On 11/03/14 16:03, Tom Lane wrote: > [ thinks for awhile... ] Oh, I know what's happening: your sort is so > large that it's being constrained by the MaxAllocSize limit on the tuple > pointer array. This has been fixed in HEAD, but it's not yet in any > shipping release. According to the log entry for commit > 263865a48973767ce8ed7b7788059a38a24a9f37, the previous limit on the number > of tuples that could be sorted in memory was INT_MAX/48 or about 44 > million; I've not done the arithmetic to check that, but it seems about > right seeing that you're having trouble with 75 million. Thanks, that makes sense. BTW, I solved my problem w/o that sort. I was just curious what happened here. Torsten -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
