[
https://issues.apache.org/jira/browse/DIRMINA-869?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Martin Ellis updated DIRMINA-869:
---------------------------------
Description:
The addresses field is a HashSet wrapped using Collections.synchronizedSet.
The modifications to the set in the bind(...) and unbind(...) method, are
guarded by synchronizing on 'this'.
However, the unbindAll() method iterates over the set, without synchronization.
This could lead to a bug if a call to unbindAll() is made while a socket is
being bind()'ed or unbind()'ed.
>From NioTcpServer:
private Set<SocketAddress> addresses =
Collections.synchronizedSet(new HashSet<SocketAddress>());
...
public void unbindAll() throws IOException {
for (SocketAddress socketAddress : addresses) {
unbind(socketAddress);
}
}
>From the unsynchronizedSet javadoc.
It is imperative that the user manually synchronize on the returned set
when iterating over it:
Set s = Collections.synchronizedSet(new HashSet());
...
synchronized(s) {
Iterator i = s.iterator(); // Must be in the synchronized block
while (i.hasNext())
foo(i.next());
}
Another fix might be to use a set from java.util.concurrent.
was:
The addresses field is a HashSet wrapped using Collections.synchronizedSet.
The modifications to the set in the bind(...) and unbind(...) method, are
guarded by synchronizing on 'this'.
However, the unbindAll() method iterates over the set, without synchronization.
This could lead to a bug if a call to unbindAll() is made while a socket is
being bind()'ed or unbind()'ed.
>From NioTcpServer:
{noformat}
private Set<SocketAddress> addresses =
Collections.synchronizedSet(new HashSet<SocketAddress>());
...
public void unbindAll() throws IOException {
for (SocketAddress socketAddress : addresses) {
unbind(socketAddress);
}
}
{noformat}
>From the unsynchronizedSet javadoc.
{noformat}
It is imperative that the user manually synchronize on the returned set when
iterating over it:
Set s = Collections.synchronizedSet(new HashSet());
...
synchronized(s) {
Iterator i = s.iterator(); // Must be in the synchronized block
while (i.hasNext())
foo(i.next());
}
{noformat}
Another fix might be to use a set from java.util.concurrent.
> NioTcpServer is not thread-safe when unbinding sockets
> ------------------------------------------------------
>
> Key: DIRMINA-869
> URL: https://issues.apache.org/jira/browse/DIRMINA-869
> Project: MINA
> Issue Type: Bug
> Components: Core
> Reporter: Martin Ellis
>
> The addresses field is a HashSet wrapped using Collections.synchronizedSet.
> The modifications to the set in the bind(...) and unbind(...) method, are
> guarded by synchronizing on 'this'.
> However, the unbindAll() method iterates over the set, without
> synchronization. This could lead to a bug if a call to unbindAll() is made
> while a socket is being bind()'ed or unbind()'ed.
> From NioTcpServer:
> private Set<SocketAddress> addresses =
> Collections.synchronizedSet(new HashSet<SocketAddress>());
> ...
> public void unbindAll() throws IOException {
> for (SocketAddress socketAddress : addresses) {
> unbind(socketAddress);
> }
> }
> From the unsynchronizedSet javadoc.
> It is imperative that the user manually synchronize on the returned set
> when iterating over it:
> Set s = Collections.synchronizedSet(new HashSet());
> ...
> synchronized(s) {
> Iterator i = s.iterator(); // Must be in the synchronized block
> while (i.hasNext())
> foo(i.next());
> }
> Another fix might be to use a set from java.util.concurrent.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
