Network Connection Failed between Offices 4

[DTGMI]

Have (2) office that have been running for a couple years without errors.

This week the link between offices has failed?

Zero changes on any components - SW programming or network equipment.

Net4 licenses are valid on both systems.

PRI cards have been temp swapped out

Internet circuit hasn't been changed and is working ...

Onsite tech has been troubleshooting for hours and still no resolution.

Is there a chance either of the SD cards are corrupted.

Thanks!

 

[AACon]

So the data connection is good, but, data has a 'retry' function where as voice (h.323 that the IPO uses for SCN) does not and those packets get lost. Make sure there is no packet loss between the locations. This can often be seen with a wireshark capture - you'll get a lot of data packet resends.
We had an issue like this at a customers site. Pulled our hair out swapping hardware and firewalls until we got a pcap and saw a bunch of retries and then verified with the carrier that yep...they had a bad router in the line somewhere (not customer, carrier).

Packet loss went away, SCN came right back up.



-Austin
I used to be an ACE. Now I'm just an Arse.

 

[IPGuru]

how exactly is the link between sites connected (please do not say you are forwarding public IP addresses to the IP Office)
you say Zero changes, but this will NOT be the case

either something has changes on your configuration of the link, or the link provider has changed something in their network



Do things on the cheap & it will cost you dear

ACSS

 

[tylamb19]

AACon is absolutely right, run a consistent ping and a packet capture. Make sure a traceroute is clear of timeouts. Troubleshoot this as if it is a network issue. I've seen this many times (not just in voice, but also in data connections as well, but voice is much more sensitive to packet loss) and usually it turns out somewhere along the line there is a failing router, bad switch, something. Could be on your side, or on the carrier side.

Once I spent days and hours troubleshooting a sporadic drop in a terminal services connection for a client, ordering 24 hour RMA turnarounds, etc and it ended up being that Comcast had a bad line card in some switch somewhere in one of their datacenters. It took a bit of convincing to make them sure of the issue but it did turn out to be exactly where I narrowed it down to in Comcast's infra. It makes me check packet loss early on every time now.