hi, we have following situation: pg 9.3.11 on ubuntu. we have master and slave. the db is large-ish, but we're removing *most* of its data from all across the tables, and lots of tables too.
while we're doing it, sometimes, we get LOTS of processes, but only on slave, never on master, that spend long time in BIND state. same queries ran on master run without any problem, and are fast.
any clues on where to start diagnosing it?
So far we've: 1. ruled out IO problems (enough io both in terms of bandwidth and iops) 2. ruled out memory (plenty) 3. vacuumed all tables, including system ones, with analyze 4. did pg_repack on most of the tables to remove bloat.
we are in process of doing the same operation to series of servers, so i'm not sure if the criteria on all of them are the same, but perhaps you can think of something we didn't?
It is looking like spinlock issue.
try to look there by "perf top"
Pavel
depesz -- The best thing about modern society is how easy it is to avoid contact with it. http://depesz.com/