On Wed, Mar 25, 2009 at 11:28:44AM -0400, Tom Lane wrote:
> What this looks like to me is an application problem. In each case
> you show, the only backends that actually *have* the lock are
> "<IDLE> in transaction", ie they are waiting for their clients to
> issue another command or close the transaction. The other queries
> won't be able to make any forward progress until those transactions
> complete and release their locks.
Tom,
Thank you. It's obvious now what a dunderhead I've been in misreading
the output of our own diagnostic tool. Thank you for the education.
We're going to sidestep the issue, for now, by going with Kevin's
suggestion to ditch the "FULL". If that doesn't work (such that I've
got to put the FULL back in and reintroduce the lockup), I'll get back
to my application, understand what's got it in knots, and untie it.
I apologize for blaming Postgres.
Best Regards,
Wayne Conrad