I accidentally wrecked the pg_class table in a database by leaving off the where clause of an update statement when setting the reltriggers back to the count from pg_trigger after a bulk data load. I can no longer connect to the database, although it does start up without error. running psql [dbname] causes extreme cpu load, and the command just hangs. Memory usage continues climbing until a oom killer crashes everything. I presume that my blunder is causing some sort of infinite loop during postmaster startup for the spawned process. Is there a way I can get direct access to the file for pg_class and reset the reltriggers to the correct values without a full dbinit process?


---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster

Reply via email to