Sorry but i didn't understand that If this is a bug from postgres version then how the same query will be worked if i recreated the tables and with same version of postgres.
This could be related to the plans that are chosen by the optimizer and this specific error might be coming only when a certain plan is chosen. When you recreated the table, the planner may have chosen a different plan because the stats were different, and hence you did not see the issue again. If you can reproduce the problem, it will be a good idea to run EXPLAIN to see the query plan.
