atcomsystems.ca/forum Forums TDM Business Telephone Systems NEC Telephones & Systems SV8100 Netlink Reset

We have two sites (siteA and siteB) connected via a VPN. We use a Netlink to connect the PBXs across the VPN. Everything worked fine for many months and then we began having significant trouble with one of the systems (siteB) resetting multiple times per day. We have our vendor looking into it with NTAC, but it's been two weeks and they haven't been able to come up with anything significant (I've done many packet captures, changed a setting here and there, etc). We are getting pretty frustrated with the situation and we're wondering - has anyone else had this kind of trouble? What can we look at? I did a search for "netlink" and found some people that had similar trouble, but there weren't really any resolutions suggested.

Additionally, does anyone know of another forum that has some good NEC resources? This and TekTips is about all I've seen.

You are more likely to receive assistance with a complete profile.

You need to setup your profile AND I would start a ping trace going across the network constantly. Normally the ONLY reason, besides power, that causes the site to reset are latency, packet loss and/or IP conflicts on said network. Also, is your VPN on an MPLS network or is it over the internet?

We had the exact same issue on a 5-site system using Cygnilink (the UX5000 flavor of Netlink).

One of the sites would reset randomly (1-2 times per day) and we could not figure out why.

In the end, we found the the customer had added a WiFi deivce at the site - and assigned it the same IP address as the PBX........

I would certainly check for an IP conflict first!

I added some stuff, sorry

Our VPN is over the internet. I've ran tests with PingPlotter for over an hour and I only see a few packets here and there lost, maybe on the order of 2/1000. I know that voice traffic is much more sensitive to loss, but something that no one has answered for me yet is whether the handshake behaves the same as voice traffic.

IP address conflict was brought up early on, I turned the system off, unplugged it, and then pinged it's IP address and I got no response, so I'm pretty sure that isn't related.

I had an issue last week (another system) where they didn't have the DHCP scope correct and someone was walking in every few days (sales guy) and his Android phone was talking an IP designated for the phone system. You need to run that ping plotter until it happens. Also, running over the internet means you have ZERO QOS control of that VPN. All bets are pretty much off on a setup like that. You've been fortunate that you have "many months" of it working well before this.

What Coral Tech said, going over the internet is a crap shoot. Look into a MPLS

There are many things are cause this your problem

- for severe network loss
- for slow response time networks
- mixture of severe/slow

I suggest you to make below changes on every system

All of the systems should be upgraded to R9 as a first step.
Then while keeping stand alone,
PRG 51-05-02 should be changed to 30s on each site.
PRG 99-03-23 should be changed (to 2 (500ms), this PRG item is only available via keyset
programming)
99-01-70 set to 1 (only available via keyset programming).
Once changes made the systems can be allowed to reconnect back to the primary.