Large messages causing crash in different ways

In the documentation there is no mention of message size limitation. Here is a sample code which cause a break on Linux:

import net.openhft.chronicle.core.UnsafeMemory;
import net.openhft.chronicle.queue.ChronicleQueue;
import net.openhft.chronicle.queue.ChronicleQueueBuilder;
import net.openhft.chronicle.queue.ExcerptAppender;
import sun.misc.Unsafe;

public class CQAppenderTest {
	static final int ITER = 2;
	static final int SIZE = 50 * 1000 * 1000;
	public static void main(String[] args) {
		ChronicleQueue queue = ChronicleQueueBuilder.single("queueDir").build();
		final ExcerptAppender appender = queue.acquireAppender();
		byte[] array = new byte[SIZE];
		long startTime = System.currentTimeMillis();
		for (int i = 0; i < ITER; i++) {
			appender.writeBytes(b -> {
			    long address = b.addressForRead(b.writePosition());
			    Unsafe unsafe = UnsafeMemory.UNSAFE;
			    unsafe.copyMemory(array, Unsafe.ARRAY_BYTE_BASE_OFFSET, null, address, array.length);
			    b.writeSkip(array.length);
			});
		}
		long estimatedTime = System.currentTimeMillis() - startTime;
		System.out.println("time = " + estimatedTime);
	}
}

And here is the error message:

#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fddf138f172, pid=25577, tid=0x00007fddf2791700
#
# JRE version: Java(TM) SE Runtime Environment (8.0_161-b12) (build 1.8.0_161-b12)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.161-b12 mixed mode linux-amd64 compressed oops)
# Problematic frame:
# V  [libjvm.so+0x803172]
#
# Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# ~/smallTests/CQTest/hs_err_pid25577.log
#
# If you would like to submit a bug report, please visit:
#   http://bugreport.java.com/bugreport/crash.jsp
#

Also, I have tried to write message with appender.writeDocument. Although it works with 50MB message, it fails with 100MB message size:

Exception in thread "main" java.lang.IllegalArgumentException: pos: 100131612, store.writePosition()=131584 queue.blockSize()=67108864
	at net.openhft.chronicle.queue.impl.single.SingleChronicleQueueExcerpts$StoreAppender.position(SingleChronicleQueueExcerpts.java:495)
	at net.openhft.chronicle.queue.impl.single.SingleChronicleQueueExcerpts$StoreAppender.access$2000(SingleChronicleQueueExcerpts.java:66)
	at net.openhft.chronicle.queue.impl.single.SingleChronicleQueueExcerpts$StoreAppender$HeaderWriteStrategyOriginal.onContextOpen(SingleChronicleQueueExcerpts.java:841)
	at net.openhft.chronicle.queue.impl.single.SingleChronicleQueueExcerpts$StoreAppender.writingDocument(SingleChronicleQueueExcerpts.java:332)
	at net.openhft.chronicle.wire.MarshallableOut.writeDocument(MarshallableOut.java:95)
	at CQAppenderTest.main(CQAppenderTest.java:22)

[peter-lawrey]

A related issue is https://github.com/OpenHFT/Chronicle-Bytes/issues/66 as write(bytes) and read(bytes) can also fail in some cases.

[RobAustin]

hence, you could increase your block size see net.openhft.chronicle.queue.ChronicleQueueBuilder#blockSize(int) , [ but you should make sure that your largest message size is no larger than 1/4 of your block size]