I have a problem with a brand new Max 6096. I am trying to phase out some Max 4000 series units. However ever since putting this 6096 online, I have experienced som serious intermittent problems. Lucent has already RMA'd the chassis of this unit because of a dead port. Until a couple of days ago, the new unit has performed without any apparent problems for about 2 weeks except for a recurring messages in my Cistron radius logfile: Wed Apr 11 11:00:25 2001: Error: Accounting: no username in record Wed Apr 11 11:01:37 2001: Error: Accounting: no username in record When I have the new 6096 online, I get these messages at fairly regular intervals, usually about 2 to 10 error messages spread out over the course of an hour. With only the 4000's online, I don't get the messages, except perhaps very rarely. This is the same message I would get when a call would hit the 'dead' port before the unit was RMA'd. The caller would hear a single high-pitch unwavering continuous tone and that is all. By monitoring the Max it is apparent that the calls were hitting the Max, but not being completed. Monday night those same errors started occuring in great numbers until that is all that was being seen in the log files. Each caller was getting the same high-pitch continuous tone. Tuesday I spent a lot of time with Southwestern Bell technicians trying to troubleshoot the problem and in the end, the best we could determine is that there seems to be a problem with the CSU's associatd with WAN3 and WAN4. Because when we took those incoming T1's and moved them to a 4000, calls began to come in ok. In all the processs of trouble shooting, I never rebooted the machine because WAN1 and WAN2 were functioning perfectly and I didn't not want to knock those users off. I feel like a reboot might have cleared some error condition in the machine but I have been using Max series 6000's for about 4 years now, and I have never had to reboot one to reset any error conditions. However at the other installation I am dealing with a nearly state-of-the-art switch and Sprint as the telco. My Maxen there run version 7.2.3 which I've found to be very stable. At this installation where I am having problems, I am dealing with an older Erikson Master/Slave switch and Southwestern Bell as the Telco. In this scenario, the orginating calls are from a small town with the Erikson slave switch. My T1 lines (plain old channelized T1, no PRI available here) come out of the Erickson master switch 60 miles away. So there is a lot of hardware between the originating call and my Maxen. The software on the new Max is running 8.0.3. However the first Max before being RMA'd was loaded with 7.2.3 and was still seeing a lot of of the errors mentioned above. So I don't think it is a software issue. All four WAN ports are configured identically on the problem Max and are as configured as closely as possible to the Maxen that are working fine at the other installation. The big difference being that I have to set Rob Ctl to Inc-W-200 as it is on my 4000 units. This is the only robbed-bit control setting that has worked at this installation. My delima is whether to try to get Lucent to RMA this unit again. I don't really know what else to do to try to troubleshoot this problem. Perhaps the Max 6000 is just not compatible with the Erikson switch or this type of slave/master configuration of switches. If anyone has any experience in dealing with connecting Max 6000's to older Erikson switches or has suggestions on how to further troubleshoot this problem, I would be grateful for any help. Thank you, Dennis Duval ++ Ascend Users Mailing List ++ To unsubscribe: send unsubscribe to ascend-users-request at bungi.com To get FAQ'd: <http://www.nealis.net/ascend/faq>