[01:05] <CIA-18> ubiquity: superm1 * r2168 mythbuntu-ubiquity/ (95 files in 9 dirs): merge with ubiquity trunk
[01:11] <CIA-18> ubiquity: superm1 * r2169 mythbuntu-ubiquity/ (bin/ubiquity debian/changelog): copy xserver-xorg debconf to new install. (only on ubiquity-mythtbuntu until bryce acks this)
[07:32] <CIA-18> ubiquity: evand * r2330 ubiquity/ (3 files in 2 dirs):
[07:32] <CIA-18> ubiquity: * Only set migration-assistant data in user-setup if updating an existing
[07:32] <CIA-18> ubiquity:  user (LP: #135149).
[07:32] <CIA-18> ubiquity: * Properly filter out partitions marked for deletion in
[07:32] <CIA-18> ubiquity:  migrationassistant.py (LP: #146222).
[08:37] <CIA-18> ubiquity: evand * r2331 ubiquity/ (debian/changelog ubiquity/frontend/gtk_ui.py):
[08:37] <CIA-18> ubiquity: * Fixed a bug where the migration-assistant page would not be updated to
[08:37] <CIA-18> ubiquity:  reflect changes, such as if the user went back and partitioned
[08:37] <CIA-18> ubiquity:  differently, if the page had already been shown.
[09:22] <CIA-18> ubiquity: superm1 * r2332 ubiquity/ (bin/ubiquity debian/changelog):
[09:22] <CIA-18> ubiquity: copy over xserver-xorg debconf items (received a +1 from bryce that this is needed for gutsy)
[09:22] <CIA-18> ubiquity: this behavior is a temporary workaround, and will be removed for hardy
[09:30] <evand> I have a hunch that ntfs-3g is the cause of /mnt/migrationassistant still being busy when ma-ask is supposed to unmount it, but I'm not quite sure yet why this would be the case.
[09:30] <evand> ma-ask/ma-apply, really
[09:34] <superm1> can you perhaps add some debugging calls to look for open fd's ?
[09:39] <evand> ah, probably.  But first, sleep.
[09:39] <evand> g'night
[09:39] <superm1> night
[09:39] <superm1> probably a good idea, i should do that too :)
[09:41] <evand> indeed!
[01:05] <xivulon> cjwatson I'd like to introduce a ubiquity/safe_mode boolean so that if that is true, no raw partition gets formatted and MBR is untouched.
[01:05] <xivulon> I'd like to avoid that someone
[01:06] <xivulon> ends up with a formatted partition because fo some (undetected) code bugs and/or preseed errors
[01:07] <xivulon> to be precise now I skip raw partition formatting IF some conditions are true, I do not want to run the danger of having those conditions not to return true...
[01:09] <xivulon> I'll turn up a warning anyway so that syslog can still be used to detect above failures without ending up with people on my doorsteps
[01:09] <xivulon> All of the above will be in my override_files of course
[01:16] <xivulon> To be precise the installer will quit immidiately with a warning msg if we try to perform any "unsafe operation" when safe-mode is on
[04:57] <CIA-18> migration-assistant: evand * r65 migration-assistant/ (debian/changelog ma-ask):
[04:57] <CIA-18> migration-assistant: * Fix logging.
[04:57] <CIA-18> migration-assistant: * Fix IFS handling around call to expr.
[04:58] <CIA-18> migration-assistant: evand * r66 migration-assistant/debian/changelog: releasing version 0.5.3
[05:28] <seb128> hi
[05:29] <seb128> is ubiquity supposed to have a "resize automatically and install" option?
[05:32] <xivulon> cjwatson, I did a bit of research, and do not think that creating large contiguous image files fast is an easy task in Linux
[05:34] <xivulon> the truth to be told that operation is easier in windows, is creating images within autoapartition-loop a strong preference?
[05:34] <cjwatson_> seb128: usually but it's not guaranteed - depends on your partition layout
[05:35] <cjwatson_> xivulon: talk to me after gutsy or at UDS
[05:35] <cjwatson_> I don't have brain-space now :)
[05:35] <cjwatson_> seb128: /var/log/partman will have a note buried in it somewhere as to why it isn't being shown
[05:35] <xivulon> eh I have to ship a release too :P
[05:35] <seb128> cjwatson_: I tried on my new laptop yesterday which had one 110G ntfs partition with 10G used
[05:36] <cjwatson_> xivulon: in general I have a strong preference for doing partitioning in Linux, but what you do in your release is your business :)
[05:36] <seb128> ok, I'll have a look if I still get the issue
[05:36] <seb128> I used gparted to resize
[05:36] <seb128> and installed on this partition
[05:36] <evand> cjwatson_: if it's reasonable and you have time, can you upload a new m-a and ubiquity?  I don't need it to go on the CD.
[05:36] <cjwatson_> seb128: how much unpartitioned space?
[05:36] <cjwatson_> or maybe ntfsresize barfed on it
[05:36] <evand> http://people.ubuntu.com/~evand/upload/migration-assistant_0.5.3.dsc
[05:37] <xivulon> cjwatson, by the way in case you did not nonticed I submitted a couple of bug reports re sendsigs
[05:37] <evand> I cannot currently prepare the latter as there's some sort of weird networking issue between me and launchpad/ubuntu.com
[05:37] <cjwatson_> evand: ok
[05:37] <cjwatson_> xivulon: I didn't notice; I can't get to my mail today
[05:37] <evand> cjwatson_: thanks, I very much appreciate it
[05:37] <xivulon> https://bugs.launchpad.net/ubuntu/+source/partman-auto-loop/+bug/150822
[05:38] <seb128> cjwatson_: none, there is a few Go dell rescue partition, the 110G window vista and like 6-7M which was listed as not partitioned IIRC
[05:39] <seb128> cjwatson_: and after the resize and ubuntu installation the windows vista boot choice is not listed which is somewhat annoying, I need to figure what happened there
[05:39] <seb128> the "dell utility" item is listed
[05:39] <seb128> and Ubuntu
[05:39] <seb128> and that's it
[05:39] <xivulon> https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/150831
[05:40] <cjwatson_> seb128: the rescue partitions mean that you almost certainly have three primary partitions
[05:40] <cjwatson_> seb128: that's enough to inhibit auto-resize at present. we may be able to relax this in hardy.
[05:41] <cjwatson_> seb128: the vista thing should be fixed in today's daily - what were you testing?
[05:41] <seb128> I think I had 2
[05:41] <seb128> the vista one and the dell one
[05:41] <cjwatson_> seb128: my Dell laptop shipped with three
[05:41] <seb128> let me check
[05:45] <cjwatson_> xivulon: both look plausible, can you deal with both of those in local overrides for gutsy?
[05:48] <seb128> cjwatson_: ah right, there is a 94.1M fat16 partition
[05:48] <seb128> cjwatson_: ok, so the resize option is not listed because of the 3 primary partitions?
[05:48] <seb128> that's slightly annoying because ubiquity let you no other choice than to format windows
[05:49] <cjwatson_> seb128: or manual partitioning, come on :)
[05:49] <cjwatson_> seb128: but yeah, I know it sucks, can't fix it for gutsy though
[05:49] <cjwatson_> there's a bug filed about it
[05:49] <cjwatson_> it's because partman used to think that the root partition absolutely had to be primary
[05:49] <seb128> you can resize from manual partitionning?
[05:49] <cjwatson_> we can relax that for gutsy because the bootloader installers deal with logicals
[05:49] <cjwatson_> seb128: yes, of course
[05:49] <cjwatson_> now
[05:49] <seb128> I didn't spot the option, closed and used gparted
[05:49] <cjwatson_> edit the partition and change the size
[05:50] <seb128> that was no obvious to me
[05:50] <seb128> and there was no label on the button suggesting I can resize
[05:50] <seb128> I didn't try that
[05:50] <cjwatson_> we have a design for making it a nice drag thing, but we need gtk/cairo help to actually implement it ...
[05:50] <seb128> ok
[05:50] <cjwatson_> it's true it's not as obvious as it could be at the moment
[05:51] <seb128> and do you have any idea why the vista partition is not listed in grub after installation?
[05:51] <evand> that may be fixed
[05:51] <evand> was this a daily desktop CD?  If so, what date?
[05:52] <cjwatson_> seb128: 16:41 <cjwatson_> seb128: the vista thing should be fixed in today's daily - what were you testing?
[05:52] <evand> ah, I missed that as well
[05:52] <seb128> ups, sorry I didn't read that
[05:52] <seb128> I did the installation yesterday afternoon
[05:52] <seb128> and that was the daily from yesterday
[05:52] <cjwatson_> it was a combination busybox/os-prober bug
[05:52] <cjwatson_> fixed yesterday (I hope)
[05:52] <cjwatson_> would appreciate re-testing today
[05:52] <seb128> I'll try a new installation now
[05:52] <seb128> I was about to do it
[05:53] <seb128> I've not configured anything on the laptop yet, I will use it to test installs until gutsy
[05:53] <cjwatson_> cool
[05:53] <seb128> cjwatson_, evand: thanks
[05:53] <cjwatson_> laptop refresh program?
[05:53] <seb128> yes ;)
[05:54] <cjwatson_> worked for me ;-)
[05:56] <xivulon> cjwatson_: overrides are fine
[05:56] <cjwatson_> xivulon: ok, good
[05:56] <cjwatson_> those two will be easy to fix when hardy opens
[05:56] <cjwatson_> evand: m-a uploaded
[05:59] <xivulon> cjwatson, quick question is killall5 picky about sendsigs.omit formatting? I mean things like: pidof NOPROCESS >> sendsigs.omit, 2 X pidof sameprocess >> sendisgs.omit, pidof morethanoneprocess >> sendsigs.omit...
[05:59] <cjwatson_>         if [ -e /var/run/sendsigs.omit ] ; then
[05:59] <cjwatson_>                 for pid in $(cat /var/run/sendsigs.omit); do
[05:59] <cjwatson_>                         OMITPIDS="${OMITPIDS:+$OMITPIDS }-o $pid"
[05:59] <cjwatson_>                 done
[05:59] <cjwatson_>         fi
[06:00] <cjwatson_> duplicates won't matter and nonexistent pids will be ignored
[06:00] <xivulon> 2 pids in one line?
[06:00] <cjwatson_> not a problem
[06:00] <xivulon> great
[06:00] <cjwatson_> it does care that they're all integers
[06:01] <cjwatson_> and you aren't allowed more than 16
[06:01] <cjwatson_> (stupid arbitrary limit, sorry, was in a hurry)
[06:01] <xivulon> you can tell when it's C and not python...
[06:02] <CIA-18> ubiquity: cjwatson * r2333 ubiquity/ (configure configure.ac): bump to 1.6.4
[06:04] <CIA-18> ubiquity: cjwatson * r2334 ubiquity/ (d-i/manifest debian/changelog): * Automatic update of included source packages: migration-assistant 0.5.3.
[06:08] <xivulon> cjwatson, maybe you want to consider the latest lupin, rev 52 diff is 1 lines and avoids having to know what $rootmount is in preseed, 53 fixes the bug above (but forgot to check whether pidof is available in initrd).
[06:10] <cjwatson_> xivulon: sorry, if it's not completely necessary, it's too late
[06:10] <xivulon> no big issue
[06:10] <cjwatson_> the uploads I'm doing for evand fix serious installer bugs and data loss
[06:10] <cjwatson_> they're not casual targets of opportunity :)
[06:11] <xivulon> well sendsigs/update-grub bugs might be very annoying, but since I am probably the only one on a fuse filesystem, that should be ok for an override
[06:12] <xivulon> no sendsigs ~ hard-reboot
[06:14] <cjwatson_> evand: all of your ubiquity changes are tested, right?
[06:14] <cjwatson_> silly question but it's worth checking :)
[06:14] <cjwatson_> (I've been known to skip testing obvious changes myself!)
[06:15] <evand> yes
[06:15] <evand> if you'd like I can test further, but I tested everything as I went
[06:15] <evand> on multiple VMs
[06:17] <cjwatson_> no, that's fine, I just wanted to be sure as I wasn't sure I understood the ubiquity m-a changes
[06:17] <cjwatson_> but I think I grok them now
[06:17] <bdmurray> Hello, I ran across a bug that might worth looking at.
[06:17] <bdmurray> bug 145836 regarding ubiquity
[06:18] <cjwatson_> bdmurray: most likely unionfs fallout
[06:18] <cjwatson_> the suggested fix cannot work unless the filesystem is fucked
[06:19] <CIA-18> ubiquity: cjwatson * r2335 ubiquity/debian/changelog: releasing version 1.6.4
[06:19] <cjwatson_> bdmurray: I've commented in the bug to that effect
[06:21] <bdmurray> Ah, that makes sense.  The dailies should not see that issue then?
[06:22] <cjwatson_> bdmurray: btw, this is exactly an example of a bug I don't think should be Triaged ;-)
[06:22] <cjwatson_> bdmurray: I hope not
[06:22] <cjwatson_> (not Triaged => because as a general rule complete system hangs in the live CD are not ubiquity bugs)
[06:23] <bdmurray> I'll look again but I thought they said they could continue to use the Live CD so that isn't a complete system hang right?
[06:25] <cjwatson_> some of them said they couldn't
[06:25] <cjwatson_> the installer basically never hangs of its own accord in the range of the progress bar when it's copying files, at any rate
[06:25] <cjwatson_> 53% is right in the middle of that
[06:26] <cjwatson_> anyway, the point is, to a developer the bug reeks of "live session is irreparably screwed by something outside the installer's control", but I wouldn't expect a QA person without direct installer development experience to know that
[06:27] <cjwatson_> it's a good example of where development experience (specifically knowledge of what's going on at that progress bar step) can help work out what's wrong
[06:30] <bdmurray> I understand your point and what Keybuk mentioned yesterday.  However, it seems like only someone familiar with the code could set the bug to Triaged.  Which what I recall from Sevilla was not the goal of the Triaged state.  I'm under the impression the goal was to identify more complete bugs so developers have a smaller pool of bugs to look at.  (Granted this is the ideal workflow.)
[06:31] <cjwatson_> honestly, I would find a state that meant "I have (or somebody I trust to know this code has) looked at this bug and agree it is a problem in this package" much more useful
[06:31] <cjwatson_> because that lets me use those bugs as a list of things to spend hacking time fixing
[06:32] <cjwatson_> whereas the Triaged state as you define it just produces a slightly different list of things I need to look at and accept, which really isn't a significant improvement over Confirmed
[06:33] <cjwatson_> it certainly isn't worth the extra bug state to me
[06:37] <bdmurray> I think the triaged state is worth discussing in more detail then.  Perhaps at UDS?
[06:41] <cjwatson_> bdmurray: *nod*
[06:41] <xivulon> "Triage is a system of sorting patients according to need when resources are insufficient for all to be treated.". Hmm that's a method not a status...
[06:42] <cjwatson_> "triaged" is the past participle of "triage" and thus has a more adjectival meaning
[06:43] <xivulon> Hmm but if treage is a "system of sorting", triaged means "sorted"... Doesn't say much about the sorting order... In fact then there are triaged "tags"
[06:43] <xivulon> but english is not my thing...
[06:45] <cjwatson_> triaged means "the nurse (bug investigator) has looked at the patient (the bug) and agreed that he (it) needs to see a doctor (a developer)"
[06:45] <cjwatson_> (I think the state is misnamed as well as less than useful ;-))
[06:45] <cjwatson_> well, actually, it's probably correctly named with its current definition
[06:47] <xivulon> Guess that's when there are no nurses around, we should stick to triaged then...
[07:24] <seb128> I'm trying a daily install on my laptop
[07:25] <seb128> looks like it's blocked on the langpack installation (87% not moving for some time), my wireless connection is not activated (didn't pick it in network manager), is that expected?
[07:25] <evand> yeah, I'm working on a fix for tht
[07:25] <evand> that*
[07:25] <evand> cjwatson_: do you think checking n-m for an active connection, since we use it by default now, would be reasonable?
[07:26] <cjwatson_> perhaps for hardy
[07:26] <cjwatson_> I think we're stuck with what we've got now
[07:26] <evand> ok
[07:27] <evand> even for kubuntu where it freezes, or is that important enough for a fix before release?
[07:35] <seb128> how do I continue the installation?
[07:35] <seb128> it's stucked at 87% there
[07:35] <evand> seb128: is this an ubuntu daily or kubuntu?
[07:37] <seb128> ubuntu
[07:37] <evand> argh, so it's just locking up rather than giving you a dialog that it cannot find the language packs?
[07:38] <seb128> yes
[07:39] <seb128> the progress bar is at 87% for a while now
[07:39] <seb128> I started the install, closed the lid, went away, came back half an hour later and it was stucked to 87%
[07:39] <seb128> and it's still there
[07:41] <evand> yikes.  I think the only thing you can do is give it a network connection and restart the install.
[07:42] <seb128> doh
[07:42] <seb128> that's not nice
[07:44] <evand> indeed, sorry.  I'm looking for a solution.
[07:47] <seb128> no problem
[07:47] <seb128> I've not stopped it
[07:47] <seb128> if you need any debug info let me know
[08:01] <cjwatson_> I don't think this is any worse than previous releases
[08:01] <cjwatson_> this isn't a new problem, though it may have morphed a bit
[08:03] <seb128> that's the first time I run into it
[08:03] <seb128> but usually I'm playing with the CD during install so I activate the network ;)
[08:07] <cjwatson_> it's fine if there's say a proxy that actively rejects the packets
[08:07] <cjwatson_> if the packets just get dropped, that's when it breaks
[08:07] <cjwatson_> we have an item on the UDS schedule for fixing this properly
[08:07] <seb128> ok, cool
[08:08] <seb128> there used to be a "cancel" button during the language pack downloads no?
[08:14] <xivulon> where do I find the UDS schedule?
[08:22] <seb128> cjwatson_, evand: do you have a bug number about the language pack download bug?
[08:24] <evand> seb128: yes, it used to have a cancel button.  And bug #145012.
[08:25] <evand> xivulon: afaik, there's no published schedule yet.
[08:26] <seb128> evand: thanks
[08:27] <cjwatson_> xivulon: on the whiteboard in the London office
[08:27] <cjwatson_> we're working on it ;-)
[08:35] <seb128> doh
[08:35] <seb128> and how do I workaround the language pack bug on a computer without network?
[08:36] <seb128> I've a desktop with a wired card but not network plug near to use right now
[08:36] <seb128> and the installation is stucked there the same way
[08:36] <evand> install in english
[08:36] <seb128> and I can't activate any network
[08:36] <seb128> that has to be rc
[08:36] <seb128> or any non english install will fail if you have no network connection?
[08:36] <cjwatson_> are any network interfaces up?
[08:36] <cjwatson_> ifconfig?
[08:37] <seb128> yes
[08:37] <seb128> eth0 is up
[08:37] <seb128> but not configured
[08:37] <cjwatson_> I wonder why?
[08:37] <cjwatson_> err, up but not configured?
[08:37] <cjwatson_> can you paste ifconfig output?
[08:37] <seb128> I mean no IP
[08:38] <seb128> no, this computer has no network :p
[08:38] <seb128> I can type though
[08:38] <seb128> something similar to
[08:38] <seb128> eth1      Lien encap:Ethernet  HWaddr 00:40:05:5E:A8:68
[08:38] <seb128>           adr inet6: fe80::240:5ff:fe5e:a868/64 Scope:Lien
[08:38] <seb128>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
[08:38] <seb128>           Packets reus:0 erreurs:0 :0 overruns:0 frame:0
[08:38] <seb128>           TX packets:0 errors:6 dropped:0 overruns:0 carrier:12
[08:38] <seb128>           collisions:102 lg file transmission:1000
[08:38] <seb128>           Octets reus:0 (0.0 b) Octets transmis:468 (468.0 b)
[08:38] <seb128>           Interruption:19 Adresse de base:0xc000
[08:38] <seb128> that is an another computer
[08:39] <seb128> I don't think details count
[08:39] <seb128> I mean the HWaddr, etc
[08:40] <seb128> I didn't touch the intefaces, I guess that's network-manager work
[08:41] <cjwatson_> or casper, conceivably
[08:41] <cjwatson_> I think probably not thouygh
[08:41] <cjwatson_> though
[08:41] <cjwatson_> seb128: anything non-comment in /etc/network/interfaces?
[08:49] <seb128> cjwatson_: the 2 lines about lo and that's everything
[08:49] <seb128> no commented lines
[09:42] <CIA-18> ubiquity: evand * r2336 ubiquity/ (configure configure.ac): bump to 1.6.5
[09:47] <CIA-18> ubiquity: evand * r2337 ubiquity/ (debian/changelog ubiquity/frontend/kde_ui.py):
[09:47] <CIA-18> ubiquity: * Revert accidental copy and paste of migration-assistant specific
[09:47] <CIA-18> ubiquity:  code (LP: #151051).