Re: [Bug 159772] Re: "redirect" binary is not statically linked

[Paul Whittaker]

2008/11/12 Daniel T Chen <[EMAIL PROTECTED]>

> Is this symptom still reproducible in 8.10?
>
> ** Changed in: thttpd (Ubuntu)
>   Status: New => Incomplete
>
> --
> "redirect" binary is not statically linked
> https://bugs.launchpad.net/bugs/159772
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Hi Daniel,

Yes it is.  [Shell transcript follows.]

Regards,

Paul.


[EMAIL PROTECTED]:~$ *ls /usr/lib/cgi-bin/redirect
ls: cannot access /usr/lib/cgi-bin/redirect: No such file or directory
[EMAIL PROTECTED]:~$ *aptitude show thttpd-util
Package: thttpd-util
State: not installed
Version: 2.25b-3
Priority: optional
Section: universe/web
Maintainer: Ubuntu MOTU Developers <[EMAIL PROTECTED]>
Uncompressed Size: 127k
Depends: libc6 (>= 2.6.1-1)
Description: tiny/turbo/throttling HTTP server (utilities)
 thttpd is a small, fast secure webserver. It features CGI support,
 URL-traffic-based throttling and basic authentication. thttpd has a very
small
 memory footprint as it only forks itself in order to execute CGI scripts.
It is
 designed to be as fast as fully featured web-servers and it performs
extremely
 well under high load.

 Homepage: 

 This package contains various support utilities for thttpd web-server.

[EMAIL PROTECTED]:~$ *sudo aptitude install thttpd-util
[sudo] password for pdw:
Reading package lists... Done
Building dependency tree
Reading state information... Done
Reading extended state information
Initialising package states... Done
Building tag database... Done
The following packages have been kept back:
  alacarte base-files dbus dbus-x11 foo2zjs libdbus-1-3 libsmbclient
  module-init-tools samba-common smbclient update-notifier
  update-notifier-common
The following NEW packages will be installed:
  thttpd-util
0 packages upgraded, 1 newly installed, 0 to remove and 12 not upgraded.
Need to get 27.8kB of archives. After unpacking 127kB will be used.
Writing extended state information... Done
Get:1 http://gb.archive.ubuntu.com hardy/universe thttpd-util 2.25b-3
[27.8kB]
Fetched 27.8kB in 0s (178kB/s)
Selecting previously deselected package thttpd-util.
(Reading database ... 157561 files and directories currently installed.)
Unpacking thttpd-util (from .../thttpd-util_2.25b-3_i386.deb) ...
Setting up thttpd-util (2.25b-3) ...
Reading package lists... Done
Building dependency tree
Reading state information... Done
Reading extended state information
Initialising package states... Done
Writing extended state information... Done
Building tag database... Done
[EMAIL PROTECTED]:~$ *ldd /usr/lib/cgi-bin/redirect
linux-gate.so.1 =>  (0xb7fa1000)
libcrypt.so.1 => /lib/tls/i686/cmov/libcrypt.so.1 (0xb7f6)
libc.so.6 => /lib/tls/i686/cmov/libc.so.6 (0xb7e11000)
/lib/ld-linux.so.2 (0xb7fa2000)

-- 
"redirect" binary is not statically linked
https://bugs.launchpad.net/bugs/159772
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 86704] Re: Empty fonts.dir file stops fonts being found

[Paul Whittaker]

2008/11/5 Adam Niedling <[EMAIL PROTECTED]>

> What version of Ubuntu are you using? Is this still an issue?
>

I've mostly moved to using Hardy now.

Unfortunately I can no longer recall exactly what I was doing that meant I
got bitten by this problem in the first place :-(  but the hardy laptop I'm
currently sitting at has the relevant files in the right places, and
correctly reports fonts via xlsfonts, so as far as I can see it's no longer
a current problem.

Thanks for chasing it up,

Paul.

-- 
Empty fonts.dir file stops fonts being found
https://bugs.launchpad.net/bugs/86704
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 56696] Re: Installer crashed

[Paul Whittaker]

Retrying installation with net connection set up properly and the FAT32
partition fscked and fixed still gives crash with identical traceback.

-- 
Installer crashed
https://launchpad.net/bugs/56696

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 56696] Re: Installer crashed

[Paul Whittaker]

Fixed.  Make root filesystem ext3, not xfs.  Details below.

--
Installer was crashing as soon as it said "Configuring boot loader...".

Experimenting with grub-installer showed that things weren't working
correctly.  This was the setup:

 - New root partition at /dev/sda2
 - Partition mostly installed; installer only failed after setting up most of 
the files on the system.
 - Mounted partition at /mnt/sda2

Running this:

grub-install /dev/sda --root-directory=/mnt/sda2

displayed message to effect that "if you run this on non-xfs
filesystems, you'll get a segfault which can be ignored; this is due to
xfs_freeze", and then hung.

Retrying install with new root partition formatted as ext3 rather than
as xfs proceeds completely successfully.  Didn't try it without the net
connection, though.

All working, for now.

-- 
Installer crashed
https://launchpad.net/bugs/56696

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 106226] Re: Background image makes pages hard to read.

[Paul Whittaker]

The three background files are corrupted.  They all have the MD5
checksum of 0752d882c1bcc139bc18d3af05457b01, whilst the background
images found at
http://www.ross.net/funnelweb/developer/binary/background.gif (etc.) all
have checksum e2ff2656cc11b545d351d1804d144887.  Restoring these puts
the page back as intended.

Other binary files look a little corrupted too, though, such as
file:///usr/share/doc/funnelweb-doc/reference/binary/title.gif - it
might be worth checking the images are all ok.

-- 
Background image makes pages hard to read.
https://bugs.launchpad.net/bugs/106226
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 105784] Re: gnuplot.info.gz not a complete info file

[Paul Whittaker]

In the meantime, manually replacing the packaged copy of
/usr/share/info/gnuplot.info.gz with a gzipped copy of the (single)
gnuplot.info file in ftp://ftp.gnuplot.info/pub/gnuplot/gpdoc.zip worked
fine for me.

-- 
gnuplot.info.gz not a complete info file
https://bugs.launchpad.net/bugs/105784
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 105784] Re: gnuplot.info.gz not a complete info file

[Paul Whittaker]

I'm also getting this problem.  Trying to view the gnuplot entry in
Emacs gives the error "info-insert-file-contents: Can't find
gnuplot.info-1 or any compressed version of it".

The relevant section of gnuplot.info.gz would appear to be:

Indirect:
gnuplot.info-1: 206
gnuplot.info-2: 291476

Neither of the ...-1 or ...-2 files are present in the package, and do
not appear to be in any package when searching at packages.ubuntu.com.

-- 
gnuplot.info.gz not a complete info file
https://bugs.launchpad.net/bugs/105784
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 159772] "redirect" binary is not statically linked

[Paul Whittaker]

Public bug reported:

Binary package hint: thttpd-util

The thttpd-util package includes a CGI program ("/usr/lib/cgi-
bin/redirect") to support page redirection.  To use this utility, you
place the "redirect" executable in your CGI directory, and add a symlink
to it from the page you want to redirect.

The supplied binary (in version 2.23beta1-6) is dynamically-linked:

 $ ldd /usr/lib/cgi-bin/redirect 
libcrypt.so.1 => /lib/tls/i686/cmov/libcrypt.so.1 (0x0071)
libc.so.6 => /lib/tls/i686/cmov/libc.so.6 (0x00d82000)
/lib/ld-linux.so.2 (0x00bc1000)

If you choose to run the thttpd server chrooted (which is the default
setting, and also securest), "redirect" will fail to run correctly and
thttpd returns an error page instead of a redirection.

Recompiling redirect.c statically produces a CGI script that works.

** Affects: thttpd (Ubuntu)
 Importance: Undecided
 Status: New

-- 
"redirect" binary is not statically linked
https://bugs.launchpad.net/bugs/159772
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 39856] Re: Panel in middle of the screen after reboot when not full-sized

[Paul Whittaker]

I've tried the following workaround, and it works for me across a
session logout/login.

Short:
Use gconf-editor to change the value of 
/apps/panel/toplevels/bottom_panel_screen0/y to the vertical screen resolution.

Notes:
I ran gconf-editor from the terminal.  There were other keys for 
bottom_panel_screen1 and top_panel_screen[0/1] visible, but I tried this one 
first and it worked for me.  When I changed the value to the current vertical 
resolution it appeared to invoke some magic, because the value automatically 
adjusted to put the panel at the bottom of the screen: I typed "1024", hit 
return, and a split second later it said "975" where it used to say "1024".  
The calculation appears to be ((Y resolution - Panel size) - 1).  Panel size is 
visible in the panel properties dialog.  As soon as the new value was entered, 
the panel moved to the new position, and I quit gconf-editor.

Thanks:
Originally found in dbloom's post at 
http://ubuntuforums.org/showthread.php?t=117997

-- 
Panel in middle of the screen after reboot when not full-sized
https://bugs.launchpad.net/bugs/39856
You received this bug notification because you are a member of Ubuntu
Bugs, which is a direct subscriber.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 112046] ifuser segfault with solitary -v flag

[Paul Whittaker]

Public bug reported:

Binary package hint: pcmcia-cs

[EMAIL PROTECTED]:~$ ifuser -v
Segmentation fault (core dumped)

This isn't causing me any problems, but it's disturbing that it
segfaults rather than just complaining.

I have a /var/crash/ file which I will attach shortly.

[I'm also opening the inability of Launchpad to let me attach files
*with* a submission as a separate bug.]

** Affects: pcmcia-cs (Ubuntu)
 Importance: Undecided
 Status: Unconfirmed

-- 
ifuser segfault with solitary -v flag
https://bugs.launchpad.net/bugs/112046
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 112046] Re: ifuser segfault with solitary -v flag

[Paul Whittaker]

** Attachment added: "Crash report from Ubuntu "Problem report" helper"
   http://librarian.launchpad.net/7533897/_sbin_ifuser.1000.crash

-- 
ifuser segfault with solitary -v flag
https://bugs.launchpad.net/bugs/112046
You received this bug notification because you are a member of Ubuntu
Bugs, which is the bug contact for Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 68027] Re: sudo update-manager -c -d crashes during xubuntu upgrade

[Paul Whittaker]

I had this problem yesterday upgrading Ubuntu Dapper to Edgy.  Note that this 
wasn't XUbuntu.
I was running the update via  `gksudo "update-manager -c" '

My experience was exactly as described above: package downloading went
fine, lots of "preparing to install..." reports with no problem, and
then lots of "Setting up..." reports before the crash.  The estimate was
down to "About 2 minutes" from an initial 30, which was fairly accurate
based on the clock I was watching.

Shortly before the crash, the reports appeared to stop (there had new
output at least every couple of seconds up to this point, and this was a
pause of many seconds with no output), and the load meter hit 100% for
several seconds (ten? fifteen?) before the installer window simply
vanished.  No error dialog, just "Baff! Where did that window go?"

Re-running update-manager from the System menu brought up a window of
the right size and shape, but it had no content.  I tried this several
times with no success.

Thinking that a reboot might solve the problem (it did on the other
machine I'd upgraded when the installer gave up), I found that the
install would not boot.  Grub was fine, and there was a new entry at the
top of the list, but the screen stayed black with a flashing cursor for
several minutes without any harddrive activity at all.  Rebooting and
trying the "recovery mode" kernel didn't work either, but from the boot
messages it was clear that the SATA modules were not present.  This
explained the kernel's final "waiting for root filesystem..." message,
since that was on /dev/sda5.  After some length of time (five minutes?)
this gave up, and dropped me into BusyBox.  There was a message about
being "unable to find tty" (I forget the precise words).

Eventually I managed to boot the system using an old kernel.  There was
a major problem with the filesystem (I assume it was the root partition,
though it could have been /usr) which dropped me into a manual fsck.  A
large number of inodes (about a hundred) needed correcting.  I didn't
catch the exact error, but I got the impression that inodes that no
longer appeared to be in use had not been freed properly, so inode
counts and free space bitmaps were wrong.

The system then booted, but X was broken.  It was trying to open
/dev/input/mice rather than /dev/input/mouse0 and so died because it had
no core pointer.  I ran `dpkg --configure -a' and that seemed to fix
things - gdm came back up - but a "bad variable" error in
/etc/gdm/Xsession stopped me from logging in.

Changing to the "default Gnome" session allowed me to log in, but the
system was obviously still confused: browsers and terminals would open,
but dropdown menus didn't (they highlighted, but no menu dropped), and I
couldn't give keyboard input (or focus - cursor stayed hollow) to
anything because the keyboard had stopped responding.  LEDs, Ctrl-
Alt-F1, everything.  After I'd been looking at it for several minutes
the console did eventually switch, the keyboard started responding
again, and I could log in to a virtual terminal, but the problem logging
in to GDM still remained.

My assumption is that the system is subtlely broken, probably due to
missing files or other filesystem problems, and needs a fresh install.
At the moment, though, I haven't had time to do this, so the system is
still as described above.

If there's any way to breathe life back into the system I'd be glad to
know it - I still have all the packages it downloaded, and wondered if a
"aptitude reinstall everything" might be possible.  I need to make sure
that there isn't any real filesystem damage, however, and am prepared to
reinstall just for peace of mind if that's necessary.

Anyway, until then I have a "live" broken system available.  If there's
anything I can check / try out / attach / whatever then please let me
know - I don't have much to lose.  I'll try holding off the reinstall
for a day or so

-- 
sudo update-manager -c -d crashes during xubuntu upgrade
https://launchpad.net/bugs/68027

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 68027] Re: sudo update-manager -c -d crashes during xubuntu upgrade

[Paul Whittaker]

Summary of last paragraph of previous comment:  "Live" broken system
available - anything people want me to try on it?

[Sorry - hadn't realised comments were clipped.]

-- 
sudo update-manager -c -d crashes during xubuntu upgrade
https://launchpad.net/bugs/68027

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 68027] Re: sudo update-manager -c -d crashes during xubuntu upgrade

[Paul Whittaker]

Attaching logs from /var/log/dist-upgrade on broken system.  apt.log
seems to indicate problems with broken dependencies and increasingly
desperate attempts to find solutions.  Things have also gone badly wrong
in term.log, as the file ends part-way through a line with 130 null
bytes.

-- 
sudo update-manager -c -d crashes during xubuntu upgrade
https://launchpad.net/bugs/68027

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 68027] Re: sudo update-manager -c -d crashes during xubuntu upgrade

[Paul Whittaker]

** Attachment added: "/var/log/dist-upgrade/apt.log (2nd machine)"
   http://librarian.launchpad.net/6538753/apt.log

-- 
sudo update-manager -c -d crashes during xubuntu upgrade
https://launchpad.net/bugs/68027

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 68027] Re: sudo update-manager -c -d crashes during xubuntu upgrade

[Paul Whittaker]

** Attachment added: "/var/log/dist-upgrade/main.log (2nd machine)"
   http://librarian.launchpad.net/6538772/main.log

-- 
sudo update-manager -c -d crashes during xubuntu upgrade
https://launchpad.net/bugs/68027

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 68027] Re: sudo update-manager -c -d crashes during xubuntu upgrade

[Paul Whittaker]

** Attachment added: "/var/log/dist-upgrade/term.log (2nd machine)"
   http://librarian.launchpad.net/6538774/term.log

-- 
sudo update-manager -c -d crashes during xubuntu upgrade
https://launchpad.net/bugs/68027

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 68027] Re: sudo update-manager -c -d crashes during xubuntu upgrade

[Paul Whittaker]

Giving up hope on fixing this install: I'm getting some screen corruption with 
elinks on the console (misplaced characters), and ld is having trouble:
  ld: warning: cannot find entry symbol _start; defaulting to 08049720
Hopefully the fresh install will be easier.  Thanks all anyway.

-- 
sudo update-manager -c -d crashes during xubuntu upgrade
https://launchpad.net/bugs/68027

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 86704] Empty fonts.dir file stops fonts being found

[Paul Whittaker]

Public bug reported:

Binary package hint: xfonts-scalable

This bug appears to be already known, but I cannot find a bug report for
it.

Existing discussion is at
http://ubuntuforums.org/showthread.php?t=313994 .

I was bitten both by bug #63408 (which is mentioned in the above thread,
and which I fixed) and also by their "empty fonts.dir" problem.
aptitude reinstall on both x-ttcidfont-conf and ttf-bitstream-vera had
no visible effect for me - "xlsfonts | grep vera" found no lines.

Using their proposed fix:

   cd /var/lib/defoma/x-ttcidfont-conf.d/dirs/TrueType
   sudo cp fonts.scale fonts.dir

and restarting X (log out and back in) solved the problem entirely.

Prior to this, fonts.dir existed but was only two bytes long.  I think
the first was the (ASCII) digit zero; the second was a newline.  (I was
looking at it through hd, and 0x30 0x0A rings a bell.  (Ok, ok, that's
0x07, I know :-))

Anyway, this seems to be a simple matter to fix, and as no-one else
(appears to have) submitted it, here it is.

PS Apologies if this is the wrong package to report the bug against;
dpkg -S  tells me that /usr/share/fonts/X11/Type1/fonts.dir is from
xfonts-scalable.

** Affects: xfonts-scalable (Ubuntu)
 Importance: Undecided
 Status: Unconfirmed

-- 
Empty fonts.dir file stops fonts being found
https://launchpad.net/bugs/86704

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 567636] Re: mumble stops responding when try to connect to server

[Paul Whittaker]

I'm also getting this on a fresh Lucid install: Run mumble, select
server, click "connect", mumble stops responding and consumes as much
CPU as is available.  If instead I run mumble, go through the audio
wizard (I don't need to change any settings, just keep hitting Next) and
then select the same server and connect, all works as it should.  I had
no problems running Mumble 1.2 on a fresh jaunty install (with mumble
from the mumble PPA) on the precise same hardware a month or two ago.
No hardware is being connected or disconnected at any point.  Microphone
input is on one card, speaker output is on another; previously this used
the default jaunty sound system (alsa?), now it's going through
pulseaudio.

I want to help get this fixed, so if anyone can tell me what information
would be useful (alsa or pulseaudio settings dump, attaching to mumble
with strace, etc.) and the command options to use, I'd be more than
happy to post back information.  I may also be able to reproduce the
earlier jaunty install for comparison if needed.

-- 
mumble stops responding when try to connect to server
https://bugs.launchpad.net/bugs/567636
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 1779548] [NEW] Desktop installer lacks an adequate warning of permanent data loss if attempting to activate existing LUKS containers.

[Paul Whittaker]

Public bug reported:

Background:

I had Ubuntu installed on a laptop using encrypted LVM storage.  Apart
from /boot, the whole disk was an encrypted LUKS volume, with all data
stored on LVM logical volumes inside that encrypted container.

I wanted to install Ubuntu 18.04 to new partitions inside the existing
LVM volume group, then migrate my data and settings across.

I was using the latest Ubuntu 18.04 desktop installer ISO image, which
I'd written to a USB stick using dd, and which had successfully passed
the check for defects.

I knew that there were multiple Ubuntu installers available, having used
them in the past, and I was aware that they offered different levels of
support for encryption and logical volumes.  Since I wanted a desktop
install, I was trying out the desktop installer first.


My steps to encounter the bug were as follows:

 1. Booted the installer and followed the installer steps until the section 
asking where I wanted to install.
 2. Noted that the installer apparently understood LUKS and LVM, since options 
for creating both of those seemed to be present.
 3. Selected the 'Something else' option, intending to try to activate my 
existing container.
 4. Highlighted the /dev/sda2 partition holding the existing LUKS container, 
and indicated that the installer should use this as a LUKS container.
 5. When prompted to enter passwords, I provided the existing LUKS password.

At this point, I still believed that the installer was about to activate
my existing LUKS partition.  The GUI support hadn't seemed amazing (for
example, the LUKS partition hadn't been autodetected), but this didn't
seem odd to me: re-using existing encrypted LVM containers isn't a
common use case, and I understand developers have a limited amount of
time available.


Most importantly for this bug: none of the installer text so far had indicated 
that these steps would *re-format* an existing LUKS container.  If they had, I 
would have realised that this wasn't the right choice of installer early enough 
to avoid data loss.


What happened next:

 6. I clicked OK to submit those passwords.  (At this point, I suspect the 
installer had permanently deleted all of my data.)
 7. When the disk partition view refreshed, I saw what I believed was my 
existing LUKS crypto device.
 8. I tried to indicate that my crypto device was being used as a physical 
volume for LVM, but the partition menu only included filesystem options such as 
'ext3' and 'ext4', so I concluded (too late) that my arrangement wasn't 
supported by the this installer.
 9. My system failed to reboot correctly, showing errors like 'evms_activate is 
not available' and 'Waiting for encrytpted source device ...'.

Even then, it wasn't clear to me that my existing LUKS volume had been
lost.  After some searching online, however, I found the following in
the cryptsetup FAQ:

"Some distribution installers offer to create LUKS containers in a way
that can be mistaken as activation of an existing container.  Creating a
new LUKS container on top of an existing one leads to permanent,
complete and irreversible data loss."

At this point it became clear what must have happened, and I stopped
trying to recover my data.


Note that I did not mind the lack of support in this installer: I just wish 
that in step 5 above it had been much clearer that it was only offering to 
create a brand new LUKS partition.

** Affects: ubiquity (Ubuntu)
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1779548

Title:
  Desktop installer lacks an adequate warning of permanent data loss if
  attempting to activate existing LUKS containers.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1779548/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 683240] Re: Installer loops forever retrying failed download

[Paul Whittaker]

** Attachment added: "Terminal output showing the install problem"
   https://bugs.launchpad.net/bugs/683240/+attachment/1750089/+files/buffer

** Visibility changed to: Public

-- 
Installer loops forever retrying failed download
https://bugs.launchpad.net/bugs/683240
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 683284] [NEW] Manual download method for hyperspec installer requires unusual file owner.

[Paul Whittaker]

Public bug reported:

Binary package hint: hyperspec

System:

 Description: Ubuntu 10.04.1 LTS
 Release: 10.04

Package:

 hyperspec 1.28

In order to work around a network problem which was preventing the
automated download of the HyperSpec tarball, I used the "download the
tarball manually" method for supplying the tarball to the install
script.

(In case it is not obvious, I'm referring to the following sequence of steps:
 1. Download tarball to right location.
 2. Install hyperspec package via aptitude.
 3. Package's postinst script finds the tarball downloaded in 1. and does not 
automatically download it from the Lispworks FTP site.
 4. HyperSpec files are installed in /usr/share/doc/hyperspec.)

As aptitude does not seem to ask the debconf questions (see bug #683240)
I read the installer's postinst file to find the path for the downloaded
file, /root/tmp/HyperSpec-7-0.tar.gz.  Whilst reading the postinst
script, I noticed that despite the file being looked for in root's home
directory, the postinst script explicitly checks that the file is *not*
owned by root, and that it is not in root's group.  If either check
fails, the file will be deleted and a new one downloaded.  This check is
done in the function "check_archive()" with the following test sequence:

[ -s "$FILE" -a ! -L "$FILE" -a ! -O "$FILE" -a ! -G "$FILE" ]

-O and -G check that "$FILE" (/root/tmp/HyperSpec-7-0.tar.gz) is owned
by the current effective UID and GID (which I believe should be root),
but then the ! then negates both of these tests.  As a result, after
downloading the tarball and putting it in the right place, I had to (as
root) chown it back to my normal user before it could be used by the
postinst script.  I also confirmed that a root-owned file at the same
location was deleted.

I would expect the file to be checked *for* root ownership (to make it
harder for users to pass malicious files to the installer), rather than
requiring it *not* to have root ownership.  There may well be a good
reason for this, and in that case I would instead expect there to be
some instructions about changing the file's owner, as most users in this
situation would use sudo to save the file in /root/tmp and therefore
give it the wrong owner.  Alternatively, if I've simply been using
aptitude as a user that wasn't envisaged (e.g. logged in as root vs. via
sudo) then a specific user should be tested for, rather than making
assumptions about what "the current" user is (or isn't).

** Affects: hyperspec (Ubuntu)
 Importance: Undecided
 Status: New

-- 
Manual download method for hyperspec installer requires unusual file owner.
https://bugs.launchpad.net/bugs/683284
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 683291] [NEW] Control files for hyperspec still refer to version 6.0.

[Paul Whittaker]

Public bug reported:

Binary package hint: hyperspec

Binary package hint: hyperspec

System:

 Description: Ubuntu 10.04.1 LTS
 Release: 10.04

Package:

 hyperspec 1.28

Whilst looking into bugs #683240 and #683284, I noticed that the
"templates" and "config" control files in the package refer to
HyperSpec-6-0.tar.gz, instead of to HyperSpec-7-0.tar.gz which is the
current version.  The postinst script, which actually downloads and
installs the file, does use the correct version, but the user guidance
as to what file the install script is checking for, for example, is out-
of-date.

** Affects: hyperspec (Ubuntu)
 Importance: Undecided
 Status: New

-- 
Control files for hyperspec still refer to version 6.0.
https://bugs.launchpad.net/bugs/683291
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 683240] Re: Installer loops forever retrying failed download

[Paul Whittaker]

Hi Marc,

On 30 November 2010 21:05, Marc Deslauriers
wrote:

> I have unmarked it as a security
> issue since this bug does not show evidence of allowing attackers to
> cross privilege boundaries nor directly cause loss of data/privacy.
>

I'd agree with that assessment.  I didn't know how paranoid we would be
about these things when I reported it, but if those are the criteria we're
using for security issues then yes, please leave it un-flagged.

Thanks for your swift response,

Best regards,

Paul.

-- 
Installer loops forever retrying failed download
https://bugs.launchpad.net/bugs/683240
You received this bug notification because you are a member of Ubuntu
Bugs, which is a direct subscriber.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 1441603] [NEW] The "CURLY LOGICAL OR/AND" glyphs across all Nanum fonts are swapped over.

[Paul Whittaker]

Public bug reported:

SYSTEM AND PACKAGE VERSIONS

Ubuntu Precise 12.04.5 LTS
Package fonts-nanum.
Tested with version 3.010-2, but appears still to be present in versions 
20131007-1 (Trusty) and 20140930-1 (Vivid).
The Firefox referred to below is version 37.0.1+build1-0ubuntu0.12.04.1.

SANITY CHECK

Just in case you are also having the same problem that I describe below, the 
following characters should all point in the SAME directions:
 - Pointing DOWN: ⋎ ∨ v (0x22CE, 0x2228, 0x76)
 - Pointing UP: ⋏ ∧ ^ (x022CF, 0x2227, 0x5e)
Those all display correctly for me in the text box I'm submitting this report 
from (in Firefox), but the first character of each group (the CURLY LOGICAL 
OR/AND character) is upside down for me in the submitted bug report.

If some characters above point in the wrong direction for you too, you
should also expect them to be displayed incorrectly in the bug
description below.

PROBLEM DESCRIPTION

The four font files

 - /usr/share/fonts/truetype/nanum/NanumGothic.ttf
 - /usr/share/fonts/truetype/nanum/NanumGothicBold.ttf
 - /usr/share/fonts/truetype/nanum/NanumMyeongjo.ttf
 - /usr/share/fonts/truetype/nanum/NanumMyeongjoBold.ttf

all have an error in their glyphs for Unicode codepoints 0x22CE and
0x22CF.

(I'm using http://www.unicode.org/charts/PDF/U2200.pdf here for
reference, which other fonts (e.g. DejaVu Sans) also agree with on these
codepoints.)

0x22CE (⋎) is the codepoint for the "CURLY LOGICAL OR" symbol (point
downwards) but the glyph given in the fonts above is the CURLY LOGICAL
AND  symbol (point upwards).

Similarly, 0x22CF (⋏) is the codepoint for the "CURLY LOGICAL AND"
symbol (point upwards) but is rendered by the fonts above with the CURLY
LOGICAL OR symbol (point downwards).

The NanumMyeongjo fonts are selected (at least for me and my colleagues)
by Firefox when it renders those glyphs, and these fonts are installed
by default in Ubuntu 12.04 Desktop.  The Firefox font choice is
unchanged even after we installed additional fonts (e.g. STIXGeneral)
which contain the glyphs in their correct form.  I have attached a
screen capture demonstrating the example.

As the glyphs are swapped (rather than simply missing) they still *look*
correct when viewed in context, and this causes subtle and quite serious
problems because it can literally change the meaning of the mathematical
formulae we use in our work.  (For example, these symbols are used for
rendering formal specifications in the Z language, and the meaning of a
specification can be fundamentally altered by swapping one of those
characters for the other.)

I realise that the primary purpose of this package is to provide Korean 
characters, not mathematical symbols, but since this bug is
 a) potentially quite serious, literally changing the meaning of displayed 
content,
 b) present when viewing content in the default browser on a default Ubuntu 
Desktop install, and
 c) presumably correctable by swapping over the two existing glyphs,
I hope it can be fixed quickly.

Many thanks,

Paul.

** Affects: fonts-nanum (Ubuntu)
 Importance: Undecided
 Status: New

** Attachment added: "Screenshot of swapped glyphs in a real web page."
   https://bugs.launchpad.net/bugs/1441603/+attachment/4369355/+files/and-or.png

** Description changed:

  SYSTEM AND PACKAGE VERSIONS
  
  Ubuntu Precise 12.04.5 LTS
  Package fonts-nanum.
  Tested with version 3.010-2, but appears still to be present in versions 
20131007-1 (Trusty) and 20140930-1 (Vivid).
  The Firefox referred to below is version 37.0.1+build1-0ubuntu0.12.04.1.
  
- 
  SANITY CHECK
  
- Just in case you are also having the same problem that I describe below, the 
following characters should all point in the same directions:
-  - Pointing down: ⋎ ∨ v (0x22CE, 0x2228, 0x76)
-  - Pointing up: ⋏ ∧ ^ (x022CF, 0x2227, 0x5e)
+ Just in case you are also having the same problem that I describe below, the 
following characters should all point in the SAME directions:
+  - Pointing down: ⋎ ∨ v (0x22CE, 0x2228, 0x76)
+  - Pointing up: ⋏ ∧ ^ (x022CF, 0x2227, 0x5e)
  Those all display correctly for me in the text box I'm submitting this report 
from (in Firefox), but the first character of each group (the CURLY LOGICAL 
OR/AND character) is upside down for me when displayed a plain HTML page.
  
  If some characters above point in the wrong direction for you too, you
  should also expect them to be displayed incorrectly in the bug
  description below.
  
- 
  PROBLEM DESCRIPTION
  
  The four font files
  
-  - /usr/share/fonts/truetype/nanum/NanumGothic.ttf
-  - /usr/share/fonts/truetype/nanum/NanumGothicBold.ttf
-  - /usr/share/fonts/truetype/nanum/NanumMyeongjo.ttf
-  - /usr/share/fonts/truetype/nanum/NanumMyeongjoBold.ttf
+  - /usr/share/fonts/truetype/nanum/NanumGothic.ttf
+  - /usr/share/fonts/truetype/nanum/NanumGothicBold.ttf
+  - /usr/share/fonts/truetype/nanum/NanumMyeongjo.ttf
+  - /usr/share/fonts/truetype/nanum/NanumMyeongjoBold.ttf
  
  all have an error 

[Bug 1441603] Re: The "CURLY LOGICAL OR/AND" glyphs across all Nanum fonts are swapped over.

[Paul Whittaker]

** Description changed:

  SYSTEM AND PACKAGE VERSIONS
  
  Ubuntu Precise 12.04.5 LTS
  Package fonts-nanum.
  Tested with version 3.010-2, but appears still to be present in versions 
20131007-1 (Trusty) and 20140930-1 (Vivid).
  The Firefox referred to below is version 37.0.1+build1-0ubuntu0.12.04.1.
  
  SANITY CHECK
  
  Just in case you are also having the same problem that I describe below, the 
following characters should all point in the SAME directions:
   - Pointing DOWN: ⋎ ∨ v (0x22CE, 0x2228, 0x76)
   - Pointing UP: ⋏ ∧ ^ (x022CF, 0x2227, 0x5e)
  Those all display correctly for me in the text box I'm submitting this report 
from (in Firefox), but the first character of each group (the CURLY LOGICAL 
OR/AND character) is upside down for me in the submitted bug report.
  
  If some characters above point in the wrong direction for you too, you
  should also expect them to be displayed incorrectly in the bug
  description below.
  
  PROBLEM DESCRIPTION
  
  The four font files
  
   - /usr/share/fonts/truetype/nanum/NanumGothic.ttf
   - /usr/share/fonts/truetype/nanum/NanumGothicBold.ttf
   - /usr/share/fonts/truetype/nanum/NanumMyeongjo.ttf
   - /usr/share/fonts/truetype/nanum/NanumMyeongjoBold.ttf
  
  all have an error in their glyphs for Unicode codepoints 0x22CE and
  0x22CF.
  
  (I'm using http://www.unicode.org/charts/PDF/U2200.pdf here for
  reference, which other fonts (e.g. DejaVu Sans) also agree with on these
  codepoints.)
  
  0x22CE (⋎) is the codepoint for the "CURLY LOGICAL OR" symbol (point
  downwards) but the glyph given in the fonts above is the CURLY LOGICAL
  AND  symbol (point upwards).
  
  Similarly, 0x22CF (⋏) is the codepoint for the "CURLY LOGICAL AND"
  symbol (point upwards) but is rendered by the fonts above with the CURLY
  LOGICAL OR symbol (point downwards).
  
  The NanumMyeongjo fonts are selected (at least for me and my colleagues)
  by Firefox when it renders those glyphs, and these fonts are installed
  by default in Ubuntu 12.04 Desktop.  The Firefox font choice is
  unchanged even after we installed additional fonts (e.g. STIXGeneral)
  which contain the glyphs in their correct form.  I have attached a
- screen capture demonstrating the example.
+ screen capture demonstrating an example of this "in the wild".
  
  As the glyphs are swapped (rather than simply missing) they still *look*
  correct when viewed in context, and this causes subtle and quite serious
  problems because it can literally change the meaning of the mathematical
  formulae we use in our work.  (For example, these symbols are used for
  rendering formal specifications in the Z language, and the meaning of a
  specification can be fundamentally altered by swapping one of those
  characters for the other.)
  
  I realise that the primary purpose of this package is to provide Korean 
characters, not mathematical symbols, but since this bug is
   a) potentially quite serious, literally changing the meaning of displayed 
content,
   b) present when viewing content in the default browser on a default Ubuntu 
Desktop install, and
   c) presumably correctable by swapping over the two existing glyphs,
  I hope it can be fixed quickly.
  
  Many thanks,
  
  Paul.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1441603

Title:
  The "CURLY LOGICAL OR/AND" glyphs across all Nanum fonts are swapped
  over.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/fonts-nanum/+bug/1441603/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 1226571] [NEW] Application looking in wrong place for install directory.

[Paul Whittaker]

Public bug reported:

Launching collatinus causes a dialog to appear saying:

   Capsam legere nequeo /usr/share/collatinus/lucretia.txt:
   No such file or directory.

Then, when the GUI has loaded, no translations of Latin words are
available.  For example, entering "amo" in the upper pane and selecting
"Lemmata > Omnia lemmatizare" produces "amo ?" in the lower pane.



The package is installing files to /usr/share/collatinus9/, not
/usr/share/collatinus/ (which does not exist).  If a symlink to
/usr/share/collatinus9 is created at /usr/share/collatinus, everything
works fine -- neither of the above problems occur, and the application
is, for example, able to automatically find, load, and provide
translations for "lucretia.txt".

** Affects: collatinus (Ubuntu)
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1226571

Title:
  Application looking in wrong place for install directory.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/collatinus/+bug/1226571/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 1226571] Re: Application looking in wrong place for install directory.

[Paul Whittaker]

Apologies; duplicate of #1091246.  Please delete if possible.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1226571

Title:
  Application looking in wrong place for install directory.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/collatinus/+bug/1226571/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 539339] Re: Exporting to PDF doesn't display text on other computers

[Paul Whittaker]

Still seeing this in version 0.4.6~pre20110721-1build1 (under Ubuntu
12.04.3, 64-bit).

Might this issue be related to the one reported (and fixed) at
https://bugs.archlinux.org/task/21693?  That mentions a patch made to
0.4.5 in late 2010; perhaps that hadn't made it into our upstream by mid
2011.  Can we adopt that work, if so?  Gaetan doesn't describe it as a
complex fix.

If we could fix this in time for 14.04 LTS, that would be very handy

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/539339

Title:
  Exporting to PDF doesn't display text on other computers

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/xournal/+bug/539339/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 550566] Re: Full-screen window prevents notification on other workspaces or monitors

[Paul Whittaker]

[I should of course have said "notify-send" in my second paragraph
above, not "notify-osd".]

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/550566

Title:
  Full-screen window prevents notification on other workspaces or
  monitors

To manage notifications about this bug go to:
https://bugs.launchpad.net/notify-osd/+bug/550566/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 550566] Re: Full-screen window prevents notification on other workspaces or monitors

[Paul Whittaker]

I can confirm that this bug renders notify-send effectively useless for
me, because I often have a terminal running full-screen (on my second
display) which prevents any notifications from appearing (on my primary
display).  I'm running 12.04.2 LTS with Unity 2D on a laptop with an
external monitor.

So, for example, in a (local) plug-in script I'm writing to display
notifications from an application, I've had to disable notify-osd
support, and instead look for an alternative mechanism.


Can anyone say whether this issue been fixed as a side effect of work on bug 
#1065425?  They seem related.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/550566

Title:
  Full-screen window prevents notification on other workspaces or
  monitors

To manage notifications about this bug go to:
https://bugs.launchpad.net/notify-osd/+bug/550566/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 998491] Re: window icon does not work with all dialogs

[Paul Whittaker]

According to strace here,

  zenity --info --window-icon /usr/share/zenity/zenity.png

makes a successful call to open() on the supplied filename
(/usr/share/zenity/zenity.png), but then also calls open() on -- and
displays -- the standard "white i on a blue disc" icon from
/usr/share/icons/Humanity/status/48/dialog-information.svg.

Changing the supplied filename to one more similar to the file displayed
above, e.g. to /usr/share/icons/Humanity/status/48/dialog-error.svg,
makes no difference to this behaviour.


Ubuntu 12.04.2 LTS
zenity 3.4.0-0ubuntu4

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/998491

Title:
  window icon does not work with all dialogs

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/zenity/+bug/998491/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 998491] Re: window icon does not work with all dialogs

[Paul Whittaker]

Further to the above, this also fails when supplying one of the short
arguments to --window-icon, as mentioned in the man page.

For me, passing in a value from the list given (error, info, question,
or warning) results in one of the files

  /usr/share/icons/Humanity/status/22/dialog-error.svg
  /usr/share/icons/Humanity/status/22/dialog-information.svg
  /usr/share/icons/Humanity/status/22/dialog-question.svg
  /usr/share/icons/Humanity/status/22/dialog-warning.svg

being accessed with open(), but then following that the file

  /usr/share/icons/Humanity/status/48/dialog-information.svg

is still loaded and displayed, as before.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/998491

Title:
  window icon does not work with all dialogs

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/zenity/+bug/998491/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 894384] Re: Impossible to force the launcher to hide

[Paul Whittaker]

Just to confirm that there's more than one person (still) affected by
this, and to provide a second use-case example.  Apologies if this bug
report is too old to be noticed any more.  Although the issue described
below doesn't prevent me from doing anything in the long term, it is
very irritating.

I'm using 12.04 64-bit with standard Unity (not 2D), and I have autohide
set (so the launcher isn't normally visible) to maximise usable screen
space on my laptop.

Sometimes I trigger the launcher by accident when trying to interact
with GUI elements awkwardly close to the left edge of the screen.  When
this happens the launcher pops out and covers the items I'm trying to
click on, at which point I have no choice but to wait... and do
nothing... until the autohide triggers and the launcher gets out of my
way again.  A common cause of this is aiming for the "back" button in
(maximised) Firefox and overshooting, but there are other times when I
need to point to several things in succession.

Ideally I would be able to bind some key to let me "cancel" opening the
launcher, so that I could keep this at hand when I'm having to work with
such applications/websites.  The faster this binding made the launcher
disappear, the better.

Using corner activation (as suggested in post #6 above) is a poor
workaround for me.  Since side activation is the default setting and I
work with (and support users on) a number of Ubuntu installations, it is
hard for me to unlearn its use.  So, when I do use corner activation on
my own machines, I inevitably only get it right after trying (and
failing) to use side activation a couple of times, after which I
remember why it isn't working and head up to the corner.  This removes
one inconvenience and introduces another.

If there isn't much appetite for adding this feature, I'd be (perhaps
even more) happy to learn of a keyboard workaround that used existing
features instead.  I had been hopeful that double-pressing Alt-F1 -- to
enter, and then exit, keyboard navigation mode -- might quickly hide the
launcher, but it does not.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/894384

Title:
  Impossible to force the launcher to hide

To manage notifications about this bug go to:
https://bugs.launchpad.net/unity/+bug/894384/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 387536] Re: Zenity should ignore markup with a flag

[Paul Whittaker]

Which package version includes that upstream bugfix?  I'm still being
bitten by this issue.  This is on a fully up-to-date 12.04.3 system.

$ apt-cache policy zenity
zenity:
  Installed: 3.4.0-0ubuntu4
  Candidate: 3.4.0-0ubuntu4

$ zenity --info --text 'x & y'

(zenity:14047): Gtk-WARNING **: Failed to set text from markup due to
error parsing markup: Error on line 1: Entity did not end with a
semicolon; most likely you used an ampersand character without intending
to start an entity - escape ampersand as &

(zenity:14047): Gtk-WARNING **: Failed to set text from markup due to
error parsing markup: Error on line 1: Entity did not end with a
semicolon; most likely you used an ampersand character without intending
to start an entity - escape ampersand as &

[(x) Information   ]
[--]
[ (info) All updates are complete. ]
[ (icon)   ]
[  ]
[[OK]  ]

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/387536

Title:
  Zenity should ignore markup with a flag

To manage notifications about this bug go to:
https://bugs.launchpad.net/zenity/+bug/387536/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 157497] Re: New windows stack up in top left corner

[Paul Whittaker]

I'm running metacity 1:2.34.1-1ubuntu11 (as part of Unity 2D) on Ubuntu
Precise amd64, and (despite post #12) I still see this issue.

I have a keybinding to launch an xterm, and if I press it four times
then I get four xterms opened in the top left of the screen, almost
entirely overlapping each other.  I have to manually drag the windows to
be able to see their content.  This is on an empty 1920x1200 desktop, so
there is plenty of free space.

Each window is very slightly offset from the previous one (1px to the
right, and 28px down) so that only their title bars are non-overlapping.

Could somebody advise as to whether I need to raise this as a new bug,
given that this ticket thinks it's been fixed?

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/157497

Title:
  New windows stack up in top left corner

To manage notifications about this bug go to:
https://bugs.launchpad.net/metacity/+bug/157497/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 846280] Re: Regression in ike 2.1.7, no longer creates IPsec connections

[Paul Whittaker]

@Mike, Geoffrey: Note that "2.1.7+dfsg-1.1" and "2.1.7+dfsg-1build1" are
not the same version.  (See versions for Oneiric and Precise at
http://packages.ubuntu.com/search?keywords=ike&searchon=names&exact=1&suite=all§ion=all
.)

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/846280

Title:
  Regression in ike 2.1.7,  no longer creates IPsec connections

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/ike/+bug/846280/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs