[17:12] <antarus_> cjwatson: qq for you or anyone else
[17:12] <antarus_> preseeds for d-i for precise
[17:12] <antarus_> we are preseeding apt-setup/use_mirror boolean false
[17:12] <antarus_> but apt-setup's 50mirror is still calling choose-mirror
[17:13] <antarus_> I'm curious if the type matters...should it be string false and not boolean false?
[17:13] <antarus_> in 50mirror, it definitely calls db_get apt-setup/use_mirror and checks for $RET = false
[17:14] <antarus_> it looks like in https://help.ubuntu.com/12.04/installation-guide/example-preseed.txt it is 'boolean false' there as well :/
[17:14] <antarus_> but for some reason choose-mirror is not being skipped for us
[17:17] <cjwatson> Type doesn't matter
[17:17] <cjwatson> Get an installer syslog with DEBCONF_DEBUG=developer as a boot parameter and I can have a look at that
[17:18] <antarus> cjwatson: yeah it looks similar to https://bugs.launchpad.net/ubuntu/+source/choose-mirror/+bug/600789
[17:18] <ubot2`> Ubuntu bug 600789 in choose-mirror (Ubuntu) ""d-i mirror/*" does not work on presseding" [Undecided,Confirmed]
[17:18] <antarus> we don't actually preseed the mirror
[17:18] <cjwatson> I'd rather look at a syslog independently
[17:19]  * antarus nods
[17:19] <cjwatson> I don't think that bug is related and it contains some misinformation by way of guesses
[17:19] <antarus> will DEBCONF_DEBUG tell me what RET contains in 50mirror? :)
[17:19] <cjwatson> Yes
[17:19] <antarus> ah
[17:19] <antarus> well that I will go that route next ;)
[17:19] <cjwatson> It's a trace of all debconf activity
[17:20] <cjwatson> Can take practice to read though, much like (say) strace - feel free to just hand me the log
[17:20] <antarus> does it log all the stuff to syslog?
[17:20] <antarus> or to the console?
[17:20] <cjwatson> syslog
[17:20] <antarus> excellent
[17:20] <cjwatson> note that it will include passwords, since it's analogous to strace and doesn't sanitise things
[17:21] <cjwatson> so don't use anything sensitive in the preseed file
[17:21] <antarus> I'll use my best judgement ;)
[18:51]  * antarus sighs at d-i
[20:27] <antarus> cjwatson: hrm, my pxe tool is apaprently downcasing the DEBCONF_DEBUG line to debconf_debug=developer, do you know if that will break that feature?
[20:47] <antarus> aha, fixed it
[21:07] <antarus> I hate d-i so much
[21:36] <antarus> Is it possible to force d-i to skip steps?
[21:52] <antarus> hrm