=== est31 is now known as est [05:22] Uhm... guys... sandbox environment is down... any news? https://qastaging.launchpad.net [09:13] Why is there no Packages file here? Index of /aims/aims-desktop/ubuntu/dists/trusty/main/binary-amd64 [09:13] http://haetae.canonical.com/aims/aims-desktop/ubuntu/dists/trusty/main/binary-amd64/ [09:40] pipedream: Only compressed indexes are provided. [09:41] There were always packages files, no? Our debmirror now started failing on this, with this PPA, not others which still have a Packages file uncompressed [09:41] is this a new development? [09:41] Ubuntu and Debian's official archives have been like this for perhaps a decade. PPAs only dropped the uncompressed variants recently. [09:42] Did you somehow configure debmirror to only use the uncompressed ones? [09:43] will check [10:23] Also, though it doesn't make a difference for this, don't use http://haetae.canonical.com/, that's an implementation detail (and I'm a little surprised it works). Use http://ppa.launchpad.net/ [11:04] cjwatson: our proxy was caching the Release file, so debmirror still thought there was a Packages file to download. [11:04] cjwatson: we hijack ppa.launchpad.net dns on campus so we can run debmirror and redirect people to local cache [11:04] then the debmirror uses haetae (since that hasn't changed in forever) [11:08] ah, I see (where forever = 2012, but OK) [11:08] pipedream: still, the uncompressed Packages is expected to still be in the Release file [11:08] pipedream: apt uses that as a sanity check after decompression [11:09] pipedream: what version of debmirror are you using? [11:10] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=375381 would suggest that this can be caused by leftover cruft in .temp sometimes [11:10] Debian bug 375381 in debmirror "fails after a disk full saying since there is no Packages file in ftp.debian.org" [Normal,Fixed] [11:44] 1:2.16ubuntu1 [11:45] 42% /srv [11:50] I wasn't specifically asserting that your disk was full; any debmirror failure that leaves it in a state with weird cruft in .temp could have this kind of effect. [11:52] yes, the real reason was a corrupted cached Releases file. I think it mentioned only Packages not Packages.gz. That seems to have fixed itself now. === morphis_ is now known as morphis === zequence_ is now known as zequence === blr_ is now known as blr [20:05] Is the owner of popcon.ubuntu.com here, or is that a different service? It's failing to upload reports for me with: File "/srv/popcon.ubuntu.com/www/popcon-submit.cgi", line 30, in <module> mkdirs(hashDir,0755) NameError: name 'hashDir' is not defined [20:19] Is building a snap supposed to work on launchpad? I tried my first. Fails on unknown-command-assemble. [20:19] I know this is bleeding edge and I should expect blood. [20:58] qengho: it's complicated due to all the snapcraft compatibility breaks. [20:58] qengho: I think we need to update it for some more recent snapcraft changes. [20:58] Faux: Not us, I'm afraid. I don't remember who runs that. #canonical-sysadmin might have some way to figure that out. [20:59] wgrant: I made some suggestions about that in a recent snappy-internal thread and apparently you had agreed a plan with Sergio? [20:59] Okay. I might just bug-report the popcon package and let them figure it out. Cheers. [21:00] cjwatson: Ish. [21:00] But I see there's another incompatible change in the pipeline this morning. [21:00] So I might wait. [21:02] My suggestion was to make bare "snapcraft" without args work again, and then that would be usable with both 1.x and current 2.x. [22:15] hello, please take a look at https://answers.launchpad.net/launchpad/+question/286299 === philipballew is now known as Guest44851