At my workplace, we have machines out in the field which were having problems that caused them to require frequent reboots.  The problem turned out to be SQL Server seg-faulting, but that's not what this entry is about.  What I wanted to talk about here were some free tools you have already or can get easily that can tell you all about your system.  One of the great features of these tools is that they are command line tools which can have their output piped to a file.  Once you have a file on disk, what you do with it is your business.  We tend to FTP and e-mail them around.

You already have these!
SystemInfo - go to a DOS Prompt and type 'systeminfo' to get a huge amount of valuable information, including processor types, physical and virtual memory, and a list of OS hotfixes which have been applied.  Have you got two "identical" machines behaving differently?  Compare the systeminfo output from both to ensure they are actually identical.  NOTE: hyperthreaded processors show up as two processors

Net statistics server/workstation - see how much network traffic your server and workstation services are dealing with, including sessions started and print jobs spooled!  Just go to a DOS prompt and type 'net statistics server' or 'net statistics workstation'

Netstat - run 'netstat -a' from a DOS prompt to see what TCP ports are opened and listening.  This was useful for a problem with an antivirus program blocking all network traffic to and from a machine due to an over-abundance of opened ports.

Download this for free!
Uptime.exe - download this tool and run it with the /s command to get some really great statistics, including a history of shutdown and boot times over a period of months.  You'll see how many times that machine has bluescreened as well as an availablility statistic.  Here is a sample of the output analysis AFTER a huge list of reboot times:
                  Since 1/12/2005:

                             System Availability: 99.9930%
                                    Total Uptime: 34d 22h:2m:17s
                                  Total Downtime: 0d 0h:3m:31s
                                   Total Reboots: 4
                       Mean Time Between Reboots: 8.73 days
                               Total Bluescreens: 0

                     Notes:
                     1/12/2005 is the earliest date in the event log where
                     sufficient information is recorded to calculate availability.

I wanted to put a referral to Deepak Sharma's blog entry about this same topic.

-- Matt Ranlett