It is looking like a 'death of a thousand cuts' issue as you put it, and we do have a plan if that is the case :)
We have limits on the JDBC stuff in fusionreactor because we haven't really needed to use it. Our DBA is fairly certain that there isn't a set of, or a specific query that is causing a problem. All of our datasources are indeed wrapped. As far as I know, we don't have any DSN set to a client variable datasource. On Mon, Nov 29, 2010 at 1:57 PM, charlie arehart <charlie_li...@carehart.org > wrote: > Barry, I would just wonder if perhaps you’re suffering the “death of a > thousand cuts”, where it’s not that you have any (or many) long-running (or > even moderately long-running queries), but instead maybe lots (and lots) of > small-running requests. Just something to consider. > > And as for using FR, I would recommend that you do NOT limit what is logged > (and be aware that there is a separate setting on the JDBC settings page for > first what is shown in the interface, and what is logged). And of course > it’s critical that you wrap **all** datasources being used—not just those > that you think “might be a problem”, as there could be requests coming in > (again, perhaps from bots) going against a DB (and app) you think is > otherwise insignificant. Finally, you should be sure to wrap (with FR) > especially any DSN set to be a client variable datasource. Again, I find > this is often a huge resource-suck, with ever page request generating > multiple i/o’s to the cglobal table. > > > > /charlie > -- You received this message because you are subscribed to the Google Groups "cfaussie" group. To post to this group, send email to cfaus...@googlegroups.com. To unsubscribe from this group, send email to cfaussie+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/cfaussie?hl=en.
