webMethods

Hi,
thanks to support:
I am facing an issue that I can’t start Enterprise Manager on UMs due to GC out of memory error.
WrapperManager Error: java.lang.OutOfMemoryError: GC overhead limit exceeded

The JVM has run out of memory. Restarting JVM.

WrapperManager Error: at java.net.SocketInputStream.read(SocketInputStream.java:222)

WrapperManager Error: at java.io.DataInputStream.readByte(DataInputStream.java:265)

WrapperManager Error: at org.tanukisoftware.wrapper.WrapperManager.handleBackend(WrapperManager.java:5362)

WrapperManager Error: at org.tanukisoftware.wrapper.WrapperManager.run(WrapperManager.java:5732)

WrapperManager Error: at java.lang.Thread.run(Thread.java:745)

WrapperManager Error: The backend was closed unexpectedly. Restart to resync with the Wrapper.

Exception in thread “Wrapper-Connection” java.lang.OutOfMemoryError: GC overhead limit exceeded

The JVM has run out of memory. Restarting JVM.

[Sun Feb 21 16:52:52 EET 2021],Session closing due to following Error: java.lang.OutOfMemoryError: GC overhead limit exceeded - 54441 => nSessionAttributes:conns=4/[nsp://10.11.251.182:9000/, nsp://fe80:0:0:0:250:56ff:feb1:7b17%2:9000/, nhp://10.11.251.182:9000/, nhp://fe80:0:0:0:250:56ff:feb1:7b17%2:9000/]

The JVM has run out of memory. Restarting JVM.

[Sun Feb 21 16:53:21 EET 2021],GC overhead limit exceeded

java.lang.OutOfMemoryError: GC overhead limit exceeded

The JVM has run out of memory. Restarting JVM.

[Sun Feb 21 16:53:23 EET 2021],Session closing due to following Error: java.lang.OutOfMemoryError: GC overhead limit exceeded - 51178 => nSessionAttributes:conns=2/[nsp://eaium2:9001/, nhp://eaium2:9001/]

The JVM has run out of memory. Restarting JVM.

Shutdown failed: Timed out waiting for signal from JVM.

JVM did not exit on request, termination requested.

JVM received a signal SIGKILL (9).

JVM process is gone.

JVM exited after being requested to terminate.

JVM Restarts disabled. Shutting down.

Hi Kareem,

please check under profiles/UM/config folder if there is a setting in either wrapper.conf or custom_wrapper.conf to specify the size of the Java HeapSize. Keep in mind that the box needs to have enough physical RAM to accomodate the increased Java Memory settings to avoid swapping.

Eventually there is another config file under the UM-directory itself.

Can you check what is causing UM to need more memory than before?
I assume that this was working previously and is failing now as I doubt that UM won´t start at all after initial installation.

Regards,
Holger

Hi Kareem,

Can you share the version and fix level of your Universal Messaging server and your Enterprise Manager.
The memory configuration is located in a file called Admin_Tools_Common.conf under <installDir>/UniversalMessaging/java/<instance>/bin (e.g. <installDir>/UniversalMessaging/java/umserver/bin/). Could you please share that file, along with a file called nenterprisemgr.conf under the same directory.
Can you also share the entire log file.
It’s odd that it would crash during startup - by default it runs with a maximum of 512 MB heap size, which is usually more than enough for an Enterprise Manager connected to a single UM realm server. If we can’t spot anything obviously wrong we may need to change the settings so that it generates a heap dump when it crashes with an OOME.

Thanks,
Stefan

Hi Stefan,

instead of neneterprisemgr.conf I think you meant enterprisemgr.conf.

Regards,
Holger

Surprising as it may sound, it’s really “nenterprisemgr.conf”
The leading ‘n’ is a convention that most users are used to by now, so I doubt it’s going to change soon…

Cheers,
Stefan

Hi Stefan,

thanks for clarification.

nenterprisemgr.cnf makes more sense than neneterprisemgr.cnf, which made me wondering about the filename.

Regards,
Holger

Ah, I see… I did mess it up indeed. I’ll correct it.

Cheers!