Jabber dispatcher clogs the whole alertqueue

We have used the jabber dispatcher since december. We soon discovered that our XMPP-server implementation (Openfire) sometimes freeze. It does not crash completely, but takes forever to answer connections. When this freeze happens and NAV tries to send an alert via jabber, the whole alertqueue clogs up and no alerts gets sent at all - no email, no sms, nothing. When the jabber-daemon is restarted all the backed up alerts are sent right away.

(This of course happened during the holidays and we have disabled the jabber dispatcher ever since.)

Is it possible to add a timeout or something to jabber_dispatcher.py to prevent this to happen?

--Ingeborg