RE: [cas-dev] identical cas servers differnet behavior

...@gmail.com] Sent: Monday, June 09, 2014 2:26 PM To: cas-dev@lists.jasig.org Subject: Re: [cas-dev] identical cas servers differnet behavior > So I have two CAS servers running the exact same CAS server war (4.0) > and tomcat 7.0.52. What Tomcat connector component are you using? T

Re: [cas-dev] identical cas servers differnet behavior

> So I have two CAS servers running the exact same CAS server war (4.0) and > tomcat 7.0.52. What Tomcat connector component are you using? The NIO connectors may use "direct buffer" space that is stored off the heap; so the heap reports nothing excessive while your overall process memory grows. T

Re: [cas-dev] identical cas servers differnet behavior

We may need to look at server socket cleaning mechanism too. Sometimes, Stale connections bring CAS Down with OOM, Lava ᐧ [image: lava kafle on about.me] lava kafle about.me/lavakafle Lava Kafle Ms by Research in Computer Science Kathmandu University cell: 98412243

Re: [cas-dev] identical cas servers differnet behavior

Are the usage patterns for the two servers identical? Also, what is your Tomcat configuration (i.e. heap size, etc.) On Mon, Jun 2, 2014 at 10:45 AM, Nancy Snoke wrote: > So I have two CAS servers running the exact same CAS server war (4.0) and > tomcat 7.0.52. I will refer to these as Server

[cas-dev] identical cas servers differnet behavior

So I have two CAS servers running the exact same CAS server war (4.0) and tomcat 7.0.52. I will refer to these as Server1 and Server2 for simplicity. Server1 regularly gets an OOM exception. It's caught and handled. The dynaTrace shows its committed memory staying at 639.03 MB. The used memo