+1
Fix looks good regd the failure scenerio. Really good stuff. :-)
Question regd. connectionTimeout and sendPartial.
Noel, are the comments on these lines an overisight ?
> + //props.put("mail.smtp.connectiontimeout", connectionTimeout +
"");
> +
//props.put("mail.smtp.sendpartial",String.valueOf(sendPartial));
I am not sure about sendPartial change or the intent, but at least
connectionTimeout change looks very safe and good to enable in 2.1.
Harmeet
----- Original <checkin> Message -----
From: <[EMAIL PROTECTED]>
Subject: cvs commit:
jakarta-james/src/java/org/apache/james/transport/mailets
RemoteDelivery.java
> noel 2003/02/20 18:22:21
>
> Modified: src/java/org/apache/james/transport/mailets Tag:
> branch_2_1_fcs RemoteDelivery.java
> Log:
...
> This should also be the code necessary to support mail.smtp.sendpartial.
...
> @@ -66,6 +63,8 @@
> private long delayTime = 21600000; // default is 6*60*60*1000
millis (6 hours)
> private int maxRetries = 5; // default number of retries
> private long smtpTimeout = 600000; //default number of ms to
timeout on smtp delivery
> + private boolean sendPartial = false; // If false then ANY address
errors will cause the transmission to fail
> + private int connectionTimeout = 60000; // The amount of time
JavaMail will wait before giving up on a socket connect()
...
> @@ -594,6 +643,10 @@
> props.put("mail.smtp.ehlo", "false");
> //Sets timeout on going connections
> props.put("mail.smtp.timeout", smtpTimeout + "");
> +
> + //props.put("mail.smtp.connectiontimeout", connectionTimeout +
"");
> +
//props.put("mail.smtp.sendpartial",String.valueOf(sendPartial));
----- Original Message -----
From: "Noel J. Bergman" <[EMAIL PROTECTED]>
> Guys,
>
> I just found and fixed a serious defect today. The failure scenario was
> this:
>
> - transport.connect() throws a MessagingException
> - failMessage returns false, so continue to loop
> - there are no additional MX records
> - the while loop exits
> - the return statement at the end of delivery returns true
> - the if (deliver(...)) statement in run() deletes the message
>
> This bug exists in all released versions of v2.1.x. I reproduced this
> situation quite readily today prior to fixing the bug.
>
> I have posted v2.1.2a2 for testing. PLEASE TEST IT ASAP, vote to release
> unless you find a problem.
>
> --- Noel
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
