We have been trying to solve a replication issue related to TEMPORARY tables.
Here's the issue Every three hours we run a processX (C++ compiled program) on the Master mysql that goes something like this: CREATE TEMPORARY TABLE tempTableA INSERT INTO tempTableA INSERT INTO normalTableB SELECT * FROM tempTableA processX terminates. Here's what we observe on the Slave side: ERROR: 1146 Table 'MainDatabase.tempTableA' does not exist Slave: error running query 'INSERT normalTableB SELECT * FROM tempTableA At this point the Slave stops replicating and we have a perl shell script running on the master every 15 minutes to determine if replication is active and restart replication if required. Our theory is that when processX terminates, the tempTableA disappears before the slave side has had a chance to pull all the replicated data across and access the information in tempTableA. N.B. We have process privileges set for the processX mysql username enabled on the slave. We are looking into 2 options at this point. 1. Figure out a reliable way to determine if the replication resulting from processX has completed. 2. Move all references of TEMPORARY tables to "scratch" tables on MainDatabase. We would rather use option 1 because we feel it is cleaner and would hopefully perturb our code less. We actually have multiple different processes each having this problem. Has anyone encountered this and come up with a solution? Thanks Peter Peter Jennings Public Communications Services (310) 954-3018 (310) 954-2159 fax [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
