Thx for this info.
I guess the spooling queue was huge. Did you try to go via JMX to see
the number of messages in that queue?
Thx, Eric
On 10/01/2013 13:54, Markus Moldaschl wrote:
Hi,
just want to let you know of an issue we had yesterday with james
3.0beta3 and how we solved it.
Upon a restart for a change of the configuration we faced the problem
where the log File said "...AMQPersistenceAdapter - The
ReferenceStore is not valid - recovering ...."
After a couple of minutes james started to work again, but did only
deliver E-Mails for about half an hour. Then strange things happened,
one or two times per hour James started to deliver mails for a couple of
seconds and stopped again. After a bit of research on that I figured out
that a flush on the James outgoing queue helped the remote delivery
thread to work until it stopped again. I took thread dumps but could not
find anything which helped me out.
During my research on that I stumbeld upon [1]. We gave it a try and it
worked. Additionally I found out that our
var/mail/store/activemq/brokers/james/journal directory used 13 GB of
space (402 Files).
I don't know yet if there is a correlation with the not valid
ReferenceStore or the huge amount of needed space in the activemq
directory. I try to figure it out. Maybe in the meantime somebody can
point me to the right direction?!?
Best regards
Markus
[1]
http://www.mail-archive.com/server-user@james.apache.org/msg13056.html
---------------------------------------------------------------------
To unsubscribe, e-mail: server-user-unsubscr...@james.apache.org
For additional commands, e-mail: server-user-h...@james.apache.org
