> Can you please elaborate on what high level diagnosis would you need?
Well, we'd need some idea of which spinlock is being contended for...
> I just ran DROP SCHEMA _<slony schema> CASCADE; and it spiked again, on a
> very low loaded box!!
That *might* mean that the problem is contention for SInvalLock, since
dropping a large schema would result in a lot of sinval traffic. Or
maybe it's something else. Do your spikes correspond to large DDL
changes?
If your platform has oprofile or DTrace or some such then getting an
execution profile with that type of tool would tell something.
regards, tom lane
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
No comments:
Post a Comment