I returned from the 2009 annual ACM Reflections | Projections conference held at the University of Illinois at Urbana-Champaign. I was asked to speak at the conference about Git.

I really enjoyed attending the conference, and not just because I learned how to play Beer Pong at the staff party. The organizers did a really good job of putting on a great event – even if the line ups for food were a bit long, but it was free food for starving students, so what would you expect. I got to meet awesome people like Bram Moolenaar, the author of Vim, Alexis Ohanian, of reddit fame, and Ryan North, the guy behind Dinosaur Comics. I was also in awe of the facilities the CS people have at UIUC.

I added a new host to bacula today. That resulted in:

    13-Oct 16:58 bacula-dir JobId 1026: Fatal error: File daemon at "oxygen:9102" rejected Hello command
    13-Oct 16:58 bacula-dir JobId 1026: Error: Bacula bacula-dir 2.4.4 (28Dec08): 13-Oct-2009 16:58:39

After looking around on the web and coming up with nothing, I noticed the version difference. The new host happened to run version 3.x.y of bacula-fd, unlike my director that runs 2.4.y. Apparently bacula doesn’t support the director being an older version than the client.

I got playing with git log and ended up creating this alias:

[alias]
    lg = log --graph --pretty=format:'%Cred%h%Creset -%C(yellow)%d%Creset %s %Cgreen(%cr)%Creset' --abbrev-commit --date=relative

Which adds a git lg command that is a prettier version of git log --oneline.

I participated in the virtualization panel at last night’s meeting of my local Linux Users Group. Pictures from last night are up on Richard Guy Briggs’ website.

I’ve put up the slides for the virtualization primer talk, and also the source document (created in Inkscape) if you wish to use the slides for something. I use negative to convert the .svg image to .png and .pdf files.

One of my old(er) USB-connected disks started to make a noise. So, it’s time to replace it.

Here are the steps I took to create an encrypted USB volume that I can attach to my laptop.

This is not as smooth as [rooting the G1]{rooting-g1}, and comes no where close as [my second attempt at rooting the G1]{simple-rooting-g1} (ie the easy way).

WARNING: I’ve said it before, but this time I want to stress it… this may brick your phone! Until I figured things out and found the right pages I had a non-booting phone.

I finally got my [Cupcake to build]{baking-cupcake}. The next step is to try to build something far tastier… naturally that something has to be the Cyanogen Cupcake/Donut blend.

Grab a coffee, sit back, and read on to find out how to build your own CyanogenMod Android ROM from Cyanogen’s Git repositories…

I’ve previously written about [how to root an android phone]{rooting-g1}, and mentioned that there was a better way which I had not tried yet. Well, I did try it today.

The procedure written by Zinx from ZenThought exploits a recently found bug in the Linux kernel (see CVE-2009-2692).

We were hanging out by the pool on my daughter’s birthday. I thought I would be an ever-so-funny-guy and push my wife into the pool. She thought she would be equally funny and pull me in with her. So far so good, all fun and games.

As soon as I landed in the water, and remembered that I had my [G1]{tag/g1} in my pocket, I instantly felt 100% less funny.

This story has a good ending, as I was able to get the phone back into perfectly working condition…

This is a follow up to my [pretty function tracing]{pretty-function-tracing} article. I base this work on the code presented there.

Some one asked me how to get the gcc -finstrument-functions feature working. If you don’t know this flag will modify the entry and exit to/from each function to make a call out to a special set of functions used for tracing.

While I’ve read about this feature, I never actually tried it. So here is what I learned…