[10:05] <xnox> cjwatson: partman-auto, the reuse recipe is giving me a fizzy fit. It can't mount the partition in read-only mode, because recovery required on readonly filesystem but white access unavailable, cannot proceed.
[10:06] <xnox> option one: remove the guards to blockdev --setro/--setrw
[10:06] <xnox> option two: use os-prober instead of manually hacking up installation detection
[10:06] <cjwatson> Consider using grub-mount if it's available
[10:07] <cjwatson> True guaranteed read-only mount
[10:07] <xnox> ok, let me try that. thanks.
[10:07] <cjwatson> os-prober switched to that when available a while back, so you could compare with its code
[10:07] <cjwatson> (So we need it to work regardless)
[10:07] <xnox> it's bug 1050562 and bug 1056571
[10:07] <ubot2> Launchpad bug 1050562 in ubiquity "Upgrading Ubuntu 12.04.1 from iso isn't given as an option" [High,Confirmed] https://launchpad.net/bugs/1050562
[10:07] <ubot2> Launchpad bug 1056571 in ubiquity "Missing "reinstall" option when installing same version of Ubuntu" [High,Triaged] https://launchpad.net/bugs/1056571
[10:07] <cjwatson> Ah
[10:07] <cjwatson> Makes sense
[10:14] <xnox> whichever way I look at it, os-prober is superior to 15reuse recipe semantics.
[10:18] <cjwatson> Performance might be an issue; os-prober + linux-boot-prober can be fairly slow
[10:18] <cjwatson> And it's very noisy in logs
[10:18] <xnox> =( true
[10:18] <cjwatson> I think if it were me I'd reimplement just the small bit you need
[10:19] <xnox> and we really don't want to run os-prober twice
[10:19] <cjwatson> Indeed
[11:08] <xnox> hmmm.... reuse yet I am still asked user setup
[11:08] <xnox> and lang and keyboard
[11:13] <psivaa> We are seeing bug 1058047 in our automated/manual testing,
[11:13] <ubot2> Launchpad bug 1058047 in ubiquity "No desktop shortcut, oem-config, oem-config-gtk in oem installation" [Undecided,New] https://launchpad.net/bugs/1058047
[11:14] <psivaa> in the process of attaching logs now,
[11:21] <cjwatson> psivaa: amd64?
[11:21] <psivaa> cjwatson, yes
[11:21] <cjwatson> psivaa: yeah, don't bother, the dailies broke today
[11:22] <cjwatson> it'll be sorted out soon
[11:22] <psivaa> cjwatson, ok, thanks, i hear due to version mismatches?
[11:23] <cjwatson> I'm not totally sure right now
[11:24] <cjwatson> I mean, yes, there's clearly a mismatch in the archive, but the new amd64 build should have been published a little while back
[11:24] <cjwatson> So I'm trying to make sure that this is just some kind of "ordinary" lag rather than a publisher bug
[11:25] <psivaa> cjwatson, ack
[11:26] <cjwatson> ah, possibly just slightly unfortunate timing
[11:27]  * cjwatson pokes the archive reports, being impatient
[11:28] <cjwatson> yeah, there we go, it's just armhf struggling now
[11:29] <cjwatson> I'll do a new daily build
[11:50] <cjwatson> ev: another wubi commit for you when you get a moment (r272)
[11:50] <ev> will do after lunch
[13:27] <xnox> ev: ubiquity will allow upgrading "devel release -> final" & "reinstall final" correctly now
[13:28] <ev> did it not before?
[13:28] <xnox> not the first bit, because it couldn't compare "Ubuntu quantal" with "Ubuntu 12.10"
[13:29] <xnox> ev: and because partman-auto was a bit broken it wasn't offering any upgrade options during quantal.... not sure if precise CDs do.
[13:29] <xnox> there is no manual - install (upgrade) test
[13:29] <xnox> with cd
[13:30] <ev> nice