No subject

<br></div>
So I&#39;d say that this is working exactly as it should be, since the join=
 order chosen by the planner has resulted in the shortest query time.<br></=
blockquote><div><br></div></div><div>A repeated explain analyze on the firs=
t query showed about 45s second time around (same). =A0However for the &amp=
;&amp; version,=A0second time around was just 2s (from 65s).</div>

<div><br></div><div>Saving 15s and losing caching seems like a bit of a fal=
se win no?</div><div><br></div></div>
</blockquote></div><br></div>

--00c09f9c9aacf83b9104876abff8--