I think the reason most people don't worry about it in production is that they don't do reloading or hot redeployment in production. Not much anyway. Development and testing are done on other servers and the production servers are updated much less frequently. Or they're running in a clustered configuration where the app server can be restarted without interupting the users.
----- Original Message ----- From: "Leonardo Quijano Vincenzi" <[EMAIL PROTECTED]> To: "Tapestry users" <[email protected]> Sent: Thursday, October 27, 2005 12:00 PM Subject: Re: AW: Memory / Caching Issues > ausias vives wrote: > > It's a known issue that reloading an application in > > tomcat causes, in the short or mid term, OOME errors. > > In your case I woudn't bother so much, and I'd focuse > > in checking that the application runs correctly > > without using the reload facility. It's very useful to > > write some load tests and using httpunit or webunit. > > > I'm wondering why this problem isn't a critical issue that has all > commiters extremely worried. OOMEs - well, at least IMO!! - are like > threading errors - race conditions, etc.. they are something that should > get everyone worried. I'd expect my applications to fail in a consistent > way, not when 'heap runs out'. > > Of course, I understand it's more worrisome if it happens in production > environments rather than only in development. But my point is that if it > happens in development, can you guarantee it won't happen in production? > > -- > Ing. Leonardo Quijano Vincenzi > Director Técnico > DTQ Software > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
