I upgraded a busy pgpool-II instance to 3.0 last week and very quickly had to back it out to 2.3.3 due to this issue. I tried 3.0.1 today and it does the same thing.
Certain queries seem to be causing the backend to remain in the PARSE state. The pgpool process itself leaves the socket open and opens new ones, doing additional work. Eventually the backend runs out of available processes and everything comes to a halt. The backend servers in this case are running PostgreSQL 8.3.11 on Linux. Mostly I'm wondering if this is a known issue and if there's a knob somewhere that can fix it. I don't see anything in the archives about it though. Thanks for any assistance. _______________________________________________ Pgpool-general mailing list [email protected] http://pgfoundry.org/mailman/listinfo/pgpool-general
