Running CA Access Gateway (SPS) randomly this one experience crashes and when that occurs, the browsers get the return code HTTP 503.
This happens randomly, and on all CA Access Gateway (SPS) in the Production environment.
The restart of the CA Access Gateway (SPS) solves the issue, but doesn't prevent to happening again.
The crash happens in the JVM:
Current thread (0x00007fa494005730): GCTaskThread "GC Thread#2" [stack: 0x00007fa4999fb000,0x00007fa499afb000] [id=17960]
Stack: [0x00007fa4999fb000,0x00007fa499afb000], sp=0x00007fa499af9c58, free space=1019k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.so+0x7608e4] G1SATBMarkQueueSet::filter(SATBMarkQueue&)+0x74
V [libjvm.so+0xcd7974] SATBMarkQueueSet::flush_queue(SATBMarkQueue&)+0x14
V [libjvm.so+0x704136] G1RemarkThreadsClosure::do_thread(Thread*)+0x36
V [libjvm.so+0xe62fbd] Threads::threads_do(ThreadClosure*)+0x4d
V [libjvm.so+0x707ffe] G1CMRemarkTask::work(unsigned int)+0x12e
V [libjvm.so+0xf1688f] GangWorker::loop()+0x5f
V [libjvm.so+0xf168ef]
V [libjvm.so+0xe655d8] Thread::call_run()+0xa8
V [libjvm.so+0xc26851] thread_native_entry(Thread*)+0xe1
CA Access Gateway (SPS) 12.8SP8CR01 on RedHat 7;
OpenJDK 64-Bit Server VM Temurin-17.0.11+9;
Dynatrace agent is configured;
Apply the 3 changes to the environment to progress to get more information about the JVM.
JVM_PERF_OPTS="-server -XX:MaxMetaspaceSize=256M -XX:+UseParallelGC"
JVM_MEM_OPTS="-Xms1024m -Xmx2048m"