Thu Jan 13 00:02:52 PST 2005
- Previous message: [Slony1-general] Reason for minXID logic?
- Next message: [Slony1-general] current slony1 limitations
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Just a note from the poor sod who's databases were being brutalized by this behavior. We found several long-lived clients that were '<IDLE> in transaction'; fixing the offending code (a perl web application that stupidly held its connections open, AutoCommit off; and a daemon that was hanging, and thus failing to close its connection) and restarting said processes has reduced our load from ~2.5 at peak to ~0.6. The sl_log_1 table went from ~720,000 rows to under 5000, which my gut says is correct, given our transaction volume. A hearty thanks to Josh, Christoper, Jan, and everybody else who helped us out. Best, jfb -- James Felix Black Programmer, iParadigms LLC (510) 287-9720 x 250
- Previous message: [Slony1-general] Reason for minXID logic?
- Next message: [Slony1-general] current slony1 limitations
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-general mailing list