[01:23] <phill_> Hi, Having issues with preseed, and kernel opts. Specifically, ath9k device failing to auth (reason 15) on 4 way handshake. it does configure with wpa_supplicant after install though... anyone seen this before? Ubuntu 14.04 LTS
[01:55] <CarlFK> phill_: you are trying to use wifi to do the install?
[16:23] <jamin> cjwatson: https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1296883
[16:25] <jamin> cjwatson: with that patch (on top of the others) oem-config should finally be fully functional.  Though I'm still experiencing issues if it don't force it to avoid using a framebuffer in my virtual machines
[16:34] <cjwatson> is this asserting that somehow switching to chvt might bring /var/lib/oem-config/run into existence?
[18:09] <jamin> cjwatson: no, without it, a ubuntu-server OEM install ends it's boot on vt7
[18:10] <jamin> as the report indicates, users then think the boot hung as they are expecting a login prompt
[18:10] <jamin> this prompt is in fact waiting on VTs 1-6
[18:11] <jamin> assuming one finds this (as I did) and then runs "oem-config-prepare" *and* has dealt with all the other oem-config issues, then reboots, they'll be prompted for the firstboot config
[18:11] <jamin> all subsequent boots work fine.
[18:12] <jamin> the entire issue here is when oem-config is present, but is *not* going to run... in *that* situation the boot ends on VT7 when in fact it should be switched to VT1 (which is what the patch does)
[18:14] <jamin> sure the presence check for /var/lib/oem-config/run is done twice, but I don't think that's a major issue, I guess the patch could be rewritten, but the essence is that if oem-config is not going to run, the system needs to change to vt1
[18:54] <jamin> cjwatson: any chance of seeing these oem-config fixes actually make it into the LTS?
[20:10] <cjwatson> jamin: yeah, it's just confusing in a delicate area, I'd prefer to see the patch a bit clearer
[20:10] <cjwatson> jamin: I think we can backport the lot of this once it's confirmed to work in utopic, yes