Hi Helge,

the message is the result of a REGISTER retransmission ; as the registration is stateless, the retransmission will be processed exactly as the original request - and usrloc reports bad registration because both requests has the same callid and cseq.

there is a thread on devel about his issue (originated by Juha) with more info and the future fix.

regards, bogdan

Helge Waastad wrote:

Hi, I have a Loadbalanced scenario with two servers of each (2x LB/Dispatchers and 2x Routers)

Each router has its own location table. They both replicate REGISTERS (forward_tcp/save_noreply).

However every now and then I get:

ERROR:usrloc:update_contacts: invalid cseq for aor

This happens for all contacts

I do not really know why, since all the transactions seems allright to me and that it happens only now and then...

Any clues to help med debug further?

br hw