Tweet
Hi guys,
I'm currently developing some Mirth destinations to be deployed in a hospital environment and have encountered an error when deploying my channels for testing:
Here are the specs of my development machine:
Windows XP SP2
Microsoft SQL Server 2005 Express
Java RE 1.5
Mirth 1.3.2
AMD Opteron Dual Core 1.8ghz CPU
512MB RAM
It is my suspicion that the memory consumed during the deployment of channels is not sufficiently free'd back to the memory pool once a channel stops deployment. I came to this conclusion when I noticed this error only occurred after frequent deployments of the same channel I was developing and testing and it is also well known that the automatic garbage collection is not the best in the java virtual machine.
Has anyone else had this problem? if so, how did you get around it? Also, does this problem occur during on-site usage of mirth? in which the mirth server has to remain online for long periods of time without rebooting and a chance for a full flush of the jvm memory space.
Would it be possible to invoke the System.gc / manual garbage collection command within javascript inside mirth to counteract this problem?
I'm currently developing some Mirth destinations to be deployed in a hospital environment and have encountered an error when deploying my channels for testing:
Code:
WARN 2007-02-09 15:21:27,890 [SslListener0-1] org.mortbay.jetty.servlet.ServletHandler: Error for /channels java.lang.OutOfMemoryError: Java heap space
Windows XP SP2
Microsoft SQL Server 2005 Express
Java RE 1.5
Mirth 1.3.2
AMD Opteron Dual Core 1.8ghz CPU
512MB RAM
It is my suspicion that the memory consumed during the deployment of channels is not sufficiently free'd back to the memory pool once a channel stops deployment. I came to this conclusion when I noticed this error only occurred after frequent deployments of the same channel I was developing and testing and it is also well known that the automatic garbage collection is not the best in the java virtual machine.
Has anyone else had this problem? if so, how did you get around it? Also, does this problem occur during on-site usage of mirth? in which the mirth server has to remain online for long periods of time without rebooting and a chance for a full flush of the jvm memory space.
Would it be possible to invoke the System.gc / manual garbage collection command within javascript inside mirth to counteract this problem?
Comment