What to do if Landmark is crashing with core dumps

Here are a few mitigation procedures to use when Landmark applications are crashing or performing very slowly and throwing core dumps on the server.

 

First, check the environment.properties file.  Validate that file is pointing to the correct version/location of Java.  Sometimes, Java is updated on the server, and the environment.properties file is not changed accordingly.  It’s not a guaranteed issue, but should be validated and resolved if found.

 

Secondly, make sure there is adequate heap space provided to the jvm.  Here are some examples for heavy processing:

 

com.lawson.was.nodeagent.jvm-init-mem=1024

com.lawson.was.nodeagent.jvm-max-mem=3072

com.lawson.was.dmgr.jvm-max-mem=4096

com.lawson.was.dmgr.jvm-init-mem=2048

com.lawson.was.<app Server>.jvm-init-mem=3072

com.lawson.was.<app server>.jvm-max-mem=12228

 

àThese values will be deployed every time a CU is applied, so you don’t have to keep updating them

 

Be sure to recycle the Landmark application server once these changes are made.

 

Note that clustering Landmark WebSphere would not do any good.  Landmark applications behave very differently from Lawson applications, and it would only waste resources to vertically scale.