Log file Viewer

[13:56:09] <sumanah> !logs

[13:56:09] <pmxbot> http://kafka.dcpython.org/channel/pypa-dev

[13:56:40] <sumanah> EWDurbin: argh, permission denied

[13:56:53] <EWDurbin> permission denied for make purge?

[13:57:12] <sumanah> EWDurbin: yes: https://hastebin.com/yakusexama.cs

[13:58:14] <EWDurbin> sumanah: what are ownership and permissions settings on `warehouse/static`

[13:58:44] <EWDurbin> sumanah: best guess is that your docker daemon is running as root, so your user `sumanah` doesn't have permission to remove the files written by the containers

[13:59:30] <di_codes> the ‘database “tests” already exists’ thing can happen when the tests are running and abruptly killed in such a way that it doesn’t have time to clean up the test databases

[13:59:37] <di_codes> a `make initdb` should fix it

[13:59:54] <sumanah> interesting, root owns `dist/` but my user (sumanah) owns everything else

[14:00:33] <sumanah> `make initdb` failed:

[14:00:33] <sumanah> psql: could not connect to server: Connection refused

[14:00:34] <sumanah> Is the server running on host "db" (172.18.0.6) and accepting

[14:00:34] <sumanah> TCP/IP connections on port 5432?

[14:00:34] <sumanah> Makefile:144: recipe for target 'initdb' failed

[14:00:34] <sumanah> make: *** [initdb] Error 2

[14:01:00] <EWDurbin> yeah, docker-compose is spawning containers with docker. generally on Linux that process is running as root. so when it writes files back to the file system as the static container does those are owned by root.

[14:05:35] <sumanah> I ran `sudo make purge` which worked, and now a `make initdb` is working away happily, thanks di_codes & EWDurbin

[14:08:54] <sumanah> huh, now there's an elasticsearch timeout https://hastebin.com/mepudufezo.sql

[14:10:45] <di_codes> sumanah: you might need to increase the amount of memory allocated to docker, elasticsearch is kind of a hog (I give warehouse ~4GB locally)

[14:51:37] <tos9> What are the ramifications of https://github.com/pypa/virtualenv/blob/master/virtualenv.py#L1087

[15:28:11] <dstufft> fungi: I hadn't broken openstack recently enough, had to sneak it in somewhere :P

[15:28:43] <fungi> dstufft: no worries! as breakages go this was mild and easily resolved

[15:29:22] <fungi> was just more surprised by the apparent lack of announcement than anything

[15:29:47] <fungi> and as always, huge thanks for improving warehouse (to all of you)!

[15:32:55] <dstufft> fungi: yea, I'm real bad at communicating stuff :(

[15:33:11] <fungi> we all are

[15:33:12] <dstufft> I always forget to do it

[15:54:45] <fungi> probably a terrible idea, but what if the ui had some scary sort of indicator next to/in place of the gravatars on the collaborators pane indicating accounts with unverified primary addresses?

[15:55:08] <fungi> that way they might be spotted by others when adding those accounts to a project

[15:57:23] <fungi> though i suppose if this is just a transitional state on the way to making unverified accounts basically unusable (and stop letting projects add them as collaborators) then that's probably on the side of overengineering

[16:52:47] <agronholm> dstufft: if you're around now, would you mind adding Travis to the list of Github apps for pypa? thanks.