So I changed the 1000*60 to 60000, and this morning, the site
responds properly.
Of course, this is kind of marginal science since a collection of
invidious daemons could
have lined up to hit the site every hour overnight. But assuming
that these changes
are responsible for the correct response, I'm unclear about what it
is in my situation
which makes them necessary. Thoughts?
(BTW, if the correct behavior persists, I'll let you all know after a
week.)
--Ken
On Jul 16, 2008, at 9:50 AM, Hassan Schroeder wrote:
On Wed, Jul 16, 2008 at 5:42 AM, Ken Bowen <[EMAIL PROTECTED]> wrote:
On Jul 16, 2008, at 8:20 AM, David Smith wrote:
I don't think "1000*60" is a valid expression for
timeBetweenEvictionRunsMillis -- at least I've never tried putting
math
expressions where an integer constant should be. It should be set
to 60000.
+1 -- that didn't look right to me, either, but I didn't try it.
All connections are /supposed/ to be closed when the interaction if
finished, but there still could be one slipping through.
Suggestions: install LambdaProbe, which provides a nice graphical
pool monitor; makes a connection leak *very* obvious.
(<http://www.lambdaprobe.org/>)
Also, change that value above, restart, and tail -f your mysql query
log and tomcat logs together, so you can confirm validation queries
are being sent, and see when they stop showing up in the log.
HTH,
--
Hassan Schroeder ------------------------ [EMAIL PROTECTED]
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
