[09:19] <cjwatson> ogra_: I think I see how to fix bug 1028905 (at least by filtering out extended partitions in list-devices as you suggest, never mind the issue of mount hanging) - are you already on it?
[09:19] <ubot2> Launchpad bug 1028905 in cdrom-detect "cdrom-detect in quantal omap4 hangs trying to look for install media on an extended partition" [High,Confirmed] https://launchpad.net/bugs/1028905
[09:20] <cjwatson> I've at least reassigned that to debian-installer-utils, which seems closer to a proper place
[09:21] <ogra_> cjwatson,  i was on it on and off butu was rather trying to hunt down the root cause with mount instead of just filtering, i can implement some filtering i guess
[09:21] <cjwatson> if you'd rather hunt down the root cause, I don't object :)
[09:21] <cjwatson> just looking for quick ways to kill -tracking bugs I guess
[09:22] <cjwatson> one reason to filter is that anything which implements tests such as "does list-devices partition return anything" shouldn't consider extended partitions
[09:22] <cjwatson> probably
[09:22]  * cjwatson goes to look at users
[09:22] <ogra_> well, i'll add a filter first then and open a task against mount we can leave open
[09:25] <cjwatson> make sure either that that task isn't targeted to quantal, or that it's a separate bug
[09:25] <cjwatson> since I don't think we want to *commit* to it for 12.10
[09:25] <cjwatson> right, I just checked all the users of list-devices *partition in upstream d-i
[09:26] <cjwatson> all of them are looking for something they can mount
[09:26] <cjwatson> so I'm pretty convinced it's correct to filter
[09:26] <cjwatson> this contributes slightly to bug 613510
[09:26] <ubot2> Launchpad bug 613510 in rescue "Rescue mode shows more partitions that those that are available as root partitions" [Low,Triaged] https://launchpad.net/bugs/613510
[09:27] <ogra_> oh, nice
[10:52] <pjotr> cjwatson: I'd like to request your attention for a bug in Ubiquity in Quantal, that causes confusion:
[10:52] <pjotr> https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1057452
[10:52] <ubot2> Launchpad bug 1057452 in ubiquity "[Quantal] Ubiquity creates confusion about which part of the disk space will go to Ubuntu" [Undecided,New]
[10:53] <pjotr> Can this be helped in time for Quantal final edition?
[10:55] <xnox> pjotr: that's a duplicate.
[10:55] <xnox> of bug 947107
[10:55] <ubot2> Launchpad bug 947107 in ubiquity "No partition labels in the resize widgets" [High,Triaged] https://launchpad.net/bugs/947107
[10:57] <xnox> I did some poking at it, and it's something nasty in gkt/cairo stack. I can get the resize widget to correct display: either space labels or the logo/titles.
[10:57] <xnox> but not both =/
[10:59] <pjotr> xnox: OK.... Well, maybe a provisional solution could be, that the explanatory text *above* the widgets, would mention clearly that the space *on the right side* will be the Ubuntu space.
[10:59] <pjotr> What do you think?
[10:59] <xnox> I think the bug needs to be fixed. Because the intended layout that is set is:
[11:00] <xnox> Logo (windows/ubuntu/other image), device (/dev/sda1), name (Ubuntu 12.04 / Windows 7) and then the size of the partition.
[11:00] <xnox> but some changes in gtk stack prevent displaying it correctly for some reason.
[11:01] <xnox> there is another bug open, that it is not accessible.
[11:01] <xnox> But i'm not sure which widgets to use to get a11y & screenreader to pick the changes up.
[11:02] <pjotr> xnox: the current text above the widgets is: "Allocate drive space by dragging the divider below:"
[11:02] <pjotr> maybe this could become: "Allocate drive space by dragging the divider below (Ubuntu is on the right):"
[11:03] <pjotr> What do you think?
[11:04] <xnox> pjotr: I don't deal with desing / user experience =))) Please read ubiquity design specification. http://goo.gl/Kokw5 And if you feel that needs changing, please open a separate bug with 'needs-design' tag.
[11:07] <pjotr> xnox: OK.... Do you have a name for me, of a developer who I can contact on that?
[11:08] <xnox> mpt is the current maintainer of the ubiquity design
[11:09] <mpt> pjotr, changing a string would break translations, and it is very late in the release cycle.
[11:09] <mpt> to do that.
[11:10] <pjotr> mpt: I know, I'm a member of the Dutch translation team myself... :-)
[11:10] <pjotr> nevertheless, I think this is pretty important
[11:11] <pjotr> creates confusion for beginners
[11:11] <mpt> yes, it is
[11:12] <pjotr> I can guarantee a full Dutch translation if the text string would be changed.... :P
[11:13] <mpt> xnox, I've found that it's unhelpful for bugs to be assigned to teams, because it gives the illusion of responsibility. Does the Canonical Foundations Team have a process for allocating bugs like 947107?
[11:14] <ogra_> cjwatson, http://paste.ubuntu.com/1230214/ how about something like that ?
[11:15] <ogra_> (not sure if we need to match other exotic extended partitions)
[11:15] <cjwatson> mpt: No
[11:15] <cjwatson> ogra_: egrep -q '^ID_PART_ENTRY_TYPE=0x(5|f|85)'
[11:15] <cjwatson> to match libparted
[11:15] <ogra_> ok
[11:15] <cjwatson> and how about 'if $match && ! device_info ...'
[11:16] <ogra_> moving it into the outer "if" ?
[11:16] <ogra_> k
[11:17] <cjwatson> yeah, I think
[11:22] <xnox> bug 947107
[11:22] <ubot2> Launchpad bug 947107 in ubiquity "No partition labels in the resize widgets" [High,Triaged] https://launchpad.net/bugs/947107
[11:23] <xnox> mpt: well, it is part of the quantal release tracking bugs and it's as best as it gets. I did spend time on it, but it's not fixed yet.... =/
[14:04] <cjwatson> Can anyone else reproduce bug 1049215?  It's a bit of a mystery to me, but it keeps showing up in ISO testing.
[14:04] <ubot2> Launchpad bug 1049215 in ubiquity "ubiquity-bluetooth-agent crashed with ImportError in /usr/lib/python3/dist-packages/gi/__init__.py: could not import gobject (error was: EOFError('EOF read where not expected',))" [Medium,Confirmed] https://launchpad.net/bugs/1049215
[14:05] <xnox> cjwatson: hmm... I guess I should burn the CD and boot with bluetooth dongle....
[14:05] <xnox> cjwatson: will try when I get home and find my bluetooth dongle....
[14:05] <cjwatson> thanks
[14:06] <xnox> assigned to myself, such that I don't forget ;-)
[15:06] <cjwatson> ev: Could you build wubi r271 at some point?  I did a translation update
[15:06] <ev> cjwatson: sure thing
[15:06] <ev> on it now
[15:06] <cjwatson> Ta
[15:07] <cjwatson> The Tibetan translators have been pretty keen
[15:07] <ev> cjwatson: I've pointed Peter M your way for a few netboot and installation guide questions
[15:07] <ev> just a heads up
[15:07] <cjwatson> OK
[15:50] <ogra_> cjwatson, hmm bug 1032874
[15:50] <ubot2> Launchpad bug 1032874 in linux "Mount command on extended partition hangs forever during install" [High,Triaged] https://launchpad.net/bugs/1032874
[15:51] <ogra_> seems i'm not the only one :(
[15:53] <cjwatson> ogra_: Yeah, I think I've seen similar things before
[15:54] <cjwatson> But it's still right for the installer to avoid this anyway
[15:54] <cjwatson> Oh, and in fact your fix will avoid the bug as stated
[15:54] <ogra_> indeed
[15:54] <cjwatson> I'll follow up to say that
[15:55] <ogra_> but its disappointing to see it hasnt been actively researched ... and just magically disappeared
[15:55] <ogra_> i.e. its likely something we miss on omap4 currently
[15:55] <xnox> cjwatson: how stable is the output of os-prober w.r.t. probing old releases? I'm debugging the re.subst in ubiquity and it seems wrong to me.
[15:56] <cjwatson> xnox: The general syntax should be stable; exact names maybe less so
[15:56] <cjwatson> Although I didn't think it was particularly unstable
[15:56] <xnox> cjwatson: ok. thanks.
[15:57] <xnox> cjwatson: I am running it against a few of my chroots and they are ok. Will get lucid chroot up, just in case it doesn't have what I want.
[16:08] <ev> cjwatson: new wubi is in place
[16:08] <cjwatson> thanks