Lazy debugging

Like all lazy good sysadmins if I have to do something more than twice I’ll start to think about either:

• Delegating to one of my team(!)
  or
• Scripting it.

A while back I read a blog post by Dmitry Vostokov about scripting the analysis of mini crash dumps but never got round to trying it until today.  The only difference from Dmitry’s post is that I took the script and then saved it as minidump_analysis.txt on my notebook.  I then fired up WinDbg and opened a minidump for analysis.  Using the following you can then call your script to be run against the loaded dumpfile

This will then run all of the commands in the script against the minidump and output it to a text file which will be the name of the dumpfile with .log appended to it.

Okay, so its not going to actually do the analysis for me but at least I don’t have to remember to run all the different commands.