Hello list. I am a first time poster, so please be gentle. I have been getting much headache with my Pipeline. I am having packet transfer issues in the order of 15%-20% packet loss through this adaptor. I am connecting to Qwest switch as well as ISP. I receive a static IP /29 block from them. Background: I have two Pipelin 75, one running 8.0-101.1 and the other 8.0-101.-2. Both experience the same issue. I have had my line tested with a loopback, 10 min BERT, and a session protocol. All these tests pass with flying colors. I have contacted quest.net (my ISP) several times to no avail. I have discovered some extended information sbout their setup. I have replaced my Pipelines with an Intel Ether Express, and had a level 2 tech for these devices come to my house to configure and do not experience this issue. This consistent issue happens in either D/L or U/L sessions and has been happening for a year+. I monitor the session by pinging my first hop and graphing the results. What I see happening is the D/L session begins. When the line is completely full of data transfer and my first hop ping replies are at their highest, I watch the bottom drop out of the transfer and no more data is transferred, the data light no longer flickers. After several seconds, the data transfer recovers itself and continues the transfer (yea TCP), then again, as the data transfer fills up the line, it will again drop out. The actual line(s) stay connected, I am not bumpped for any reason, I do not have to redial, or reestablish the connection. If I am D/L a 50M file in Windows, I see the D/L stats ramp up to 14KB/s. at the completion of the D/L, my avg. stats will drop to near 4KB/s for the complete D/L session. I was lucky enough to have been able to talk with the Qwest NOC team to discuss this. I discovered two things of interest. 1). I am given a placeholder for the Configure...Rem Addr setting of the Pipeline. I have asked several Qwest people if they can tell me the subnet bit of the placeholder and have come up MT handed. I currently have the placeholder set to 207.225.92.6/27. When I connect, I currently get up to three different gateway addresses, 67.2.0.19, 67.2.0.18, or 67.2.0.16. I believe the only way to be able to have these IPs available is with a /27 or less bit mask setting which is why I have the placeholder set to /27. 2). The Qwest NAS runs load balancing between the three 67.2.0.X addresses. I generally am connected to 67.2.0.16. What I discovered is that my primary SPID line is "linked" to 67.2.0.16, however, the seconday SPID line is offloaded to one of the other two NAS with the use of serial connections between the NAS and a virtual tunnel. I am actually "linked" to two seperate NAS, most if not all of the time. One additional test I tried was to set the 67.2.0.16 as my Configure...Rem Addr address, then keep redialing until it is actually my gateway for the TA. I thought this would clear up the issue based on 1). above. No such luck. I anticipate any response. Thank you, Drew Flickema ++ Ascend Users Mailing List ++ To unsubscribe: send unsubscribe to ascend-users-request at bungi.com Archives: http://www.nexial.com/mailinglists/