drop sip calls after 32 seconds

[mode1]

I have been battling this for awhile at a customer site. Site has IP office R9.1.7. I have the same setup at my office using same sip provider and same release of ip office with no trouble. The sip provider recently changed to a new peering sip server. I believe this is the trouble. I pointed my customer's sip trunks to my office and internet and my sip trunks work and my customer does the same thing with the drop after 32 seconds. I called the provider and they did not have a reason why. They say they see back and forth 200 messages then a bye message. I have attached a monitor trace of the dropped call. Any help would be greatly appreciated.

 

[janni78]

You never recieve an ACK on you 200 OK, probably since your sending your internal IP in the o=UserA 725318007 2398831140 IN IP4 192.168.2.100

"Trying is the first step to failure..." - Homer

 

[Westi]

Firewall?
Try just for fun statically forwarding 5060 to the system

Joe W.

FHandw, ACSS (SME)


"This is the end of the world, make sure to buy your T-shirt before it is too late"
Original expression of my daughter

 

[mode1]

Westi I have done this. I have working sip trunks from same provider on their legacy sip server. I have attached a call using the working sip trunk and hanging up after 33 seconds. I used the same settings as my working sip trunk for the non-working sip trunk.

 

[mode1]

Jani thanks for the reply. What would change then as I have a working sip trunk with the same configuration and same provider bu they went to new sip server?

 

[teletechman]

try turning on RTP keep alives.
Mike

 

[mode1]

I turnrd on keep alives and tried different times.

 

[jamie77]

Your SIP provider is not getting your responses from the system through the firewall, so they end the call as they assume it hasn't connected properly. If you wireshark outside the firewall, you will probably see they try multiple times before ending the call.

SIP ALG/Firewall/STUN. One of these is your issue.

Seen this quite a few times!!

Jamie Green

[bold]A[/bold]vaya [bold]R[/bold]egistered [bold]S[/bold]pecialist [bold]E[/bold]ngineer

 

[mode1]

Sip alg is turned off on the netgear fvs336gv3. I made inbound and outbound rules pointing port 5060 to the phone system internal ip. I am using the following stun server that I ran stun on. IP 146.101.248.221 port 3478. After running stun it comes back full cone nat and it shows my public ip and public port as 5060. I made multiple adjustments to the binding refresh rate and last try was 30 seconds. I am at a loss.

 

[Westi]

If you have a cheap router on hand switch the netgear with it and see if the problem still consists or not.
Then you know if you have to continue troubleshooting the router/firewall or if you should look into the IPO.

I have not seen the IPO doing this without the "help" of a firewall so my bet is that the firewall is doing it but you never know.

try to see if there is a firmware update for the Netgear because 4.3.2-7 only supports 5 VoIP phone calls later releases do 50

Joe W.

FHandw, ACSS (SME)


"This is the end of the world, make sure to buy your T-shirt before it is too late"
Original expression of my daughter

 

[mode1]

Should canuseeme.org or the like work for check if port 5060 is open? West whats weird is that I have working SIP trunks in my office. I pointed the none working ones to my office for testing purposes. The original sip trunks are working and I poseted a monitor trace earlier in my post. Also I posted a trace of the none working trunks.They are both set up exactly the same. Line 17 is working and line 18 none working. I set uri's on both sip trunks to all *'s.

 

[mode1]

I upgraded the firewall to the newest firmware as well.

 

[mode1]

So I put the phone system on the direct internet for testing purposes and low and behold the call did not disconnect. Any Netgear experts out there?

 

[mode1]

So I got this to work. I added to the sip line under transport use network topology info to lan 1. It worked for a day then it stopped working again. I am attaching a monitor trace of a working call. I would greatly appreciate it if someone could look at it and see if it looks good. The co looked in call logs and saw service unavailable.

 

[mode1]

Here is a second trace.

 

[critchey]

You said it worked for a day and then stopped? Is it possible that your public IP address is dynamic? IP address changes and then you lose the connection would make sense here.

The truth is just an excuse for lack of imagination.

 

[mode1]

I have a static ip. Thanks for the response.

 

[fieldtechonIR]

You're not sending the public IP the IPO is behind. Set the topology for the lan you are using to static port block, Enter the public IP the IPO is behind, then set the SIP line to use the topology of that lan port. Binding refresh 30 sec, set the public ports, use a stun server address but don;t run stun.

In this scenario make sure SIP ALG/Transforms are off.

 

[mode1]

FieldtechonIR if I read the knowledge base if I set this way I will have to open all the RTP ports. Is that true or have you set up this way with success

 

[mode1]

I rebooted the firewall and the trunks did not come up. I rebooted the phone system and they started working. It showed they went out of service at 11:59pm. They were working from 11am till then. The weirdest thing about all these issues is that I have sip trunks from the same provider as the troublesome trunks and never have a problem. The difference between the two is that mine are on their legacy switch and the troublesome ones are on their new switch. I sent my sip configs to them and they state that they meet the required settings on the new metaswitch.