Excerpts from Tom Lane's message of mar jul 26 13:43:08 -0400 2011:
> Uh, yes. I've lost count of the number of times I've seen people hit
> one-instruction-wide race condition windows, SIGSEGV crash based on
> accessing only one byte past the valid data structure, etc etc.
I think you need a better locking protocol on that counter of yours.
--
Álvaro Herrera <alvherre@commandprompt.com>
The PostgreSQL Company - Command Prompt, Inc.
PostgreSQL Replication, Consulting, Custom Development, 24x7 support