Noel J. Bergman wrote:
>Steve,
>
>You're probably seeing the FAQ item. Are you running the CVS HEAD?
>
Yep.
>It
>should happen far less frequently with changes made recently to the code.
>Basically, an exception occurs because of a known race condition. A thread
>picks up a mail item from the spool. After it gets the item, it attempts to
>lock the item. If another thread has it locked, then the second thread will
>give up and go back for another item. However, there is a window where one
>thread can pickup an item from the spool, and between when it gets the spool
>item and when it attempts to lock it, the thread that actually owns the
>spool item removes it from the spool and unlocks it. So the second thread
>finds that it is able to lock it, but when it goes to get it, the item is
>gone.
>
>The code:
>
> String key = spool.accept();
> MailImpl mail = spool.retrieve(key);
> if (infoEnabled) {
>
>should be changed to:
>
> String key = spool.accept();
> MailImpl mail = spool.retrieve(key);
> if (mail == null) continue;
> if (infoEnabled) {
>
>since we know that spool.retrieve(key) can return null. Try patching your
>copy of the code, and see if that doesn't fix it.
>
Thanks!
Patched the code locally and the NPE dissapears.
However, the mailstore exception remains - here is an extract of the
trace together with some additional comments in-line:
[INFO ] (root.james.smtp-server): Hello Name is: home
[DEBUG ] (root.james.smtp-server): Max message size is: 0
[INFO ] (root.james.smtp-server): Connection from racine.noos.net
(212.198.2.71)
[DEBUG ] (root.james.smtp-server): Sent: 220 home SMTP Server (JAMES
SMTP Server 2.1a1-cvs) ready Fri, 16 Aug 2002 09:34:06 +0200 (CEST)
[DEBUG ] (root.james.smtp-server): Command received: EHLO smtp.noos.fr
[DEBUG ] (root.james.smtp-server): Sent: 250 home Hello smtp.noos.fr
(racine.noos.net [212.198.2.71])
[DEBUG ] (root.james.smtp-server): Command received: MAIL
FROM:<[EMAIL PROTECTED]>
[DEBUG ] (root.james.smtp-server): Sent: 250 Sender
<[EMAIL PROTECTED]> OK
[DEBUG ] (root.james.smtp-server): Command received: RCPT
TO:<[EMAIL PROTECTED]>
[DEBUG ] (root.james.smtp-server): Sent: 250 Recipient
<[EMAIL PROTECTED]> OK
[DEBUG ] (root.james.smtp-server): Command received: DATA
[DEBUG ] (root.james.smtp-server): Sent: 354 Ok Send data ending with
<CRLF>.<CRLF>
Retrieving from: C:\TEMP\Mail1029483246359-056372.m64
Saving to: file://var/mail/spool//Mail1029483246359-0
Modified: false
[DEBUG ] (root.james.mailstore): Mail Mail1029483246359-0 stored.
[INFO ] (root.james.james): Mail Mail1029483246359-0 pushed in spool
[INFO ] (root.james.smtp-server): Mail sent to Mail Server
[DEBUG ] (root.james.smtp-server): Sent: 250 Message received
[DEBUG ] (root.james.smtp-server): Command received: QUIT
[DEBUG ] (root.james.smtp-server): Sent: 221 home Service closing
transmission channel
[DEBUG ] (root.james.smtp-server): Closing socket.
[DEBUG ] (root.james.mailstore): Retrieving mail: Mail1029483246359-0
The following error is occuring inside the
File_Persistent_Object_Repository class, get( String key ) method. I'm
using the cornerstone version from the Avalon Excalibur CVS, but I
noticed that there is a excalibur-datasource-1.0.jar in the James lib
directory which contains classes that also exist in cornerstone.jar.
The setup I have is ignoring the excalibur-datasource-1.0.jar - is this
likely to be a problem?
Steve.
[ERROR ] (root.james.mailstore): Exception retrieving mail:
java.lang.RuntimeException: Exception caught while retrieving an object,
cause:
java.io.StreamCorruptedException: unexpected end of block data, so
we're deleting it... good riddance!
java.lang.RuntimeException: Exception caught while retrieving an object,
cause: java.io.StreamCorruptedException: unexpected end of block da
ta
at
org.apache.avalon.cornerstone.blocks.masterstore.File_Persistent_Object_Repository.get(File_Persistent_Object_Repository.java:68)
at
org.apache.james.mailrepository.AvalonMailRepository.retrieve(AvalonMailRepository.java:260)
at
org.apache.james.transport.JamesSpoolManager.run(JamesSpoolManager.java:268)
at
org.apache.avalon.excalibur.thread.impl.ExecutableRunnable.execute(ExecutableRunnable.java:47)
at
org.apache.avalon.excalibur.thread.impl.WorkerThread.run(WorkerThread.java:80)
>
> --- Noel
>
>-----Original Message-----
>From: Stephen McConnell [mailto:[EMAIL PROTECTED]]
>Sent: Friday, August 16, 2002 2:15
>To: James Developers List
>Subject: Re: James under Merlin
>
>
>
>
>In JamesSpoolManager, run method:
>
> MailImpl mail = spool.retrieve(key);
>
>The mail variable is is the source of the NPE.
>
>Steve.
>
>
>Stephen McConnell wrote:
>
>
>>
>>Paul Hammant wrote:
>>
>>
>>>Stephen,
>>>
>>>
>>>>I've been playing around with James under Merlin.
>>>>
>>>>During the process I came accross a runtime problem:
>>>>
>>>> 1. The class AvalonUsersStore is referenced as a Component
>>>> somewhere or other but doesn't implement the Component
>>>> interface
>>>>
>>>
>>>
>>>Err should be able to fix that one.
>>>
>>
>>
>>It rather easy!
>>
>>
>>>
>>>> 2. After starting James inside Merlin, then sending an email
>>>> to an arbitary name on the server James is running on,
>>>> I get NPE - which is where I figure I may need some help
>>>> from you guys ...
>>>>
>>>
>>>
>>>This is where we wich we could unit test each comp.
>>>
>>
>>
>>Something to keep in mind is that I'm using the CVS version of
>>cornerstone components. I had to do a number of updates to
>>cornerstone including
>>
>> (a) addition of the Component interface to soem of the blocks
>> becuase they are used in James - so adding Component is
>> basically a backward compatibility thing
>>
>> (b) addition of some meta-info into a couple of the blocks so
>> that Merlin can automatically build the services without
>> having Phoenix around
>>
>>Apart from the above, the only differences are that I'm using a few
>>more excalibur jar files that are not included in the James lib
>>(because they are included in Phoenix and not in Merlin) and
>>naturally, this bypasses all of the Phoenix config, assembly, and
>>server files.
>>
>>
>>>A guy at BP has written a generic AbstractPhoenixTestCase (where
>>>Phoenix is mocked). I'll get him to write it up and someone might do
>>>the same thing inside Phoenix CVS or maybe Merlin CVS ? :-)
>>>
>>
>>
>>I'll put in place a Merlin projects directory containing the James
>>content and update the CVS.
>>
>>
>>>Failing that we'll have to do good old fashioned debugging. Question
>>>do you use IntelliJ ? Have you got interactive debuging working with
>>>Phoenix?
>>>
>>
>>
>>No on both points!
>>In the meantime I'll start tracking it down.
>>
>>Cheers, Steve.
>>
>>
>>
>>>
>>>- Paul
>>>
>>>
>>>--
>>>To unsubscribe, e-mail:
>>><mailto:[EMAIL PROTECTED]>
>>>For additional commands, e-mail:
>>><mailto:[EMAIL PROTECTED]>
>>>
>
>--
>
>Stephen J. McConnell
>
>OSM SARL
>digital products for a global economy
>mailto:[EMAIL PROTECTED]
>http://www.osm.net
>
>
>
>
>--
>To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
>For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
>
>
>--
>To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
>For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
>
--
Stephen J. McConnell
OSM SARL
digital products for a global economy
mailto:[EMAIL PROTECTED]
http://www.osm.net
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
