.NET CLR Data counters are wrong - Official

"Original counters (in V1.0 and V1.1) reported wrong numbers in almost all scenarios."

http://www.gotdotnet.com/team/changeinfo/Backwards1.1to2.0/default.aspx#00000123

"We’ve created a new set of Performance Counters in V2.0 that are specific to each provider. We're obsoleting all the counters under the .NET CLR Data category..."

[via Phil]

How on earth did this go unnoticed past the testers?

Published Sunday, June 26, 2005 12:50 PM by RoyOsherove
Filed under: ,

Comments

Sunday, June 26, 2005 6:08 AM by Eran Sandler

# re: .NET CLR Data counters are wrong - Official

I think its best to put these things into perspective.

Not ALL counters were wrong. I know of a few accumulating counters that never got decremented when they should have.

Counters such as the various Generation Heap Sizes, bytes promoted to each heap, % Time spent in GC and some other GC and memory related counter did show correct numbers (and its even verifyable).
Sunday, June 26, 2005 8:09 AM by Wallym

# re: .NET CLR Data counters are wrong - Official

As I remember, a lot of people knew that the performance counters in 1.x didn't all work properly. As I see it, this is an issue that could slip by because performance counters aren't typically worked with in the main stream app development area.
Sunday, June 26, 2005 8:49 AM by Roy Osherove

# re: .NET CLR Data counters are wrong - Official

Wally: I could see how one or two would have broke, but *most* of them is simply weird.
Sunday, June 26, 2005 1:55 PM by jayson knight

# re: .NET CLR Data counters are wrong - Official

This explains a lot of some whacky behavior I saw on a project I worked on a couple of years ago; we never could track down some of the craziness we were seeing (mainly db connections/pooling). Good find.