 12:06 |
| wgrant | (the file won't be removed from the pool for a few hours after the deletion) | 12:06 |
| pipedream | is there a better way? | 12:07 |
| pipedream | or should I just do that | 12:07 |
| wgrant | This is only the third file that we know to be affected. Have you checked if the other binaries for that package are OK? | 12:08 |
| pipedream | no | 12:08 |
| pipedream | don't have quick access to do that now | 12:08 |
| pipedream | do you want to have a look / investigate before I delete them? | 12:10 |
| wgrant | pipedream: Delete away. Unfortunately there's not much we can discover from the files on disk. | 12:12 |
| wgrant | We have a fix in progress to prevent it from happening again, though we don't totally understand the underlying cause. | 12:12 |
| pipedream | okey dokey | 12:13 |
| wgrant | We suspect an OpenStack swift overload condition resulting in truncated downloads, but it's a bit more complicated somehow. | 12:13 |
| pipedream | thanks! happy debugging | 12:19 |
| === bladernr` is now known as bladernr | ||
| === nacc_ is now known as nacc | ||
| === pavlushka is now known as kryten | ||
| === kryten is now known as Guest62041 | ||
| === Guest62041 is now known as pavlushka | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!