Reference to - BUG #18349: ERROR: invalid DSA memory alloc request size 1811939328, CONTEXT: parallel worker
Craig Milhiser <craig@milhiser.com>
From: Craig Milhiser <craig@milhiser.com>
To: pgsql-bugs@lists.postgresql.org
Date: 2024-09-22T11:58:41Z
Lists: pgsql-bugs
Has the referenced bug in this discussion[1] been released? The discussion mentions it was fixed. I read the release notes but do not recognize this fix as documented. I do not want to hijack that thread. I am using v16.3 - AWS Aurora. I have opened a support case with AWS also. I have 4 queries that get this error message. The symptoms are the same - thousands of temp files are created before the error is returned. Just because I am getting the same error/symptoms does not mean it is the same problem. If the fix has been released I can check against the AWS Aurora version with their support staff. I investigated one of the queries so far. When I turn off parallel execution this query completes. I am working on a reproducible example to send to the community. The query is sensitive to the values in the predicates. Most of the time these queries work. So far, I have not been able to get an MRE but still working on it. Thanks Craig [1] https://www.postgresql.org/message-id/18349-83d33dd3d0c855c3%40postgresql.org
Commits
-
Fix extreme skew detection in Parallel Hash Join.
- 98c7c7152d2d 18.0 landed
- 4ac5d33a8b08 17.1 landed
- 53edc948580f 16.5 landed
- 1831545ca186 15.9 landed
- 20d94899414a 14.14 landed
- 45329466fdc9 13.17 landed
- 4fa80a6d7d4b 12.21 landed