15 Rules of IT

1) If a computer or system is working on Monday, something has gone wrong and you just don’t know about it.

2) When you don’t know what you’re doing, document it.

3) Computer errors must be reproduce-able, they should fail the same way each time.

4) First call tech support, then panic.

5) Experience is directly proportional to the number of systems/computers ruined.

6) Always keep a backup of your data, and hope everyone else does the same.

7) To write a program really well, have your wife/husband/Mother/Father test it.

8) If you can’t get the system to match the Statement of Work (SOW), redo the SOW.

9) In case you experience doubt, make it sound convincing.

10) Do not believe in miracles–rely on them.

11) The technical term is H.O.S.E.D. (Hardware Or Software Error Detected), use it often and amaze your friends.

12) When it starts working; and hopefully it will, you fixed it in case anyone asks.

13) No troubleshooting experience is a complete failure. At least it can serve as a negative example.

14) Any expensive piece of software will break before any use can be made of it.

15) Team work is essential, it allows you to blame someone else.


This is an original post, feel free to reproduce as long as you give credit.

Tech Ed 2007 – Orlando – notes from the field

Its that time of year again. The masses gather to learn from the best, talk to the experts, network with others. This year I am not only speaking at the conference, I am also helping at the TCL Yellow Section – Windows Server 2008 Failover Clustering POD. Here is a sample of what people want to know:

Q.  Can I upgrade from Windows Server 2003 to 2008?

A. NO! Migration is the only and best way to go.


Q. Will Exchange Server 2007 run on Windows Server 2008 Beta 3?

A. Exchange Server 2007 does not run on the Beta of Windows Server 2008. Exchange will run on Windows Server 2003 and the final (RTM) copy of Windows Server 2008.


Q. Can I cluster Virtual Machines running Microsoft Virtual Server 2005 R2?

A. Yes you can!


Q. What is really new about Failover Clustering in Windows Server 2008?

A. Great question, here is some highlights:

  • No service account – now uses a local system account

  • No support for Parallel Attached SCSI (PAS), this allows for better hardware control and failover

  • 16 nodes if you use a 64-bit edition of Windows Server 2008

  • Support for GPT disks, yes larger than 2TB disks are now supported natively

  • Validate – ensure your systems are cluster ready and help you follow best practices

  • 4 Quorum models, no more single point of failure for the quorum

    • Majority quorum model

    • Majority of Nodes

    • Witness disk

    • File Share Witness

  • Improved IPSec, no more 7 minute timeout when failed over!

  • Stretched clusters can have nodes in different network segments with various speeds

  • Cluster Migration Tool – to help you migrate 🙂

  • MMC 3.0 snap-in, no longer a seperate application

  • Setup/install 2003 was 23 steps, now 3 steps. Yes I said 3 steps