[JayVal]

We have this OAS running Out of Memory issue in Production on a daily basis and we had to restart the Server everytime. OTM becomes slow to the point that the application was not responding.

A check across the system showed the OAS JVM was using 100% of 1 CPU
And the OAS heap was maxed out on memory:

[Full GC 1024447K->1024447K(1024448K), 4.0970930 secs]
[Full GC 1024447K->1024447K(1024448K), 4.0963020 secs]
[Full GC 1024447K->1024447K(1024448K), 4.0915010 secs]

Garbage Collection (GC) was taking place, but little or no un-used memory was being Unloaded.

Scanning the OAS console.log, a pattern was seen where memory began rising.

10:47 AM – OK at this point
[GC 665769K->456278K(1023424K), 0.0744990 secs]

As we approach 11:21 AM, OAS starts using more total memory and releases less memory via GC.
The GC times increase also indicating more work is needed to find memory to Unload.

11:21 AM
[GC 875287K->672253K(1024448K), 3.7772500 secs]

By 11:26 AM, OAS has begun maxing the heap before GC runs,
and OAS begins performing FULL GC”s.

[Full GC 1024447K->815040K(1024448K), 3.4803020 secs]

At this point, GC is still Unloading memory and the application is still responding.

As time goes on, OAS is able to Unload less and less memory.

As of 11:38 AM, OAS is getting little or no memory back via GC.
At this point, the application has already stopped responding.

Please help.