Re: BUG #7626: Query planner never returns, uses 100% CPU - Mailing list pgsql-bugs

From Tom Lane
Subject Re: BUG #7626: Query planner never returns, uses 100% CPU
Date
Msg-id 21286.1351533628@sss.pgh.pa.us
Whole thread Raw
In response to BUG #7626: Query planner never returns, uses 100% CPU  (brian@omniti.com)
List pgsql-bugs
brian@omniti.com writes:
> Running this causes the thread to use 100% CPU and never returns (or at
> least not for longer than my patience runs out).
> This returns just fine on 8.4.1 and on 9.2beta.

Hmm ... seems to be a consequence of commit
3b8968f25232ad09001bf35ab4cc59f5a501193e.  I wrote in that patch

+        * This looks expensive, but in practical cases there won't be very many
+        * distinct sets of outer rels to consider.

but evidently that was too optimistic :-(.  Will need to think about
suitable heuristics for limiting the sets of outer relations we
consider building parameterized paths with.

            regards, tom lane

pgsql-bugs by date:

Previous
From: Bill MacArthur
Date:
Subject: Re: BUG #7626: Query planner never returns, uses 100% CPU
Next
From: "Kevin Grittner"
Date:
Subject: Re: BUG #7623: Inconsistency on transaction isolation documentation