[00:28] <xivulon> cjwatson, I am quite sure the file is not corrupted (still getting out of disk errors)
[00:43] <xivulon> I mounted the xp vm under my main jaunty setup, run fsck on root.disk, then mounted root.disk and accessed without any problem
[00:44] <xivulon> then unwound everything and tried to access root.disk from kvm/grub2 via loopmount module, but got "out of disk"
[00:47] <xivulon> not sure how to check for lba, it's the stock kvm bios (vgabios cvs 8-jul-09) but do not see any obvious way of displaying the settings
[00:47] <xivulon> soren ^
[01:16] <xivulon> as for the fs corruption that happens after reboot, the error is "contains a filesystem with errors" (EXT2_ERROR_FS || !ext2fs_test_valid())
[09:39] <xivulon> cjwatson, morning, any hint for ^^^?
[09:41] <xivulon> evand, davmor2, do you experience fs corruption in a wubi installation after a reboot?
[09:42] <xivulon> it seems most (all?) of those are just dirty flag
[09:42] <davmor2> xivulon: is it down to time check on superblock
[09:43] <xivulon> ah so it's not just me, glad to hear that
[09:44] <xivulon> at the moment grub.cfg has to be updated manually, I will provide a patch for casper-support
[09:44] <davmor2> xivulon: https://bugs.launchpad.net/bugs/423247
[09:44] <ubottu> Ubuntu bug 423247 in clock-setup "Superblock last mount times cause fsck to fail" [High,Fix released]
[09:45] <xivulon> re grub2 loopmount, I still have "out of disk" errors even after running fschk on the root.disk file (from outside the vm)
[09:46] <xivulon> soren, how can I check whether vgabios used in kvm supports LBA?
[09:51] <CIA-33> ubiquity: evand * r3433 ubiquity/debian/ (changelog ubiquity.templates):
[09:51] <CIA-33> ubiquity: Do not provide an inaccurate remaining time estimate for the file copy
[09:51] <CIA-33> ubiquity: process (LP: #199834). Thanks rugby471!
[10:08] <soren> xivulon: vgabios doesn
[10:08] <soren> t. It's for the graphics card :)
[10:08] <soren> xivulon: The bochsbios should.
[10:10] <CIA-33> ubiquity: evand * r3434 ubiquity/ubiquity/frontend/base.py: Remove reference to a no longer used variable.
[11:40] <xivulon> ahh!
[11:41] <xivulon> soren thanks!
[12:24] <davmor2> evand, xivulon: Running today's iso wubi is working again however after reboot on the linux side.  Hit ubuntu you get the grub menu hit the top option and I get the following error.  error: You need to load the kernel first  press any key to continue...  This takes you back to the grub menu
[12:25] <evand> davmor2: is this after the install completes in its entirety (after ubiquity has run)?
[12:26] <davmor2> evand yes
[12:26] <evand> if you press e instead of hitting enter on the option, what does it show you?
[12:27] <davmor2> set quiet=1
[12:27] <davmor2> insmod ntfs
[12:27] <evand> that's it?
[12:27] <davmor2> set root=(hd0.3)
[12:27] <davmor2> evand hang on I'm copying it all
[12:28] <davmor2> , not . on hd0
[12:28] <davmor2> search --no-floppy --fs-uuid --set f6863c88863c4b7f
[12:29] <davmor2> linux /vmlinuz-2.6.31-9-generic root=/host/ubuntu/disks/root.disk ro quiet splash
[12:30] <davmor2> initrd /initrd.img-2.6.31-9-generic
[12:30] <davmor2> evand that's the lot
[12:31] <evand> davmor2: can you please file a new bug with that information, and attach the wubi log file
[12:31] <davmor2> yes no probs
[12:39] <xivulon> hi davmor2
[12:39] <davmor2> xivulon: hello
[12:40] <xivulon> as mentioned grub.cfg has to be edited
[12:40] <xivulon> will have to submit a patch for lupin to that automatically
[12:40] <xivulon> in the meantime you can change thing manually
[12:40] <xivulon> the line should read:
[12:41] <xivulon> linux /ubuntu/disks/boot/vmlinuz root=/dev/sda1 loop=/ubuntu/disks/root.disk  [other arguments]
[12:42] <xivulon> replace sda1 with whatever is appropriate, similarly, for the initrd, prepend "/ubuntu/disks/boot/"
[12:43] <xivulon> loop might be = /host/ubuntu/disks/root.disk (don't remember on top of my head which one it is)
[12:44] <davmor2> xivulon: wilko I'll try it now
[12:45] <davmor2> evand: bug 426238 fell free to do what ever I'll add the work round to it when I know it works :)
[12:45] <ubottu> Launchpad bug 426238 in wubi "ubuntu fails to boot after install." [Undecided,New] https://launchpad.net/bugs/426238
[12:52] <xivulon> evand, FYI I have been waiting to get grub2 loopmount going first, as in that case the grub config will change
[12:53] <xivulon> was hit by 423247 and lba but hopefully today will get that moving
[12:55] <CIA-33> usb-creator: rgreening * r163 trunk/usbcreator/frontends/kde/frontend.py: Re-enable ISO image support in KDE fronte-end againg. Got lost in one of the prev merges.
[13:10] <davmor2> xivulon: added the temp work round to the bug
[14:02] <mterry> evand, hey, thanks for pushing in the plugins branch!
[14:02] <evand> sure thing
[14:12] <StevenK> evand: Did anything happen with UNR + wubi, by the way?
[14:22] <xivulon> StevenK we played with davmor2, and we still have the size issue
[14:22] <xivulon> evand shall we enable DVDs?
[14:22] <xivulon> or do you prefer for me to check the squasfs instead?
[14:22] <StevenK> I thought the size issue was sorted out
[14:23] <StevenK> I've managed an install off a USB key
[14:23] <xivulon> if I am not mistaken the patch compensated for free space
[14:23] <xivulon> which might still be an issue if you have a large usb disk, with little free space
[14:24] <StevenK> My issue on Friday was that I could install UNR via Wubi, but I couldn't boot
[14:24] <xivulon> ah that's known, I am on it
[14:24] <StevenK> Woot
[14:24] <StevenK> xivulon: Is there a bug about it?
[14:24] <xivulon> See  426238
[14:25] <xivulon> if you look at the irclog above I gave davmor2 instructions for manual override
[14:25] <xivulon> as mentioned I am playing with grub2 loopmount, once that is sorted I will take care of update-grub
[14:27] <rgreening> evand: any progress with working around format backend issue in usb-creator?
[14:29] <evand> rgreening: not quite yet.  davidz hasn't responded, so I'm going to have to look into it myself.  I'm caught up in getting policykit support working at the moment, but I'll make sure I find time for it today/tomorrow.
[14:30] <rgreening> oh, policy-kit is a bigger priority. :)  Will the kde FE be able to take advantage then?
[14:31] <rgreening> evand: as soon as you have something ready for policy-kit, let me know and I can add the kde bits and test...
[14:31] <evand> will do!
[14:31] <rgreening> assuming it's straight forward
[14:31] <rgreening> ty evand
[14:31]  * rgreening cant wait to remove kdesudo
[14:33] <davmor2> StevenK: it may also be that the wubi builds were failing so it dropped to version 142 rather than rev 150
[14:33] <evand> yeah, that's quite the kludge
[14:33] <davmor2> try again with todays you might have more luck :)
[14:34] <davmor2> StevenK: also I add the work round to the bug above
[15:01] <kirkland> evand: usb-creator seems broken again
[15:01] <kirkland> evand: known issue?
[15:01] <evand> define broken
[15:01] <evand> there are a few fixed issues that are in trunk
[15:03] <kirkland> evand: http://rookery.canonical.com/~kirkland/fail.png
[15:03] <evand> interesting.
[15:03] <evand> can you pastebin /tmp/usb-creator.log
[15:05] <kirkland> evand: http://paste.ubuntu.com/267299/
[15:07] <evand> kirkland: does the CD image you're using not have an isolinux directory?
[15:07] <kirkland> evand: dunno ... it's the current karmic desktop amd64 iso
[15:07] <kirkland> evand: i can crack it open
[15:08] <evand> kirkland: can you pastebin isoinfo -J -l -i whatever.iso ?
[15:08] <kirkland> evand: yeah it has isolinux
[15:08] <evand> curious.
[15:09] <kirkland> evand: http://paste.ubuntu.com/267304/
[15:09] <kirkland> evand: i downgraded to jaunty's usb creator and it's working
[15:11] <evand> oh, I've missed the obvious in the log file
[15:12] <evand> it's not copying any relevant files over
[15:21] <CIA-33> usb-creator: evand * r164 usb-creator/usbcreator/ (backends/devicekit/backend.py install.py): Add debugging information on install choices.
[15:34] <CIA-33> usb-creator: evand * r165 usb-creator/po/ (37 files): Update translations from Launchpad.
[15:45] <CIA-33> usb-creator: evand * r166 usb-creator/debian/changelog: releasing version 0.2.5
[15:54] <rgreening> evand: you know that in the current version of crewator it shows drives with 0 capacity? Just wondering if we should filter them out appropriately.
[15:54] <rgreening> and best way to go about it ( i assume in back-end)
[15:56] <evand> it's 0 bytes free
[15:56] <evand> not 0 bytes capacity
[15:56] <evand> but it's for entire disks
[15:56] <evand> where there's no concept of free space
[15:56] <evand> suggestions welcome
[15:57] <evand> (see the devicekit backend's _add_drive function for the relevant code)
[16:00] <rgreening> evand: if capcity == 0 and free space == 0 then should we filter?
[16:02] <evand> I don't think we ever want to hide a partition (disks are a different story, but one step at a time :)).  If we need to delete some files, then we should put up the "need more space" message below the target list box with an open button next to it that spawns a file browser.
[16:02] <rgreening> evand: hmm... I get /dev/sr0 which is my CD-ROM showing up in target list
[16:03] <evand> rgreening: is that with the latest bzr?
[16:03] <rgreening> yep
[16:04] <superm1> evand, wrg to the permissions drop on the UI for ubiquity, do they actually get fully restored by the end for that 'reboot' command?  (referencing bug 412825)
[16:04] <ubottu> Launchpad bug 412825 in upstart "Clicking reboot doesn't actually reboot the system" [Undecided,New] https://launchpad.net/bugs/412825
[16:04] <superm1> I thought they were
[16:04] <rgreening> ah-ha... evand, if I load the CD-ROM with a proper imatall media first, then launch creator, it works as expected.
[16:05] <rgreening> however evand, if I do not have a CD inserted, it shows up in the targets list and does not allow me to add the CD to the source list (as its already in the target)
[16:05] <evand> rgreening: hrm, r160 should have fixed that
[16:06] <evand> I'll look into it
[16:06] <evand> superm1: checking
[16:06] <rgreening> evand: ok
[16:06] <rgreening> I see the same in both kde and gtk FE
[16:12] <CIA-33> ubiquity: evand * r3435 ubiquity/ (debian/changelog ubiquity/frontend/gtk_ui.py): Call reboot as root. Thanks Mario Limonciello!
[16:12] <evand> superm1: so it is not.  Thanks a bunch!
[16:12] <superm1> evand, wow crazyness.
[16:12] <superm1> how did that ever work before?
[16:12] <rgreening> evand: the capacity && free == 0 question, both are referencing disks here. E.g. the /dev/sr0 and /dev/sdb. sr0 is my CD-ROM and sdb is my card reader. Both show in target list and 0 capacity and 0 free which is correct since neither have media in them at the time. SO they really shoul dnever hit the target list (I believe)
[16:12]  * evand shakes his head.
[16:13] <evand> no idea
[16:13] <evand> rgreening: I disagree on /dev/sdb.  It allows us to use it as a target for disk images (do recall that usb-creator supports those) and format the device when it's unpartitioned.
[16:14] <evand> we should fix it so that cd rom devices do not show up in the targets list though
[16:14] <rgreening> evand: my point is, that I have no card in it and it shows up with 0 free and 0 capacity
[16:14] <evand> ah
[16:14] <rgreening> so exactyl how is that beneficial :) format thin air :P
[16:14] <evand> I see your point
[16:14] <rgreening> haha
[16:14] <rgreening> ;)
[16:14] <evand> apologies for that
[16:15] <rgreening> np
[16:15] <superm1> maybe only show /dev/sdb when you select a .img file in the first place then
[16:15] <superm1> if it's an ISO, then don't ever offer it
[16:15] <evand> superm1: that's the plan, but I'd like to get it working without issue first
[16:16] <superm1> ah ok :)
[16:16] <rgreening> evand: also, the fact that the CD-ROM cannot be picked up in source list if the CD is inserted after launching the creator is a problem (could be related to it being already in the target list I guess)
[16:16] <evand> (it was also need to show it when a device was unpartitioned or had no vfat partitions, so you could format the device)
[16:17] <evand> rgreening: not sure why that's happening.  I'll add it to my list of things to check out.
[16:22] <CIA-33> usb-creator: evand * r167 usb-creator/ (debian/changelog usbcreator/backends/devicekit/backend.py): Ignore 0 byte devices. Thanks Roderick B. Greening!
[16:26] <rgreening> evand: yeah, a device with no partitions, will still have a > 0 capacity :)
[16:26] <rgreening> ty for the fix
[16:32] <rgreening> evand: tested. 0 capacity filtering works as expected now. cheers.
[16:32] <evand> hoora
[16:32] <evand> y
[16:32] <rgreening> haha
[16:33] <rgreening> so, now just need to get it to pick up CD insertion after the fact...
[16:33] <rgreening> I'll take a peek and see if I can figure it out
[16:35] <evand> cool, thanks
[16:35] <rgreening> np. gonna reboot.. bbiab
[16:46] <CIA-33> ubiquity: mterry * r3436 trunk/ (3 files in 2 dirs): oem-config: restart HAL when finished
[16:51] <superm1> mterry, with halsectomy going on, perhaps you should add hal as a depends to oem-config for now as a reminder that script needs to be dropped when hal goes away
[16:52] <mterry> superm1, fair enough
[16:55] <CIA-33> ubiquity: mterry * r3437 trunk/debian/control: oem-config: add hal to Depends for last checkin -- when hal is finally dropped, we should remove this and update the ubi-restart-hal script that restarts it
=== nxvl_ is now known as nxvl
[17:56] <rgreening> evand: ping
[18:01] <rgreening> evand: In creator gtk frontend, there's a timer for update_free. Is this still necessary? If so, I need to mirror this code in kde front-end.
[18:01] <rgreening> otherwise, it should be removed from gtk version, along with the dead code (add/delete_timeout)
[18:07] <rgreening> evand: I think the CD-ROM detection issue required a dbus signal in detect_devices for "DeviceChanged" ... maybe...
[18:21] <rgreening> evand: yep. DeviceChanged will catch it... but it seems the device is constantly in a changed status CD or no CD inserted.
[18:21] <rgreening> so, it's not quite correct...
[18:21] <rgreening> ideas evand?
[19:04] <rgreening> evand: have a look at this... http://paste.ubuntu.com/267435/
[19:05] <rgreening> evand: it's my attempt at fixing the optical drive detection not working if inserted after creator is run. It works, but may not be the best approach.
[19:06] <rgreening> evand: I believe the devicechanged polling for optical drive needs to deal with add and remove, cause if you also eject the optical media, the normal remove device doesn't apply/work either.
[19:07] <rgreening> since the device file exists regardless of any media in the drive, and the backend only detects new or removed device file entries.
[19:07] <rgreening> anyway, have a look. Let me know if this is ok to commit for now until we get a better solution (I have in my local bzr checkout rdy to apply)
[19:12] <CIA-33> usb-creator: rgreening * r168 trunk/usbcreator/frontends/kde/frontend.py:
[19:12] <CIA-33> usb-creator: Make kde frontend call detect_devices from private ref (self.__backend).
[19:12] <CIA-33> usb-creator: Add update_loop timer to mirror gtk frontend (is this still required by either frontend?)
[19:16] <CIA-33> usb-creator: rgreening * r169 trunk/debian/changelog: update changelog
=== robbiew is now known as robbiew-afk