FYI, current theory is customer’s corporate IT gave Verizon the wrong MAC address when they activated the network extender. Should know tomorrow if this was all a wild goose chase.
No matter what I did, I saw ISAKMP packets going out but no apparent response. If the MAC address is coded into those packets, I’m guessing the Verizon VPN servers were just ignoring the VPN tunnel setup requests. From: Eric Kuhnke Sent: Tuesday, February 09, 2016 4:28 PM To: af@afmug.com Subject: Re: [AFMUG] Verizon "network extender" also most modern chipsets and smartphones will pick up data from 2 or 3 of the GLONASS satellites that happen to be visible at any given time, using their timing to augment GPS accuracy. http://www.themoscowtimes.com/business/article/gps-only-devices-to-be-hit-with-tax/421275.html http://russianamericanbusiness.org/web_CURRENT/articles/878/1/GPS-import-duties-to-promote-Russia%92s-GLONASS short version is, russia slapped an extra 25% import tax on every smartphone with a GPS receiver (which is pretty much all of them) that didn't support GLONASS. The RF baseband companies like Qualcomm quickly responded with GLONASS enabled GPS functionality so that their clients could continue to sell in Russia and all of its economically-dependent client states (kazakhstan, tajikistan, etc). On Tue, Feb 9, 2016 at 1:56 PM, Forrest Christian (List Account) <li...@packetflux.com> wrote: Modern GPS chipsets do coldstart within 35 seconds with a good view of the sky, and no previous data. This is the main performance gain a GPS chipset with a significant number of 'channels' will get you. On Tue, Feb 9, 2016 at 2:38 PM, Eric Kuhnke <eric.kuh...@gmail.com> wrote: Same idea as if you take a handheld dedicated GPS receiver (not one that partially uses aGPS / cellular location based assistance) from one side of the world to the other, powered off, and turn it on again... can take 5-10 minutes to reacquire lock even when on a flat rooftop with view to 12 satellites. On Tue, Feb 9, 2016 at 6:50 AM, Adam Moffett <dmmoff...@gmail.com> wrote: It might not be just a matter of getting the location. If they use the 1pps clock from GPS to calibrate an oscillator before they start transmitting, then it would legitimately take 20-30 minutes. Telrad BTS's are like that too. Pisses me off if I ever have to reset the power. On 2/9/2016 12:12 AM, Jason McKemie wrote: For whatever reason, the receivers that they use in some of these don't seem to be "modern" at all. They frequently take an excessively long time to get a lock. On Monday, February 8, 2016, Eric Kuhnke <eric.kuh...@gmail.com> wrote: Modern GPS receivers work surprisingly well, if not very accurately, from inside a single floor wood framed house... My oneplus one will pick up 6 satellites while standing in a central hallway 15'+ from any window. Should be accurate enough to get a location within 75'. All bets are off if it is a concrete framed apartment building or something like that. I still find it amazing that anything works at -162 RSL. Thanks to tiny channel size and very basic modulation. On Feb 8, 2016 6:46 PM, "Bill Prince" <part15...@gmail.com> wrote: Canopy NAT seems to break it with regularity. It might also fail if the GPS location that it reports is not within a 1/4 mile of where the customer address is. Also requires enough GPS (like near a window) to get a GPS lock. bp <part15sbs{at}gmail{dot}com> On 2/8/2016 3:34 PM, Ken Hohhof wrote: What are the typical reasons for these not to work?� From the user guide it appears to use IPSEC, so I assume anything that prevents a VPN? � Verizon support told the customer they needed a Class A address.� WTF?� Did they maybe mean it can't be a class A address?� Customer uses 10.x.x.x addresses internally, behind Cisco ASA firewall (which I don't manage). � I do see some udp/500 and udp/4500 packets, I think that means something is using UDP for IPSEC NAT traversal? -- Forrest Christian CEO, PacketFlux Technologies, Inc. Tel: 406-449-3345 | Address: 3577 Countryside Road, Helena, MT 59602 forre...@imach.com | http://www.packetflux.com
