United StatesChange Country, Oracle Worldwide Web Sites Communities I am a... I want to...
JDK-7065972 : Some race condition may happen in SSLSocketImpl class

Details
Type:
Bug
Submit Date:
2011-07-12
Status:
Closed
Updated Date:
2013-10-29
Project Name:
JDK
Resolved Date:
2011-08-16
Component:
security-libs
OS:
windows_xp,windows_7
Sub-Component:
javax.net.ssl
CPU:
x86
Priority:
P3
Resolution:
Fixed
Affected Versions:
6u23,7
Fixed Versions:

Related Reports
Backport:
Backport:
Backport:
Relates:

Sub Tasks

Description
During SSL and client authentication network communication, there is a race condition may happen, the exception are looking like this:

Exception in thread "HandshakeCompletedNotify-Thread" java.util.ConcurrentModificationException
	at java.util.HashMap$HashIterator.nextEntry(Unknown Source)
	at java.util.HashMap$EntryIterator.next(Unknown Source)
	at java.util.HashMap$EntryIterator.next(Unknown Source)
	at com.sun.net.ssl.internal.ssl.SSLSocketImpl$NotifyHandshakeThread.run(Unknown Source)

This is a partial fix for bug 7060523, which will only apply to JDK repository.

                                    

Comments
EVALUATION

This is a regression of CR 6357710.  Need a fix in both deploy and JDK code. Deploy team also filled a CR (7060523) for the issue. SQE has approved to defer CR 7060523 for JDK7.

Note that this should be a partial fix for bug 7060523, which will only apply to JDK repository. Deploy code will address the issue in CR 7060523 separately.

The problem in applet exists since 6u23, not a showstopper for 7.

The problem will only happen when the applet request TLS handshaking/renegotiation very frequently. No wide scope impact. No impact on normal JSSE/HTTPS applications except the Java plugin. The issue in JSSE has been there since the SSLSocketImpl.java was created.
                                     
2011-07-12
We should be able to use the test results from testcase: http://129.148.174.102:8000/dennis/bug/dolphin/6.0update/6357710/unitTest.html (deployment test JDK-6357710).
But I am not able to locate the results for jdk 8 in aurora.
Written to dennis/jitender for pointers
                                     
2013-06-20
noreg-hard. There is no regression test available to verify this. 
Marking this as verified based on comments in the bug
                                     
2013-10-29



Hardware and Software, Engineered to Work Together