On 6/12/2014 12:57 PM, Christopher Nielsen wrote:
With a problem like this, I am not exactly positive how to proceed. I
am really looking forward to hearing your thoughts, and opinions, if
you can share them.
keep a shell session open with a superuser role (eg, postgres), and next
time its hammered, try SELECT * FROM PG_STAT_ACTIVITY
look for things that are waiting = t and such.
also, if the server is 100% CPU bound, see what processes are using all
that CPU. ps -auxw should show the sql commands in addition to
'postgres' or 'postmaster'. you can compare the PID shown in ps output
with the pid (or procpid if an older release) in pg_stat_activity, too.
--
john r pierce 37N 122W
somewhere on the middle of the left coast
