l0b0's questions

After following the official guide for fixing a bug in Ubuntu I'm stuck at how to actually submit some existing work (in the form of local Git commits on an "ubuntu/devel" branch for a "main" pool* package). The guide seems to assume familiarity with Quilt and Debian packaging tools+infrastructure, since the first command to start actual work, edit-patch 99-new-patch, fails with

No patchsystem detected, cannot create new patch (no dpatch/quilt/cdbs?)

which seems to be misleading. It also seems pretty complicated to map to version control systems (1, 2, 3).

So: Is there a complete and straightforward set of steps to submitting some Git commits for review?

* Is "pool" the right name? None of apt, apt-cache or dpkg seem to tell.

I've found a bazillion references to get a list of the installed packages, but how can I print a single list of all known packages with their "package state" (not-installed, installed, half-installed etc. as defined by dpkg) in a shell, ideally like this:

awk   not-installed
bash  installed
cc    half-installed
[...]

dpkg --get-selections and dpkg --list only lists installed packages.

dpkg --get-selections '.' does not work.

apt-cache dump does not print whether packages are installed, and also prints a lot of irrelevant stuff.

I'm using Travis CI, which is running Ubuntu 12.04 LTS Server Edition 64 bit with for example dpkg-query 1.16.1.2.

Update: This issue will not be answered conclusively; I have moved to another distro and have not observed this problem since. I was never able to fix it with the insightful answers available at the time, but your fuel efficiency may vary (YMMV).

crontab -e and crontab -l work just fine:

$ crontab -l | grep -v '^#'
* * * * * /usr/bin/env
* * * * * echo 'Hello from crontab'

However, I see two messages like this every minute in /var/log/syslog:

Mon DD hh:mm:01 username CRON[PID]: Permission denied

So the crontab is being read, but somehow it can't execute anything at all (of course I verified the commands when logged in as the same user). Any idea why?

/etc/cron.allow and /etc/cron.deny do not exist.

crontab is set group setuid:

$ stat --format '%A %U %G' /usr/bin/crontab
-rwxr-sr-x root crontab

The crontabs directory seems to have the right permissions:

$ stat --format '%A %U %G' /var/spool/cron/crontabs
drwx-wx--T root crontab

The crontab itself is owned by me (not surprisingly, since I'm able to edit it):

$ sudo stat --format '%A %U %G' /var/spool/cron/crontabs/$USER
-rw------- username crontab

I am not a member of the crontab group.

These lines appear in /var/log/auth.log every minute (thanks @Alaa):

Mon DD hh:mm:01 username CRON[1752]: pam_unix(cron:session): session opened for user username by (uid=0)
Mon DD hh:mm:01 username CRON[1752]: PAM bad jump in stack

Maybe PAM is broken? pam-auth-update (thanks @coteyr) lists all of these, and all of them are enabled:

• Unix authentication
• GNOME Keyring Daemon - Login keyring management
• eCryptfs Key/Mount Management
• ConsoleKit Session Management
• Inheritable Capabilities Management

Can any of them be safely disabled? I'm not using any encrypted filesystems.

Based on a Debian bug entry I tried running debconf-show libpam-runtime, and I got the following error message:

debconf: DbDriver "passwords" warning: could not open /var/cache/debconf/passwords.dat: Permission denied

The contents of /etc/pam.d/cron:

# The PAM configuration file for the cron daemon

@include common-auth

# Read environment variables from pam_env's default files, /etc/environment
# and /etc/security/pam_env.conf.
session       required   pam_env.so

# In addition, read system locale information
session       required   pam_env.so envfile=/etc/default/locale

@include common-account
@include common-session-noninteractive 

# Sets up user limits, please define limits for cron tasks
# through /etc/security/limits.conf
session    required   pam_limits.so

session [success=1 default=ignore] pam_succeed_if.so service in cron quiet use_uid

The files mentioned (/etc/environment, pam_env.so, /etc/default/locale, pam_limits.so, pam_succeed_if.so) are all readable by my user.

On another host with Ubuntu 13.04, with the same user crontab, no /etc/cron.{allow,deny}, same permissions as above, and not being a member of the crontab group, it works just fine (logs the commands but not the output in /var/log/syslog).

By changing the first crontab line:

* * * * * /usr/bin/env >/tmp/env.log 2>&1

and checking that /tmp is world writeable:

$ sudo -u nobody touch /tmp/test
$ ls /tmp/test
/tmp/test
$ ls -ld /tmp
drwxrwxrwt 15 root root 12288 May 27 10:18 /tmp

I've verified that the crontab commands are not run at all: The Permission denied messages still show up in /var/log/syslog, but /tmp/env.log is not created.

Based on a random listing of /etc/pam.d settings I found the following discrepancies:

$ grep '^[^#]' /etc/pam.d/sshd 
@include common-auth
account    required     pam_nologin.so
@include common-account
@include common-session
session    optional     pam_motd.so # [1]
session    optional     pam_mail.so standard noenv # [1]
session    required     pam_limits.so
session    required     pam_env.so # [1]
session    required     pam_env.so user_readenv=1 envfile=/etc/default/locale
@include common-password
$ grep '^[^#]' /etc/pam.d/common-session
session [default=1]         pam_permit.so
session requisite           pam_deny.so
session required            pam_permit.so
session optional            pam_umask.so
session required    pam_unix.so 
session optional    pam_ecryptfs.so unwrap
session optional            pam_ck_connector.so nox11
$ grep '^[^#]' /etc/pam.d/common-account
account [success=1 new_authtok_reqd=done default=ignore]    pam_unix.so 
account requisite           pam_deny.so
account required            pam_permit.so
$ grep '^[^#]' /etc/pam.d/common-session-noninteractive 
session [default=1]         pam_permit.so
session requisite           pam_deny.so
session required            pam_permit.so
session optional            pam_umask.so
session required    pam_unix.so 
session optional    pam_ecryptfs.so unwrap

PAM packages installed:

$ dpkg --get-selections | grep --invert-match deinstall | cut --fields 1 | grep pam
libpam-cap
libpam-ck-connector
libpam-gnome-keyring
libpam-modules
libpam-modules-bin
libpam-runtime
libpam0g
python-pam

I tried reinstalling these - didn't help:

$ sudo apt-get install --reinstall $(dpkg --get-selections | grep --invert-match deinstall | cut --fields 1 | grep pam)

I can't purge and then reinstall these because of unmet dependencies.

tl;dr Is there a generic guide to set up GNOME background services to work with Awesome WM on any Ubuntu?

The official guide doesn't seem to have been updated since Ubuntu 11.10. Another guide is slightly more up to date, but after upgrading to 12.10:

1. GNOME panels on the top and bottom of the first monitor. They shouldn't be there.
2. Awesome WM background image.
3. Awesome panels on the second monitor.
4. Awesome controls work (Mod4+number to go to tag, etc.).

After upgrading to 13.04 today it's even worse:

1. GNOME panels on the top and bottom of the first monitor. Still shouldn't be there.
2. No panels on the second monitor.
3. Awesome WM background image.
4. feh error message when logging in even though it is installed. Sorry, can't copy it here since I can't reproduce (see below).
5. Windows are placed at X=0 and Y=something like -10 (the window title bars are about 90% offscreen). This means I can't access the menu in the top panel without closing all open applications (since they can't be moved by dragging with either Mod4 or Meta held down).
6. Some GNOME keyboard shortcuts work (Ctrl+Meta+Del, Ctrl+Meta+t), but not others (Meta+F1).
7. No Awesome keyboard or mouse controls work (Mod4+number, Mod4+Enter, Mod4+mouse drag).

Points 3 and 4 only happened the first time after rebooting into 13.04; on the second login I get the Ubuntu background image and no feh error message.

The command I cobbled together to setup "Awesome GNOME" based on the guide (reformatted for readability):

sudo sh -c '
apt-get install awesome awesome-extra gnome-session-fallback notification-daemon && \
printf "%s\n" "[Desktop Entry]" "Version=1.0" "Type=Application" "Name=awesome" "TryExec=awesome" "Exec=awesome" | tee /usr/share/applications/awesome.desktop && \
printf "%s\n" "[Desktop Entry]" "Name=Gnome with Awesome" "Comment=Gnome with Awesome as window manager" "TryExec=gnome-session" "Exec=gnome-session --session=awesome" "Type=Application" | tee /usr/share/xsessions/gnome-awesome.desktop && \
printf "%s\n" "[GNOME Session]" "Name=Awesome" "RequiredComponents=gnome-panel;gnome-settings-daemon;" "RequiredProviders=windowmanager;notifications;" "DefaultProvider-windowmanager=awesome" "DefaultProvider-notifications=notification-daemon" | tee /usr/share/gnome-session/sessions/awesome.session'

@RolandSommer's answer fixed most of the issues. In the meantime I worked around a crash by commenting out all awful.util.spawn entries.

There's an (as far as I can tell) undocumented feature of gnome-control-center which can be used to show a specific panel directly. For example, gnome-control-center sound will display the "Sound" panel. I've been trying to figure out what the "Keyboard Layout" panel is called, without success:

gnome-control-center keyboard # Just the "Typing" and "Shortcuts" tabs
gnome-control-center layout # Not found
gnome-control-center keyboardlayout # Not found
gnome-control-center 'keyboard layout' # Not found

Where can I find the list of panel names to use with this command?

In Synaptic it's possible to select a bunch of packages for installation (or reinstallation), then just click File → Generate package download script to get a set of commands to download the corresponding deb files. This is useful for transferring missing packages to an offline system for installation, and it also resolves dependencies.

Unfortunately, as far as I can tell the process has some drawbacks:

1. The online system has no way of knowing which of the dependencies are installed on the offline system (unless they're both somehow in sync). In the worst case this means several roundtrips to download all the necessary files.
2. synaptic is not installed by default on a new system, so I'd have to use two roundtrips if I wanted to generate the package download script on the offline system - Once for the Synaptic deb file, and again with the files downloaded by the generated script.

apt-get download is not an alternative since it doesn't resolve dependencies.

Is there some way to

1. generate a package download script with the packages available on the Ubuntu live CD, or
2. transfer the packaging state on the offline machine to the online one and create a package download script with the latest upstream state?

The latter option would be preferred since it would also get the latest versions of each package.

I have had an AMD Radeon HD 6950 (i.e., Cayman series) for a couple years now, and I have tried a lot of combinations of drivers and settings with terrible results. I'm completely at a loss as to how to proceed. The open source driver has much better 2D performance, but it offloads all OpenGL rendering to the CPU.

What I've tried so far:

• All the latest stable Ubuntu releases in the period, plus one Linux Mint release.
• All the latest stable AMD Catalyst Proprietary Display Drivers, and currently 13.1.
• The unofficial wiki installation instructions for every Ubuntu version and the semi-official Ubuntu instructions.
• All the tips and tweaks I could find for Minecraft (Optifine, reducing settings to minimum), VLC (postprocessing at minimum, rendering at native video size), Catalyst Control Center (flipped every lever in there) and X11 (some binary toggles I can no longer remember).

Results:

• Typically 13-15 FPS in Minecraft, 30 max (100+ in Windows with the same driver version).
• Around 10 FPS in Team Fortress 2 using the official Steam client.
• Choppy video playback, in Flash and with VLC.
• CPU use goes through the roof when rendering video (150% for 1080p on YouTube in Chromium, 100% for 1080p H264 in VLC).
• glxgears shows 12.5 FPS when maximized.
• fgl_glxgears shows 10 FPS when maximized.

Hardware details from lshw:

• Motherboard ASUS P6X58D-E
• CPU Intel Core i7 CPU 950 @ 3.07GHz (never overclocked; 64 bit)
• 6 GB RAM
• Video card product "Cayman PRO [Radeon HD 6950]", vendor "Hynix Semiconductor (Hyundai Electronics)"
• 2 x 1920x1200 monitors, both connected with HDMI.

I feel I must be missing something absolutely fundamental here. Is there no accelerated support for anything on 64-bit architectures? Does a dual monitor completely mess up the driver?

$ fglrxinfo
display: :0  screen: 0
OpenGL vendor string: Advanced Micro Devices, Inc.
OpenGL renderer string: AMD Radeon HD 6900 Series
OpenGL version string: 4.2.11995 Compatibility Profile Context
$ glxinfo | grep 'direct rendering'
direct rendering: Yes

I am currently using the open source driver, with the following results:

• Full frame rate and low CPU load when playing 1080p video.
• Black screen (but music in the background) in Team Fortress 2.
• Similar performance in Minecraft as the Catalyst driver. In hindsight obvious, since both end up offloading the rendering to the CPU.

My /var/log/Xorg.0.log after upgrading to AMD Catalyst 13.1. Some possibly important lines:

(WW) Falling back to old probe method for fglrx
(WW) fglrx: No matching Device section for instance (BusID PCI:0@3:0:1) found

The generated xorg.conf. The disabled "monitor" 0-DFP9 is actually an A/V receiver, which sometimes confuses the monitor drivers when turned on/off (but not in Windows). All three "monitor" devices are connected with HDMI.

Edit: Chris Carter's suggestion to use the xorg-edgers PPA (Catalyst 13.1) resulted in some improvement, but still pretty bad performance overall:

• Minecraft stabilizes at 13-17 FPS, but at least the CPU load is "only" at 45-60%.
• Still 150% CPU use for 1080p video rendering on YouTube in Chromium.
• Massive improvement for 1080p H264 in VLC: 40-50% CPU use and no visible jitter
• glxgears performance about doubled to 25-30 FPS when maximized.
• fgl_glxgears still at ~10 FPS when maximized.

GNOME Terminal 3.4.1.1 in Ubuntu 12.04 has an IMO weird mouse scroll behavior: A single mouse wheel motion scrolls ceiling($LINES / 10) lines, rather than a set number. That is, if the terminal window is between 31 and 40 lines, it will scroll four lines at a time. If it's 100 lines, it will scroll 10 lines at a time. How can I force it to work like other applications, scrolling N lines regardless of the window height?

The Ubuntu security notices web page has an option to show only notices for a single release. Is there a way to get notices for a single release as a feed, without setting it up myself? That is, is there an official feed containing all and only Ubuntu 12.04 (for example) security notices?

The end goal is a tool which warns when any of a set of packages (not necessarily the ones installed on a system) for a single Ubuntu release are affected by a security notice.

I'm using gLabels to create some small pieces of paper to give away to people - Just two lines of about 15-20 characters each. One of the important parts of the design is that there are no "o" ("oh") characters, but there are "0" ("zero") characters in an all-lowercase URL. Therefore I need a non-monospaced (and very legible in print) font where "0" cannot be confused with "o" even without reference.

The closest thing I've found so far is Andale Mono and Inconsolata , but they are both monospace, and as such not very easy to read in print.

Alternatively, to find a font myself, is there some way to preview all fonts in the Ubuntu repositories without installing them?

VT100 terminals have a relatively simple way to save terminal contents with escape codes to a file using Meta-p:

XTerm.VT100.printAttributes: 2
XTerm.VT100.printerCommand: cat - > file
XTerm.VT100.translations: #override Meta <KeyPress> P: print() \n

I've looked around quite a bit, and can't find a solution for this to work in the default GNOME Terminal + XTerm + Bash.

I do not want just an image with the contents of the window.

It looks like GNOME Terminal might eventually get a fix for this, but there's no information about which version it will be included in, and it's been almost three years since the last update...

The /home/users directory on my host is an NFS mount connected to a server which is not in my control. When deleting files in $HOME (/home/users/foo) with Nautilus I get the following message:

Cannot move file to trash, do you want to delete immediately?

The file "..." cannot be moved to the trash.

Is there some way to enable trash for this mount point? I have write access to the partition, but even after creating the expected /home/users/.Trash-$(id -u) and setting the sticky bit, Nautilus still fails to use the trash.

When reading this question and the Trash Specification I found that $HOME/.local/share/Trash is the standard location (unless $XDG_DATA_HOME is set), and it looks like Nautilus is only half-aware of this: It does indeed write a $HOME/.local/share/Trash/info/[filename]<.number>.trashinfo every time I try to delete a file, but the files are all empty and I still keep getting the error message.

After sudo apt-get install awesome awesome-extra and logging into it (with no previous configuration that I know of), there's no menu entry for Firefox. All my other stuff is there - Mutt, Meld, jEdit, BOINC, etc., only Firefox is missing. I have no ~/.config/awesome/ directory, so there should be no override in place.

I've run sudo download-mibs and downloaded two additional MIB files (INFORMANT-OS-MIB, WTCS-MIB):

$ ls -lA /var/lib/mibs/site
total 456
-rw-r--r-- 1 root root 460726 2012-05-22 16:52 INFORMANT-OS-MIB
-rw-r--r-- 1 root root   1088 2012-05-23 08:55 WTCS-MIB

And created a symlink in the standard MIB directory:

$ ls -lA /usr/share/mibs/site
lrwxrwxrwx 1 root root 18 2012-05-23 08:58 /usr/share/mibs/site -> /var/lib/mibs/site

But it's not able to translate this OID:

$ snmptranslate -Of 1.3.6.1.4.1.9600.1.22.11.1.1.19
.iso.org.dod.internet.private.enterprises.9600.1.22.11.1.1.19

In order to research what caused the installation of a given package, I'd like to get a list of packages which depend on that package. I couldn't find anything obvious in man dpkg.

After installing Catalyst 11.12 and studying instructions (1, 2) to enable hardware acceleration I ended up with the following command:

sudo apt-get install xvba-va-driver libva-glx1 vainfo && \
vainfo && \
vlc --verbose 2 --ffmpeg-hw video.1080p.x264.mkv

The output confirms that it's using the VA API:

avcodec decoder: Using VA API version 0.32 for hardware decoding.

But even after a reboot the video still looks choppy. There's no vlc output to indicate that the codec is having trouble, or that it's skipping frames. Any idea what could be the problem?

Using vlc 1.1.12 with default settings.

Also tried (in sequence):

1. sudo apt-get install libva-glx1
2. sudo apt-get install --reinstall libgl1-mesa-glx:i386 libgl1-mesa-glx:amd64 libgl1-mesa-dri:i386 libgl1-mesa-dri:amd64 xserver-xorg-core
3. Rebooting

No luck yet.

Specs from /proc/cpuinfo : Intel Core i7 CPU 950 @ 3.07GHz

Firefox claims it's the default browser (Edit -> Preferences -> Advanced -> Check Now), Preferred Application also lists it as the default, but x-www-browser runs Chrome. Is there really more than one setting for the Ubuntu default browser?

Twice now an upgrade of the proprietary ATI Catalyst driver has been a major pain, even with the help of excellent guides.

These commands all run successfully (as in $? -eq 0):

sudo apt-get remove --purge fglrx fglrx_* fglrx-amdcccle* fglrx-dev* \
xorg-driver-fglrx
sudo apt-get install build-essential cdbs fakeroot dh-make debhelper \
debconf libstdc++6 dkms libqtgui4 wget execstack libelfg0 \
dh-modaliases # Already installed
sudo apt-get install ia32-libs # Ditto
sudo sh ./ati-driver-installer-11-7-x86.x86_64.run --buildpkg Ubuntu/natty
sudo dpkg -i fglrx*.deb

After a restart it crashed during the graphical boot phase. Restarted in safe mode, and restored the old configuration:

sudo mv /etc/X11/xorg.conf{.fglrx-N,}

Now it boots, and goes straight into the correct video mode, but the version number shown in CCC still says 11.5!

Tried uninstalling the drivers, and noticed some warnings:

dpkg: warning: while removing fglrx, directory '/usr/lib32/fglrx' not empty so not removed.
dpkg: warning: while removing fglrx, directory '/usr/lib/fglrx/etc/ati' not empty so not removed.
dpkg: warning: while removing fglrx, directory '/usr/lib/fglrx/etc' not empty so not removed.
dpkg: warning: while removing fglrx, directory '/usr/lib/fglrx' not empty so not removed.

Tried to install without generating the packages:

sudo sh .../ati-driver-installer-11-7-x86.x86_64.run

It says the packages are still installed! Verified with the following command:

$ apt-cache pkgnames --installed | grep fglrx
fglrx
fglrx-dev
fglrx-amdcccle

What can I do at this point? I even tried to remove the /usr/lib32/fglrx and /usr/lib/fglrx directories, but that didn't help.

Thinking about what would be useful to display in a screensaver, I realized that it would be nice to know how long the screensaver has been running. gnome-screensaver-command --time takes care of that. Now, how can I create a simple screensaver based on this value (plus perhaps some ps output), updating it at regular intervals?

While a form control like <select> or <input> is focused the arrow keys, Page Up, Page Down, Home and End all navigate inside the contents of that control rather than the web page itself. This is very useful, but makes it difficult to navigate web pages by keyboard - After focusing a form control, you'll have to press Tab or Shift-Tab until you get to a URL to be able to navigate with the keyboard again. If there is no URL on the web page (shock, horror) the mouse seems to be the only way out.

So is there a keyboard shortcut to give focus back to the page itself?