KDE Tip – Middle-Click to Close Tabs in Konqueror

To close tabs in Konqueror with a middle-click, run this command:

  • kwriteconfig --file konquerorrc --group "FMSettings" --key MouseMiddleClickClosesTab --type bool true

More cool tips can be found at the KDE Wiki : Hidden configuration page.

UPDATED: For KDE 3.5.8, remove the quotes around FMSettings.

  • kwriteconfig --file konquerorrc --group FMSettings --key MouseMiddleClickClosesTab --type bool true

How Long to Crack WPA?

Trolling through websites, I found an interesting link on this aircrack-ng tutorial page – a calculator to estimate the amount of time it would take to brute-force the WPA keys, based on a variety of parameters.

The calculator is HERE – you will need to turn on JavaScript if you are using No-Script (IE users, just ignore that tip… ^__^) .

I do not know how accurate or valid it is, but I imagine the aircrack-ng folks wouldn’t have linked to it if they thought it was junk….

Kismet and D-Link DWL-G122 USB WLAN Adapter

I decided to try out kismet today. I had a D-Link USB wireless adapter handy (a DWL-G122 version B1), but no driver. Some quick Googling turned up the RaLink Linux support web site, but I had trouble compiling the drivers there (I tried both the RT2501USB(RT2571W/RT2671) and the RT2500USB(RT2571/RT2572) (source Code)). So instead I tried the latest beta rt2570 driver from the rt2x00.serialmonkey.com web site:

  1. Untar after downloading.
  2. Go to the Module directory.
  3. “make”
  4. “make install” (gotta be root).
  5. “modprobe rt2570”
  6. “iwconfig” should then show it as rausb0.

From there, the README file included in the tarball has instructions for having the hardware activated on boot.

Of course, you need to have iwconfig (urpmi network-tools to get it on Mandriva)…

I then installed kismet via urpmi, and after a little trolling through man pages, and finding the config file (hint – “updatedb” followed by “locate kismet”), I edited the /etc/kismet.conf file and changed the following lines:

  • # User to setid to (should be your normal user)
    suiduser=<put in a non-root username here>
  • # source=sourcetype,interface,name[,initialchannel]
  • #source=prism54g,eth0,eth0 (for my Planex Intersil card)
    #source=rt2570,rausb0,DWL-G122-B1 (my first guess – shoulda read the README below first)
    source=rt2500,rausb0,rausb0 (correct setting)

I found the right sources in the README doc (/usr/share/doc/kismet-3.1.060413/README) – READ THIS FILE FIRST!

Running kismet without the suid user set failed (it needs to be started as root, but then will drop down to the specified user’s privileges for security reasons). Trying with the correct suid user and the source set to the prism54 worked, but of course, my wireless connection was now gone. I changed to use the DLink interface, and kismet came up, discovered my wireless network (which stilled worked), and even revealed that my neighbors do not appear to have wireless.

Of course, our walls are pretty thick, too….

Next, I’ll get a notebook and try a little war-strolling in my neighborhood. Everyone needs a hobby, right?

PS – Seriously, read the kismet README first – it has the correct sources and list the serialmonkey site as the place to get the driver.

PPS – lsusb output for the D-Link:

  • Bus 005 Device 003: ID 2001:3c00 D-Link Corp. [hex] DWL-G122 802.11g rev. B1 [ralink]

Firefox Bolt-On – Tab Mix Plus

I tried this out, and while it has a *lot* of configuration settings (making it easy to cause your browser to misbehave, but nothing you cannot undo), it is also extremely cool. I was able to ditch several other addons (fewer is better, in my book), all associated with tab preferences and opening closed tabs, and tweak Firefox until it works just the way I want it. I especially like the saved sessions and embedded progress indicators on each tab. Tabs can also be highlighted if unread. There is plenty more to fool around with, so go nuts.

Hats off to the Tab Mix Plus developers! Get it HERE.

Microsoft Patch Badness – Who Is To Blame?

Talk about timing. I just did a write-up on product quality from Microsoft, and right around the corner is a nasty zero-day bug that even affects Vista. Today, after Microsoft releases a patch, it is revealed that the patch BREAKS THINGS (specifically, Realtek audio and Ethernet devices seem to be most affected – they stop working). Question is, is it Microsoft’s fault? The easy answer is yes, but how can Microsoft be expected to know the inner workings of all of the driver manufacturers? Are the third party vendors not also capable of bad code, and of being constrained by the same pressures that likely result in less-than-the-best code from Microsoft?

Microsoft has long argued that many “bugs” in their software are really caused by third-party manufacturers – this problem with their patch seems to highlight that point rather nicely. Of course, further investigation is the only thing that will really show if it is a good patch stepping on a bad driver, or vice versa….

It also seems to highlight a strength of open source, as the chances of this happening if both driver and patch were open are greatly reduced.

Zero-day for Microsoft – Three Months Warning Not Enough?

Today, SANS went to InfoCon YELLOW, due to an exploit involving how Microsoft Windows OS’s handle malformed ANI files. It seems to affect nearly any OS Microsoft makes, so long as they are at the latest patch levels. This includes Vista, and includes IE7. IE7 on Vista in Protected Mode seems to offer protection.  The exploit is silent, and allows arbitrary code execution.

According to the article on SANS, Microsoft was warned about this back in December by Determina.  Yup, three months ago (this is being generous, since it is now April).

Way to go, Microsoft.