On Tue, Nov 11, 2014 at 9:11 AM, Josh Berkus <j...@agliodbs.com> wrote:
> On 11/10/2014 01:40 PM, Alvaro Herrera wrote: > > Josh Berkus wrote: > >> All, > >> > >> pg version: 9.3.5 > >> RHEL 6.5 > >> 128GB/32 cores > >> Configured with shared_buffers=16GB > >> Java/Tomcat/JDBC application > >> > >> Server has an issue that whenever we get lock waits (transaction lock > >> waits, usually on an FK dependancy) lasting over a minute or more than > >> 10 at once, *all* queries on the server slow to a crawl, taking 100X to > >> 400X normal execution times. > > > > Current FK checking makes you wait if the referenced tuple is modified > > on any indexed column, not just those that are actually used in > > foreign keys. Maybe this case would be sped up if we optimized that. > > > >> * This applies even to queries which are against other databases, so > >> it's not purely a lock blocking issue. > > > > Oh. > > Yeah, I think this is more likely a problem with the general lock table > and shared_buffers than anything to do with actual lock-blocks. > > Any chance you can run 'perf record -a' on it? Cheers, Jeff
