Issue for starting Tomcat app server
During app server startup, JVM reaches to the point where it is actioning Hazelcast clustering and then it just hangs and does not proceed.
Hangs at this spot:
2017-10-31 15:23:55,988 [ myserver] [ STANDARD] [ ] ( internal.mgmt.PRNodeImpl) INFO - Created provisional node: "myserver pega 2017-10-31 15:23:34.232 GMT", [PRVSNL-13806D80-CDAE-4DA2-82ED-4CD41773D337]
2017-10-31 15:23:55,988 [ myserver] [ STANDARD] [ ] ( internal.mgmt.PRNodeImpl) INFO - Starts joining cluster
2017-10-31 15:23:56,072 [ myserver] [ STANDARD] [ ] (cluster.internal.PRClusterImpl) INFO - No other cluster members detected in the database.
2017-10-31 15:23:56,072 [ myserver] [ STANDARD] [ ] (cluster.internal.PRClusterImpl) INFO - This node is starting first.
System console also reports this error which does not appear in logfile:
Error in `/amp/soft/jdk1.7.0_15/bin/java': malloc(): memory corruption: 0x00007ffa34230ba0
======= Backtrace: =========
======= Memory map: ========
00400000-00401000 r-xp 00000000 fd:1e 134249301 /amp/soft/jdk1.7.0_15/bin/java
00600000-00601000 rw-p 00000000 fd:1e 134249301 /amp/soft/jdk1.7.0_15/bin/java
023d4000-023f5000 rw-p 00000000 00:00 0 [heap]
5e0000000-6aaab0000 rw-p 00000000 00:00 0
Steps to Reproduce
- Bring down all JVMs.
- Truncate the pr_sys_statusnodes table and restart the system. It hangs on the statement "This node is starting first".
A third-party product issue is shown - the java7 interaction with the OS related to /libnet.so(Java_java_net_NetworkInterface_getAll call.
Make the following change to the operating environment:
Update JVM to a newer version (latest java8 release) to resolve the issue.
Published February 12, 2018 - Updated October 8, 2020
0% found this useful
Have a question? Get answers now.
Visit the Collaboration Center to ask questions, engage in discussions, share ideas, and help others.