the corner office : tech blog

I've cracked open Programming in Scala again, after a long hiatus during which No Programming Took Place At Home Because I Was Doing Other Important Stuff. Watch this space, yet again...

Perf issues in Eclipse 4.x:

• Juno Performance

• Something is really broken with Eclipse foundation

• Why C++ Is Not "Back"

Hardy Heron goes EOL in a month's time and I'm not looking forward to the server upgrade effort that'll entail. 2008 feels like aaaages ago (I guess it was). For all my grumbling about Ubuntu, I'm still rather grateful that a company I pay no money to has made my life so easy for so many years.

A frightening-as-you-slowly-begin-to-understand-the-severity run-down of recent Ruby on Rails security problems:

This Sounds Like A #$()#%ing Disaster

That is primarily because this is a #$()#%ing disaster.

Damien Katz: The Unreasonable Effectiveness of C and a follow up.

Very interesting read. This point in particular got me to thinking about Java circa 2013:

C++ is also a complicated mess, so when you adopt C++ for its libraries and community, you have to take the good with the bad and weird to get the benefits. And there is a lot of disagreement what constitutes bad or weird. Your sane subset of the language is very likely to be at odds with others ideas of a sane subset.

One day I'll learn that when prompted to upgrade something, the safest option is almost always 'no' (especially, I might perhaps unfairly add, if it has anything to do with Oracle). The latest version of VirtualBox has broken VBoxVMService and it's now impossible to run the service and the VirtualBox GUI at the same time. If the service is running then trying to open the GUI now gives you an error

Failed to create the VirtualBox COM object. The application will now terminate.

with detail

Callee RC: 
CO_E_SERVER_EXEC_FAILURE (0x80080005)

A real pity.

Details in the VBoxVMService wiki: Running VirtualBox GUI together with VBoxVmService.

Update: I reverted to VirtualBox 4.1, but VMs wouldn't start, moaning about VERR_SSM_UNSUPPORTED_DATA_UNIT_VERSION and the like. A bit of googling and the solution is just to right-click on each VM and choose 'Discard Saved State'.

• dz: How to tame java GC pauses

  Using advises above and HotSpot's CMS, I was able to keep GC pause on 32Gb Oracle Coherence storage node below 150ms on 8 core server.

• Attila Szegedi: Everything I Ever Learned about JVM Performance Tuning @twitter

It's a pity - with very few exceptions I've managed to avoid having gaps in my monthly archive list, but September just went *poof*.

Ironically, I wasn't entirely idle... the blog software isn't something I'm inclined to devote too much time to, but it needed a few things doing, and to make it all happen I needed an easier way to set things up and drop in changes. So I set up one VM wired up much like the one at my hosting provider. Then I set up another VM (Crunchbang, for the heck of it) which thinks that my domain names point to the first VM, so I could test redirects and virtual hosts etc etc. And then some scripts to push the right bits and bobs on the right place on the prep and real hosts.

Then a tweaked CSS file which is something that's needed doing for a while, and which has been made much easier with Chromium and Firefox's built-in web dev tools. I haven't done anything resembling web development in many many many years, and while Firebug was pretty good back then, the world has come a long way. It was fun spending a few nights tweaking the CSS and experimenting, but it did remind me why it's not my day job. Anyhow, two open source web browsers and they're both awesome. So here it is: voila!

This case really shows only one thing. The patent system, especially when it comes to software, is utterly and totally broken.