I'm using pgsql 6.4.2 on Solaris via the JDBC driver. I have some queries running simultaneously in separate threads, and this occasionally generates a WaitOnLock error: ERROR: WaitOnLock: error on wakeup - Aborting this transaction The first thread begins a transaction, does a bunch of inserts, then commits the transaction and notifies the second thread of the inserted data. The second thread does a bunch of selects on the data that were just inserted. There is some overlap in the data: some of the data that the second thread selects might be being worked on by the first thread in a seubsequent transaction, which is how I expect this lock clash is being caused. What I'd like to know is how to go about debugging or working around this. Is there some way to get the backend to report more information (such as tables and rows) about the lock problem? What about increasing the timeout? -- Ari Halberstadt mailto:[EMAIL PROTECTED] <http://www.magiccookie.com/> PGP public key available at <http://www.magiccookie.com/pgpkey.txt>
