Hi!
Looks like 1 GB of heap is not enough for your app to work normally. Give
it more memory, it's a Java application after all :)
It's true, that persistence helps you forget about eviction policy and your
data is safe, even when memory comes to an end. But this mechanism affects
only off-heap memory.
Note, that if you start Ignite as an embedded application, it will share
heap with your app.
So, don't restrict Ignite in memory consumption too much, it needs some
heap to operate normally.
Denis
вт, 2 янв. 2018 г. в 14:05, userx <gagan...@gmail.com>:
> Hi All,
>
> Ignite has thrown an OutOfMemory Error, the stack of which is mentioned
> below
>
> 2017-12-29 10:19:11,655 ERROR
> [tcp-disco-sock-reader-#40%63d769bd-bea6-450e-ab8b-0697da94ef1e%] {}
> org.apache.ignite.spi.discovery.tcp.TcpDiscoverySpi - Runtime error caught
> during grid runnable execution: Socket reader [id=133279,
> name=tcp-disco-sock-reader-#40%63d769bd-bea6-450e-ab8b-0697da94ef1e%,
> nodeId=f50f84ea-d362-4fdd-a41e-1df63043d78e] java.lang.OutOfMemoryError:
> Java heap space
>
> 2017-12-29 10:19:36,733 ERROR
> [nio-acceptor-#63%63d769bd-bea6-450e-ab8b-0697da94ef1e%] {}
>
> org.apache.ignite.internal.processors.rest.protocols.tcp.GridTcpRestProtocol
> - Runtime error caught during grid runnable execution: GridWorker
> [name=nio-acceptor,
> igniteInstanceName=63d769bd-bea6-450e-ab8b-0697da94ef1e,
> finished=false, hashCode=164131648, interrupted=false,
> runner=nio-acceptor-#63%63d769bd-bea6-450e-ab8b-0697da94ef1e%]
> java.lang.OutOfMemoryError: Java heap space
> at
>
> org.jsr166.ConcurrentHashMap8$KeySetView.iterator(ConcurrentHashMap8.java:3612)
> at
>
> org.apache.ignite.internal.util.GridSetWrapper.iterator(GridSetWrapper.java:94)
> at
>
> org.apache.ignite.internal.util.nio.GridNioServer$AbstractNioClientWorker.reset0(GridNioServer.java:2525)
> at
>
> org.apache.ignite.internal.util.nio.GridNioServer$SizeBasedBalancer.run(GridNioServer.java:3850)
> at
>
> org.apache.ignite.internal.util.nio.GridNioServer$GridNioAcceptWorker.accept(GridNioServer.java:2613)
> at
>
> org.apache.ignite.internal.util.nio.GridNioServer$GridNioAcceptWorker.body(GridNioServer.java:2580)
> at
> org.apache.ignite.internal.util.worker.GridWorker.run(GridWorker.java:110)
> at java.lang.Thread.run(Thread.java:748)
> Ignite_OOM.csv
> <http://apache-ignite-users.70518.x6.nabble.com/file/t1165/Ignite_OOM.csv>
>
> I have analyzed the same in Eclipse MAT and have attached the Dominator
> objects.
>
> The Ignite server instance has been running on a max heap of 1 GB
> (Persistent Store). The topology is
>
> 2017-12-29 10:17:28,813 INFO
> [disco-event-worker-#69%63d769bd-bea6-450e-ab8b-0697da94ef1e%] {}
> org.apache.ignite.internal.managers.discovery.GridDiscoveryManager -
> Topology snapshot [ver=1375, servers=3, clients=10, CPUs=168, heap=150.0GB]
>
> The clients just come and go. The clients are running in a jvm with max
> size
> of 15G.
>
> I have been using it for 6 months now and it was flawless. There are 3
> server nodes in the topology.Last week, it threw this error, so I am a bit
> confused as to why it failed in this configuration because even if the data
> surpasses 1 GB, it shouldn't fail on OOM.
>
>
>
>
> --
> Sent from: http://apache-ignite-users.70518.x6.nabble.com/
>
