JVM crash in PlainSocketImpl_socketConnect

[Sarah Allen]

Hi,

We have a webmail product that uses JavaMail and we're seeing very
intermittent crashes in production (see stack trace below).  It doesn't
appear to be out of memory from looking at the crash log and we can't
(yet) reproduce in the lab, so I thought I would ask on this list in
case it sounds familiar to anyone.  This was with Java version
1.5.0_11-b03 (more details on server env below).

Thanks in advance,
Sarah

Stack: [0x77e90000,0x77eb0000),  sp=0x77eacfc0,  free space=115k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V  [libjvm.so+0x152a80]
V  [libjvm.so+0x164d1c]
V  [libjvm.so+0x20b96c]
V  [libjvm.so+0x20a828]
V  [libjvm.so+0x20bd5c]
C  [libjava.so+0xee30]  Java_java_lang_Throwable_fillInStackTrace+0x14
J  java.lang.Throwable.fillInStackTrace()Ljava/lang/Throwable;
J  java.lang.Exception.<init>(Ljava/lang/String;)V
v  ~I2CAdapter
j  java.io.IOException.<init>(Ljava/lang/String;)V+2
j  java.net.SocketException.<init>(Ljava/lang/String;)V+2
v  ~StubRoutines::call_stub
V  [libjvm.so+0x19b95c]
V  [libjvm.so+0x1f706c]
V  [libjvm.so+0x2f2814]
C  [libjava.so+0xd57c]  JNU_NewObjectByName+0xb4
C  [libjava.so+0x14960]  JNU_ThrowByNameWithLastError+0x724
C  [libnet.so+0xcf78]
C  [libnet.so+0xacc8]  Java_java_net_PlainSocketImpl_socketConnect+0x6b4
j  java.net.PlainSocketImpl.socketConnect(Ljava/net/InetAddress;II)V+2080671304
j  java.net.PlainSocketImpl.socketConnect(Ljava/net/InetAddress;II)V+0
j  java.net.PlainSocketImpl.doConnect(Ljava/net/InetAddress;II)V+10
j  java.net.PlainSocketImpl.connectToAddress(Ljava/net/InetAddress;II)V+23
j  java.net.PlainSocketImpl.connect(Ljava/net/SocketAddress;I)V+71
j  java.net.SocksSocketImpl.connect(Ljava/net/SocketAddress;I)V+332
j  java.net.Socket.connect(Ljava/net/SocketAddress;I)V+165
j  java.net.Socket.connect(Ljava/net/SocketAddress;)V+3
j  com.sun.mail.util.SocketFetcher.createSocket(Ljava/net/InetAddress;ILjava/lang/String;IILjavax/net/SocketFactory;Z)Ljava/net/Socket;+93
v  ~C2IAdapter
J  com.sun.mail.util.SocketFetcher.getSocket(Ljava/lang/String;ILjava/util/Properties;Ljava/lang/String;Z)Ljava/net/Socket;
J  com.sun.mail.imap.IMAPStore.getProtocol(Lcom/sun/mail/imap/IMAPFolder;)Lcom/sun/mail/imap/protocol/IMAPProtocol;
J  com.sun.mail.imap.IMAPFolder.open(I)V


OS:                        Solaris 9 4/04 s9s_u6wos_08a SPARC
           Copyright 2004 Sun Microsystems, Inc.  All Rights Reserved.
                        Use is subject to license terms.
                             Assembled 22 March 2004

uname:SunOS 5.9 Generic_118558-39 sun4u  (T2 libthread)
rlimit: STACK 8192k, CORE infinity, NOFILE 60001, AS infinity
load average:1.13 1.16 0.87

CPU:total 4 has_v8, has_v9, has_vis1, has_vis2, is_ultra3

Memory: 8k page, physical 8388608k(4675696k free)

vm_info: Java HotSpot(TM) Server VM (1.5.0_11-b03) for solaris-sparc, built on Dec 15 2006 01:18:11 by unknown with unknown Workshop:0x550

===========================================================================
To unsubscribe, send email to listserv@... and include in the body
of the message "signoff JAVAMAIL-INTEREST".  For general help, send email to
listserv@... and include in the body of the message "help".

[Bill Shannon]

This is probably a question for the JDK/JVM team.  Do you have a support
contract?

I believe there are newer version of JDK 1.5 available so you could start
there.  Also, be sure your Solaris patches are up to date.

I'm not an expert in interpreting these dumps but it looks like the
failure is occurring while it's trying to throw an exception to report
a connection failure.  Are you getting an unexpected number of connection
failures, or are the failures associated with connecting to certain hosts?

Does the JVM die with SIGBUS or SIGSEGV or something like that?

Are you running the JVM standalone or do you have it embedded in another
application?


Sarah Allen wrote:

> Hi,
>
> We have a webmail product that uses JavaMail and we're seeing very
> intermittent crashes in production (see stack trace below).  It doesn't
> appear to be out of memory from looking at the crash log and we can't
> (yet) reproduce in the lab, so I thought I would ask on this list in
> case it sounds familiar to anyone.  This was with Java version
> 1.5.0_11-b03 (more details on server env below).
>
> Thanks in advance,
> Sarah
>
> Stack: [0x77e90000,0x77eb0000),  sp=0x77eacfc0,  free space=115k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
> C=native code)
> V  [libjvm.so+0x152a80]
> V  [libjvm.so+0x164d1c]
> V  [libjvm.so+0x20b96c]
> V  [libjvm.so+0x20a828]
> V  [libjvm.so+0x20bd5c]
> C  [libjava.so+0xee30]  Java_java_lang_Throwable_fillInStackTrace+0x14
> J  java.lang.Throwable.fillInStackTrace()Ljava/lang/Throwable;
> J  java.lang.Exception.<init>(Ljava/lang/String;)V
> v  ~I2CAdapter
> j  java.io.IOException.<init>(Ljava/lang/String;)V+2
> j  java.net.SocketException.<init>(Ljava/lang/String;)V+2
> v  ~StubRoutines::call_stub
> V  [libjvm.so+0x19b95c]
> V  [libjvm.so+0x1f706c]
> V  [libjvm.so+0x2f2814]
> C  [libjava.so+0xd57c]  JNU_NewObjectByName+0xb4
> C  [libjava.so+0x14960]  JNU_ThrowByNameWithLastError+0x724
> C  [libnet.so+0xcf78]
> C  [libnet.so+0xacc8]  Java_java_net_PlainSocketImpl_socketConnect+0x6b4
> j  
> java.net.PlainSocketImpl.socketConnect(Ljava/net/InetAddress;II)V+2080671304
>
> j  java.net.PlainSocketImpl.socketConnect(Ljava/net/InetAddress;II)V+0
> j  java.net.PlainSocketImpl.doConnect(Ljava/net/InetAddress;II)V+10
> j  java.net.PlainSocketImpl.connectToAddress(Ljava/net/InetAddress;II)V+23
> j  java.net.PlainSocketImpl.connect(Ljava/net/SocketAddress;I)V+71
> j  java.net.SocksSocketImpl.connect(Ljava/net/SocketAddress;I)V+332
> j  java.net.Socket.connect(Ljava/net/SocketAddress;I)V+165
> j  java.net.Socket.connect(Ljava/net/SocketAddress;)V+3
> j  
> com.sun.mail.util.SocketFetcher.createSocket(Ljava/net/InetAddress;ILjava/lang/String;IILjavax/net/SocketFactory;Z)Ljava/net/Socket;+93
>
> v  ~C2IAdapter
> J  
> com.sun.mail.util.SocketFetcher.getSocket(Ljava/lang/String;ILjava/util/Properties;Ljava/lang/String;Z)Ljava/net/Socket;
>
> J  
> com.sun.mail.imap.IMAPStore.getProtocol(Lcom/sun/mail/imap/IMAPFolder;)Lcom/sun/mail/imap/protocol/IMAPProtocol;
>
> J  com.sun.mail.imap.IMAPFolder.open(I)V
>
>
> OS:                        Solaris 9 4/04 s9s_u6wos_08a SPARC
>           Copyright 2004 Sun Microsystems, Inc.  All Rights Reserved.
>                        Use is subject to license terms.
>                             Assembled 22 March 2004
>
> uname:SunOS 5.9 Generic_118558-39 sun4u  (T2 libthread)
> rlimit: STACK 8192k, CORE infinity, NOFILE 60001, AS infinity
> load average:1.13 1.16 0.87
>
> CPU:total 4 has_v8, has_v9, has_vis1, has_vis2, is_ultra3
>
> Memory: 8k page, physical 8388608k(4675696k free)
>
> vm_info: Java HotSpot(TM) Server VM (1.5.0_11-b03) for solaris-sparc,
> built on Dec 15 2006 01:18:11 by unknown with unknown Workshop:0x550
>
> ===========================================================================
> To unsubscribe, send email to listserv@... and include in the body
> of the message "signoff JAVAMAIL-INTEREST".  For general help, send
> email to
> listserv@... and include in the body of the message "help".

===========================================================================
To unsubscribe, send email to listserv@... and include in the body
of the message "signoff JAVAMAIL-INTEREST".  For general help, send email to
listserv@... and include in the body of the message "help".

[Sarah Allen]

Hi Bill,

There is a support contract in place for this deployment -- I'll follow
that path.  I haven't reported a bug before, do you happen to know if I
should just report it on the website and have it escalated through
technical support or if they have their own channels and I should wait
till I figure out what those are and do it that way?

Unfortunately we cannot yet reproduce the bug, the servers were
restarted immediately after the crash, so we don't have any more data
that what is in the crash logs.

We're running Tomcat and it's a SIGSEGV failure:

#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
#  SIGSEGV (0xb) at pc=0xfe952a80, pid=23309, tid=1031
#
# Java VM: Java HotSpot(TM) Server VM (1.5.0_11-b03 mixed mode)
# Problematic frame:
# V  [libjvm.so+0x152a80]


Thanks so much for your quick response.

Sarah

Bill Shannon wrote:

> This is probably a question for the JDK/JVM team.  Do you have a support
> contract?
>
> I believe there are newer version of JDK 1.5 available so you could start
> there.  Also, be sure your Solaris patches are up to date.
>
> I'm not an expert in interpreting these dumps but it looks like the
> failure is occurring while it's trying to throw an exception to report
> a connection failure.  Are you getting an unexpected number of connection
> failures, or are the failures associated with connecting to certain
> hosts?
>
> Does the JVM die with SIGBUS or SIGSEGV or something like that?
>
> Are you running the JVM standalone or do you have it embedded in another
> application?
>
>
> Sarah Allen wrote:
>> Hi,
>>
>> We have a webmail product that uses JavaMail and we're seeing very
>> intermittent crashes in production (see stack trace below).  It
>> doesn't appear to be out of memory from looking at the crash log and
>> we can't (yet) reproduce in the lab, so I thought I would ask on this
>> list in case it sounds familiar to anyone.  This was with Java
>> version 1.5.0_11-b03 (more details on server env below).
>>
>> Thanks in advance,
>> Sarah
>>
>> Stack: [0x77e90000,0x77eb0000),  sp=0x77eacfc0,  free space=115k
>> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code,
>> C=native code)
>> V  [libjvm.so+0x152a80]
>> V  [libjvm.so+0x164d1c]
>> V  [libjvm.so+0x20b96c]
>> V  [libjvm.so+0x20a828]
>> V  [libjvm.so+0x20bd5c]
>> C  [libjava.so+0xee30]  Java_java_lang_Throwable_fillInStackTrace+0x14
>> J  java.lang.Throwable.fillInStackTrace()Ljava/lang/Throwable;
>> J  java.lang.Exception.<init>(Ljava/lang/String;)V
>> v  ~I2CAdapter
>> j  java.io.IOException.<init>(Ljava/lang/String;)V+2
>> j  java.net.SocketException.<init>(Ljava/lang/String;)V+2
>> v  ~StubRoutines::call_stub
>> V  [libjvm.so+0x19b95c]
>> V  [libjvm.so+0x1f706c]
>> V  [libjvm.so+0x2f2814]
>> C  [libjava.so+0xd57c]  JNU_NewObjectByName+0xb4
>> C  [libjava.so+0x14960]  JNU_ThrowByNameWithLastError+0x724
>> C  [libnet.so+0xcf78]
>> C  [libnet.so+0xacc8]  Java_java_net_PlainSocketImpl_socketConnect+0x6b4
>> j  
>> java.net.PlainSocketImpl.socketConnect(Ljava/net/InetAddress;II)V+2080671304
>>
>> j  java.net.PlainSocketImpl.socketConnect(Ljava/net/InetAddress;II)V+0
>> j  java.net.PlainSocketImpl.doConnect(Ljava/net/InetAddress;II)V+10
>> j  
>> java.net.PlainSocketImpl.connectToAddress(Ljava/net/InetAddress;II)V+23
>> j  java.net.PlainSocketImpl.connect(Ljava/net/SocketAddress;I)V+71
>> j  java.net.SocksSocketImpl.connect(Ljava/net/SocketAddress;I)V+332
>> j  java.net.Socket.connect(Ljava/net/SocketAddress;I)V+165
>> j  java.net.Socket.connect(Ljava/net/SocketAddress;)V+3
>> j  
>> com.sun.mail.util.SocketFetcher.createSocket(Ljava/net/InetAddress;ILjava/lang/String;IILjavax/net/SocketFactory;Z)Ljava/net/Socket;+93
>>
>> v  ~C2IAdapter
>> J  
>> com.sun.mail.util.SocketFetcher.getSocket(Ljava/lang/String;ILjava/util/Properties;Ljava/lang/String;Z)Ljava/net/Socket;
>>
>> J  
>> com.sun.mail.imap.IMAPStore.getProtocol(Lcom/sun/mail/imap/IMAPFolder;)Lcom/sun/mail/imap/protocol/IMAPProtocol;
>>
>> J  com.sun.mail.imap.IMAPFolder.open(I)V
>>
>>
>> OS:                        Solaris 9 4/04 s9s_u6wos_08a SPARC
>>           Copyright 2004 Sun Microsystems, Inc.  All Rights Reserved.
>>                        Use is subject to license terms.
>>                             Assembled 22 March 2004
>>
>> uname:SunOS 5.9 Generic_118558-39 sun4u  (T2 libthread)
>> rlimit: STACK 8192k, CORE infinity, NOFILE 60001, AS infinity
>> load average:1.13 1.16 0.87
>>
>> CPU:total 4 has_v8, has_v9, has_vis1, has_vis2, is_ultra3
>>
>> Memory: 8k page, physical 8388608k(4675696k free)
>>
>> vm_info: Java HotSpot(TM) Server VM (1.5.0_11-b03) for solaris-sparc,
>> built on Dec 15 2006 01:18:11 by unknown with unknown Workshop:0x550
>>
>> ===========================================================================
>>
>> To unsubscribe, send email to listserv@... and include in
>> the body
>> of the message "signoff JAVAMAIL-INTEREST".  For general help, send
>> email to
>> listserv@... and include in the body of the message "help".
>
> ===========================================================================
>
> To unsubscribe, send email to listserv@... and include in the
> body
> of the message "signoff JAVAMAIL-INTEREST".  For general help, send
> email to
> listserv@... and include in the body of the message "help".

===========================================================================
To unsubscribe, send email to listserv@... and include in the body
of the message "signoff JAVAMAIL-INTEREST".  For general help, send email to
listserv@... and include in the body of the message "help".

[Bill Shannon]

Sarah Allen wrote:
> Hi Bill,
>
> There is a support contract in place for this deployment -- I'll follow
> that path.  I haven't reported a bug before, do you happen to know if I
> should just report it on the website and have it escalated through
> technical support or if they have their own channels and I should wait
> till I figure out what those are and do it that way?

Honestly, I don't know for sure, but I suspect that if you file a bug
report on bugs.sun.com and then reference that information when you
contact tech support, that will likely make it easier to get the right
details to the right people.

===========================================================================
To unsubscribe, send email to listserv@... and include in the body
of the message "signoff JAVAMAIL-INTEREST".  For general help, send email to
listserv@... and include in the body of the message "help".