Basics of Debugging

For some time now, we get some weird Netapp problems that are difficult to debug in our environment. We run NIS, LDAP, and active directory to form a super hybrid environment to support Linux, Mac, Solaris, and Windows computers.

Awhile back, we had a problem of Office Applications were really slow to save which is documented on my old vox blog. So for a couple diagnostics on that one:

reproducing the error on different windows users using different protocols to mount the data (nfs, cifs, smb) wireshark/netbt on the Windows host to capture packets

Now we are getting another office problem but this time on the Mac (both Tiger a nd Leopard). I’m going to run through the same steps as last time and see what I’ll find out this time. I’ll probably post this adventure in another post though.

So basically, a run down of steps when trying to diagnose a problem.

  • reproduce the problem – Being able to successfully reproduce the problem is key in diagnosing it. How else will you be able to fix it if it doesn’;t happen reliably?
  • determine if O/S specific – This is can greatly narrow down whether it will be a common fix for several O/S’s or just the one you can reproduce the problem on.
  • determine if software specific – Using different software can help to see if it’s just one specific vendor that causing the problem or does it happen using all

    different software. If the problem is with Adobe Reader, try Foxit Reader to see if it happens there too.

  • network related – Learning how to use wireshark or snoop is a critical

    element to analyze what information is passing between the computers.

    You never know if there is an error message being sent in the clear!

This entry was posted in tips and tagged . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *