Doubt's log
| February 2003 | ||||||
| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
| 1 | ||||||
| 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| 9 | 10 | 11 | 12 | 13 | 14 | 15 |
| 16 | 17 | 18 | 19 | 20 | 21 | 22 |
| 23 | 24 | 25 | 26 | 27 | 28 | |
| Jan Mar | ||||||
Microsoft Bloggers:
Benjamin J. J. Voigt
Better Living Through Software
Console.WriteLine("Hello World");
Denatonium benzoate: Kent Sharkey"s blog
Don Box's Spoutlet
Harry Pierson's DevHawk Weblog
Incessant Ramblings
InkBlog : The Random Musings of David Weller
jimmygrewal.com
Joe Beda's EightyPercent.Net
John Lambert
Karsten Januszewski's UDDI Web Log
meta-douglasp
michaelw.net
min jeschwad
Objective
Peter Drayton's Radio Weblog
Pushing the Envelope
Rob Fahrni, at the core.
ScottGu's Blog
Tantek Çelik
Technovangelist (Matt Williams) WebLog
Y. B. Normal
Dogfooding and Showstoppers. Figuring out what has happened in situations like this is both an art and a science. Since the server farm is located somewhere other than Redmond, you have to investigate everything remotely. All you have to figure out what went wrong are the perfmon logs and a raw memory process dump before it terminated (since these production servers had no debuggers attached). Analyzing them is a bit like a TV episode of CSI -- involving forensic science (what perfmon counter showed what and when) and a bit of detective work (trying to figure out what happened over the course of a day's run -- in particular in the last few minutes and seconds before things started going bad). [ScottGu's blog]
9:52:44 AM
comments ==
9:52:44 AM