JVM crashes when(persistence-enabled=true) <persistence-enabled>true</persistence-enabled> in Jboss 7.1.1 final standalone.xml in RHEL 6

[freakyprogrammer]

I configured HornetQ in my Jboss 7.1.1 final to my JMS queue. I copied whatever the configuration related to hoqnetq in the standalone-full.xml into standalone.xml. But when I start the server JVM is crashing with the following message.

But If I specify  <journal-type>NIO</journal-type> server started without any issue even <persistence-enabled>true</persistence-enabled> .  JVM crashing in the following cases

1) <persistence-enabled>true</persistence-enabled> with no <journal-type/> in Standalone.xml

2) <persistence-enabled>true</persistence-enabled> with <journal-type>ASYNCIO</journal-type>.

but working fine with

<journal-type>NIO</journal-type>

Unhandled exception

Type=Segmentation error vmState=0x00000000

J9Generic_Signal_Number=00000004 Signal_Number=0000000b Error_Value=00000000 Signal_Code=00000001

Handler1=00007F5F27756F60 Handler2=00007F5F273F7900 InaccessibleAddress=0000000000000058

RDI=0000000000000040 RSI=0000000000000000 RAX=0000000000000040 RBX=0000000000000110

RCX=0000000000000020 RDX=0000000000000000 R8=00007F5F240572C0 R9=0000000000000000

R10=00007F5F240575A0 R11=00007F5F183270F4 R12=0000000000000003 R13=0000000000020102

R14=00007F5F278D3000 R15=00007F5F27EA4080

RIP=00007F5F1832710D GS=0000 FS=0000 RSP=00007F5F24057770

EFlags=0000000000210202 CS=0033 RBP=00007F5F24057810 ERR=0000000000000004

TRAPNO=000000000000000E OLDMASK=0000000000000000 CR2=0000000000000058

xmm0 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm1 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm2 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm3 45646e756f466665 (f: 1866884736.000000, d: 1.976011e+26)

xmm4 0043495353414c43 (f: 1396788352.000000, d: 2.145686e-307)

xmm5 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm6 00007f5f24057cc0 (f: 604339392.000000, d: 6.919222e-310)

xmm7 00007f5eb80050b4 (f: 3087028480.000000, d: 6.919132e-310)

xmm8 0000000042400000 (f: 1111490560.000000, d: 5.491493e-315)

xmm9 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm10 0000ff0000000000 (f: 0.000000, d: 1.385239e-309)

xmm11 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm12 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm13 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm14 0000000000000000 (f: 0.000000, d: 0.000000e+00)

xmm15 0000000000000000 (f: 0.000000, d: 0.000000e+00)

When I check for installed packages in my RHEL I see libaio-0.3.107-10.el6.x86_64 installed on my machine. Please help me to understand the reason to this.

[clebert.suconic]

I don't have much information, but I can only think of two possibilities:

I - you're using a non supported file system not supported by libaio.. (but you shouldn't get a Crash this way)

II - you could try recompile the .so following the chapter on hornetq.

At this point I have no information, and no one else reporting this issue to me. libaio has been quite stable actually.

[freakyprogrammer]

First I would like to thank you for quick reply. As I am working on Linux OS for first time to develop my Java Application I am not much fimilar with Linux file systems . How I can choose a file system which supports libaio? But as seen it's working fine with NIO can I continue working with this journal type? Can you please give some info about cases where I have to use ASYNCIO over NIO?

[clebert.suconic]

You can continue fine with NIO.

We usually recomment ext3 or ext4 without any compression or criptography (Just plain journalling). Any other system has to be looked up.

there are no problems on using NIO. usually AIO has a bit better performance, but we also simulate asynchronous IO through our implementation on TimedBuffer class.