[jboss-jira] [JBoss JIRA] Commented: (JGRP-647) Potentially Meaningless InterruptedIOException Can Be Logged Repeatedly

[Galder Zamarreno (JIRA)]

    [ http://jira.jboss.com/jira/browse/JGRP-647?page=comments#action_12398433 ] 
            
Galder Zamarreno commented on JGRP-647:
---------------------------------------

For PingWaiter, whenever start() is called, it creates a brand new 
thread and starts it and after find initial members, it nulls the thread so 
will be recreated again next time find new members is requested, 
which as seen can happen very often, i.e. when MERGE2 tries to find new 
initial members.

I need to check how this works in JGroups 2.5.x and 2.6.x

> Potentially Meaningless InterruptedIOException Can Be Logged Repeatedly
> -----------------------------------------------------------------------
>
>                 Key: JGRP-647
>                 URL: http://jira.jboss.com/jira/browse/JGRP-647
>             Project: JGroups
>          Issue Type: Bug
>    Affects Versions: 2.5, 2.6, 2.4.1 SP3, 2.4.1 SP4, 2.5.1
>            Reporter: Jimmy Wilson
>         Assigned To: Bela Ban
>            Priority: Minor
>             Fix For: 2.7, 2.4.2
>
>         Attachments: cluster.log.tar.gz
>
>
> TP.down doesn't catch/handle the InterruptedIOException that can be thrown from the underlying send call when interrupted:
>         try {
>             if(use_outgoing_packet_handler)
>                 outgoing_queue.put(msg);
>             else
>                 send(msg, dest, multicast);
>         }
>         catch(QueueClosedException closed_ex) {
>         }
>         catch(InterruptedException interruptedEx) {
>         }
>         catch(Throwable e) {
>             if(log.isErrorEnabled()) {
>                 String dst=msg.getDest() == null? "null" : msg.getDest().toString();
>                 log.error("failed sending message to " + dst + " (" + msg.getLength() + " bytes)", e.getCause());
>             }
>         }
> This can lead to it being logged repeatedly at the ERROR level:
> 2007-11-14 10:02:30,040 ERROR -> (PingSender) [UDP] failed sending message to null (0 bytes)
> java.io.InterruptedIOException: operation interrupted
> at java.net.PlainDatagramSocketImpl.send(Native Method)
> at java.net.DatagramSocket.send(DatagramSocket.java:612)
> at org.jgroups.protocols.UDP._send(UDP.java:316)
> at org.jgroups.protocols.UDP.sendToAllMembers(UDP.java:286)
> at org.jgroups.protocols.TP.doSend(TP.java:937)
> at org.jgroups.protocols.TP.send(TP.java:926)
> at org.jgroups.protocols.TP.down(TP.java:712)
> at org.jgroups.stack.Protocol.receiveDownEvent(Protocol.java:499)
> at org.jgroups.stack.Protocol.passDown(Protocol.java:533)
> at org.jgroups.protocols.PING.sendMcastDiscoveryRequest(PING.java:214)
> at org.jgroups.protocols.PING.sendGetMembersRequest(PING.java:208)
> at org.jgroups.protocols.PingSender.run(PingSender.java:59)
> at java.lang.Thread.run(Thread.java:595) 
> 2.5/2.6 don't handle the InterruptedIOException either, but since this was reported with EAP 4.2 (JGroups 2.4.1.SP3), I 'd like to see it in 2.4.2 if possible.  That way, it can make it into a 4.2/4.3 EAP CP.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira