Random calls go directly to voicemail

[Chulo]

I've been spending time testing VT today and I've had some interesting results.
I have 2 voip lines: one is Viatalk, the other is from Lingo. The Lingo line I've had since 2005 and I've never once had a missed phone call- ever. Viatalk however, will OFTEN send calls to voicemail with no clear pattern. Tech support has had me buy a linksys router that they claimed would solve the problem (it didn't) and then suggested that the new router may be incompatible too (wow). I then purchased a Linksys SPA-2102 which I connected directly to my modem to eliminate any router interference.

(Internet --> VT adapter --> Lingo adapter --> network)

That didn't work either, but Lingo is still flawless while running behind the VT adapter- which tells me my isp is not the issue. DracoFelis on the boards suggested to shorten my registration time to around 5 minutes. That hasen't worked. Another call to VT and they tell me that since my IP address changes, that it would cause the connection to be lost. I thought this seemed a little fishy considering that it would probably have to affect Lingo too, but I called time warner anyway to ask about a static ip address. They gave it to me half price for 6 months so I figured I'll give it a try and if it didnt work I could just cancel it.
So today, even with static ip- more of the same. I called the VT line 1 one minute after a registration and it went to voicemail. I immidately called again and it rang through (no new registration, it just rang this time for some reason with no change on the network side). I duplicated the error two more times several minutes apart. Each time, the second attempt would go through without a reregistration. Taking everything into consideration (add the fact that my sister has Viatalk running on verizon fios with the exact same issues), I've come to the conclusion that these issues are entirely on the VT side- maybe busy servers? But if that's the case, all of you would be having these problems too right? So who else here is on the Newyork2 server and having this problem? Or if you know of some solution that I haven't yet tried PLEASE let me know (as long as it doesnt involve wasting more money on new equipment) lol.

[bubbanc]

Did you try disabling dynamic routing? I would also find out which one of the VT servers has the best response time from your location. To find their locations, you can use nslookup (from either unix/Linux or Windows from cmd):

nslookup -q=srv _sip._udp.richmond-1.vtnoc.net
Server: 10.25.10.27
Address: 10.25.10.27#53

Non-authoritative answer:
_sip._udp.richmond-1.vtnoc.net service = 0 1 5060 richmond-2.vtnoc.net.
_sip._udp.richmond-1.vtnoc.net service = 10 1 5060 chicago-1.vtnoc.net.
_sip._udp.richmond-1.vtnoc.net service = 10 1 5060 houston-1.vtnoc.net.
_sip._udp.richmond-1.vtnoc.net service = 10 1 5060 newyork-1.vtnoc.net.
_sip._udp.richmond-1.vtnoc.net service = 10 1 5060 newyork-2.vtnoc.net.
_sip._udp.richmond-1.vtnoc.net service = 10 1 5060 sanjose-1.vtnoc.net.
_sip._udp.richmond-1.vtnoc.net service = 0 1 5060 richmond-1.vtnoc.net.

Then you can run pings to these servers and traceroute to them to determine which has the least number of hops and the lowest ping times.

I use richmond-1 as my ping times are around 33ms. newyork-2 is next inline with about 37ms.

I also use my service with a WRT54G and the pap2t behind the router. I forward 69,5060-5080, and 10000-20000 to the static IP address of the pap2t (VT can provision you for a static IP or you can try to setup a DHCP reservation if your router supports it).

Sorry you're having troubles. I too started off with issues but they have seemed to iron themselves out. I had Vonage for 3 years with no issues, so I was a little surprised that I had to deal with issues, but tech support & the folks on this board have been a great help.

Good luck.

[Agrajag]

Today was one of those days. Had two calls that seemed to take a LONG time to connect. Also had several friends call leaving messages saying, "Hey, your phone never rang and went right to voicemail." All of this while I was sitting there either using the phone or next to it just watching it alert me to a new message.

[Chulo]

Today was one of those days. Had two calls that seemed to take a LONG time to connect. Also had several friends call leaving messages saying, "Hey, your phone never rang and went right to voicemail." All of this while I was sitting there either using the phone or next to it just watching it alert me to a new message.

Was it around 6:45? I had a voicemail at 18:47:27, though somebody else got through at 18:48:10.

[Chulo]

Bubbanc, thank you for the input. I pinged all the servers and I think newyork-2 gives me the best results at 10ms with 14 hops (I live in new york). The two richmonds have only 13 hops, but they ping around 15- should I bother changing over? Interestingly, when I try newyork-1 it shows "Pinging Chicago-3" and it pings 31 ms.

[bubbanc]

I would make sure you're configured on newyork-2. I've also noticed that newyork-1 and chicago-3 are one in the same (or at least in the same location as they have the same ping times/traceroute). You can swap your server from the web interface then look under the soft phone configuration to see exactly what server you are on.

[connervt]

newyork-1 now is resolving in Chicago. Last week, it was resolving in NY, but being routed (for me, anyway) first to Level 3's gateway in Chicago, then back to NY.

At least this way now is faster...

[bubbanc]

It looks like whatever was causing their ping times to go up for me has been fixed:

http://forums.hostrocket.com/attachment.php?attachmentid=448&stc=1&d=1178639071
http://forums.hostrocket.com/attachment.php?attachmentid=449&stc=1&d=1178639071

[sbradshaw]

It looks like whatever was causing their ping times to go up for me has been fixed

Can you tell if it was with your provider or was it somewhere else in the path?

[GregM]

Since so many different network providers are involved and their traffic routes are unpredictable, it would be nice to see VT use a webapp that can determine the most efficient route to the most efficient server and set the CP server for that particular user's situation.

[jnuzzi]

Since so many different network providers are involved and their traffic routes are unpredictable, it would be nice to see VT use a webapp that can determine the most efficient route to the most efficient server and set the CP server for that particular user's situation.

Now that would be a useful feature!

[bubbanc]

Can you tell if it was with your provider or was it somewhere else in the path?

I can't tell as I don't remember the path from before, but both now go from Raleigh -> Charlotte -> Atlanta -> Chicago. They are different hosts but I would suspect that they are fairly close by. ny1 is an alias to chicago-3.

Richmond-1 is still the best for me at 14 hops, 34ms average. It goes from Raleigh -> Charlotte -> Atlanta -> DC for me (vs 16 hops and 40ms average with ny2)

[bubbanc]

Can you tell if it was with your provider or was it somewhere else in the path?

I can't tell as I don't remember the path from before, but both now go from Raleigh -> Charlotte -> Atlanta -> Chicago. They are different hosts but I would suspect that they are fairly close by. ny1 is an alias to chicago-3.

I did check and the weekly graph shows them both going up around noon on Sunday, so whatever changed happened then and just got fixed.

Richmond-1 is still the best for me at 14 hops, 34ms average. It goes from Raleigh -> Charlotte -> Atlanta -> DC for me (vs 16 hops and 40ms average with ny2)

[Agrajag]

Just had another one a few minutes ago. Strange as this was from my exterminator and they called just yesterday and got through. Today? Direct to voicemail. Also getting other calls.

[Chulo]

I detached my Lingo adapter today to see if maybe it was causing some kind of interference to the network. It didn't take long to see that it didn't help since 2 calls went to voicemail this morning at around 8:40 (all good since).
Bub, I'm gonna try the richmond server and see if it helps since that's what you use. Also, I used port forwarding and eventually the dmz with two different routers (my sister also uses the dmz and QOS with her Verizon Fios router- no help). I no longer use the router in front of the adapter because I purchased a linksys 2102 connected first in line to the modem. It routs traffic to Lingo (no problems) and my network.

[bubbanc]

Have you tried disabling the dynamic routing in the CP and power cycling you unit?

If you have your admin password you can try a few tweaks that I read in another thread that dealt with registration issues. This may or may not help with your issue but I thought I'd at least mention it: (http://forums.hostrocket.com/showpost.php?p=106306&postcount=52):

Under the SIP tab - Reg Min Expires from 1 to 50, Reg Max Expires 7200 to 1200, Reg Retry Intvl from 30 to 66, Reg Retry Long Intvl from 1200 to 150

Under Line 1 tab: Make Call without Reg from no to yes, Register Expires from 600 to 450

You may need to turn provisioning to NO under the Provision tab or else these will be overwritten by VT's regular config updates. You can always swap this back to YES and power cycle the unit and it will go back to normal.

[Chulo]

Yeah I have those settings in use already. I called lingo to see if they would tell me what might be different and they don't know anything. I keep trying to tell them I just want the settings and they only want to send another adapter. Very interesting now is that I changed my server to Richmond-1 a few hours ago (by changing domain value in the adapter) and for the past 15 minutes the registration is lost. It worked fine at first, but I try rebooting and it wont renew. Do I have to change my server in some other area as well or is this an anomaly? I had pretty good latency to richmond so it should be fine.

Edit- I set it back to newyork and it renewed. I also just disabled dynamic routing now; I haden't tried that yet. If that doesnt work then I'll try richmond again tomorrow.

[Chulo]

I saw in another thread somebody mentioned changing the default sip port (on mine its 5060). I was wondering if anybody using Time warner cable has done this and what has given them the best results. TWC claims they dont block any ports but they sell voip phone service as well, and it would seem to explain my problem (maybe lingo is using a different port). And maybe Verizon is doing it with Fios as well? I guess if one major company can do it they all can since theres no real regulation. I can't believe I haven't gotten this working yet :eek:

[sbradshaw]

And maybe Verizon is doing it with Fios as well?

I personally don't experience any problems doing my VOIP over Verizon FiOS using the default ports. ;)

I have heard of some Internet providers blocking ports 5060 -5061. So it does not hurt to try a different port, IMO, if things are pointing to this being the problem. You can work with VT on getting your ports changed.

[Chulo]

I personally don't experience any problems doing my VOIP over Verizon FiOS using the default ports. ;)

I have heard of some Internet providers blocking ports 5060 -5061. So it does not hurt to try a different port, IMO, if things are pointing to this being the problem. You can work with VT on getting your ports changed.

Yeah, I'm gonna give them a call now and see what they suggest.
Since I'm also trying to solve the issue for my sister who has fios, I was wondering if you woulnd't mind letting me know if you made any extra changes to the router they gave you (assuming they gave you the actiontec 424). we've already assigned a static ip and put it in the dmz with highest qos priority and upgraded firmware on both the router and adapter. I'll have vt change her ports when I call for mine in a few. I'm hoping the ports will do the trick.

[sbradshaw]

Yeah, I'm gonna give them a call now and see what they suggest.
Since I'm also trying to solve the issue for my sister who has fios, I was wondering if you woulnd't mind letting me know if you made any extra changes to the router they gave you (assuming they gave you the actiontec 424). we've already assigned a static ip and put it in the dmz with highest qos priority and upgraded firmware on both the router and adapter. I'll have vt change her ports when I call for mine in a few. I'm hoping the ports will do the trick.

I actually did not do much to get it to work. I use the same router as your sister. I did assign an IP address to my PAP2T from my router based on the MAC address. I also had a new STUN server setting provisioned to my PAP2T ( stun.softjoys.com:3478 ). Additionally, I use my own DNS settings (of my ISP) in my PAP2T instead of the default provisioned ones from VT. With all of this, I have not found the need to put my PAP2T in the DMZ (and no port forwarding as well).

I found that the QOS of this router caused intermittent problems for me when I tried to use it...so I stopped using it some time ago. Quality of my calls has been exceptional with my tweaks. Hope this info helps. Let me know if you have other questions.

[mslaga]

I'm curious if anyone using Viatalk with an ISP that provides VoIP services notices more voice quality issues than others. I have heard that several ISP VoIP providers alter their quality of service settings to non-standard dscp points. That way they can differentiate between their VoIP service and other VoIP services and provide theirs with a higher QoS setting. It would be interesting to sniff someone's outside interface to see if what some of the ISP VoIP providers are marking their VoIP traffic.

[Chulo]

"I have heard that several ISP VoIP providers alter their quality of service settings to non-standard dscp points."
I don't know too much about it, but I know that twc and cablevision do regulate their voip to always have top priority (they told me this when they were trying to sell me the service), though I'm not sure exactly how they do it. As far as the affect on other voip, I can speak from experience that there are no negative effects at all. I've been using Lingo since 2005 and it's actually been MORE reliable than my landline was (used to have idt who took forever to get verizon to fix anything). Having said that, Viatalk has had alot of problems with incomming calls, the source of which I'm not sure of. I changed my SIP port this morning, so if that fixes the issue, then I could say for sure that it was Time Warner blocking the port.

Sbradshaw: Thanks for the info. I too set the static ip in the router as well as in the adapter with the router's dns settings. However the STUN server idea is new to me. Do you think it gets better results than the DMZ? How would I go about getting a stun server? Thanks again for the help!

[mslaga]

In theory, they only match based on DSCP or ToS settings. If you could find out what those are, you could match your device's QoS settings and you would get the same network priority as their VoIP service.

[Chulo]

In theory, they only match based on DSCP or ToS settings. If you could find out what those are, you could match your device's QoS settings and you would get the same network priority as their VoIP service.

That sound's like a great idea. Has it been tested? My brother actually has TWC's digital phone on my network, you think it's possible to get the settings from his modem/ata? Also, I don't want to jump the gun, but I changed my sip port and I haven't had a problem for over a day!
Well, Newyork-2 went down, but thats unrelated lol. I'll give it a week before I say for sure if it fixed the problem :D

[GregM]

You must have brought down the NY server by changing your SIP port.:p;)

Shhhhhtuff happens.

I haven't noticed any problems with VoIP with Vonage or VT riding on Comcast's network. They probably wouldn't want the heat if they were caught being anti-competitive with other providers' traffic.

[Chulo]

I spoke too soon, the port change did NOT fix the problem. I've researched the issue more and I've found many forums around the net discussing this problem. It seems that sometimes, VT just has a problem with their servers that is not user or caller specific. They are either overloaded or they have a software issue that causes sporadic problems with the service.

If anybody at VT is reading this:
Please understand that this problem has worsened over the time that I've had your service. I'm a big fan of Viatalk (I've referred two family members that give me hell for it every day), but the periods of unreliable service are becomming a major cause for concern.
I don't mean to sound like I know what's best for your company, but I feel that if your techs could use their resources to fix bugs in the system rather than developing new features (though I do appreciate the ones you have), you would EASILY become a competitor in the VoIP market simply through word of mouth. If you sell RELIABLE VoIP service near the price you do (regardless of extra features), and then continue to give customers 10% off for referrals, VT could literally become a household name.
Think about this, I referred 2 customers easily right off the bat. If the service would have worked well for us, there would have been a chain of referreals from people looking to save money on their phone, and an influx of new customers. But while a happy customer tells a friend, an unhappy customer tells 10 friends, and the chain ended. Anyway, thanks for your attention and keep up the good work.

[Agrajag]

I again point out that I believe a power user feature here with respect to logging would help identify and solve this problem. If we can point VT to groups of specific examples of this then they can look at the same log entries we're seeing and figure out what happened on their side at that time.

It's a problem now that we cannot reliably reproduce so they'd need to provide some sort of tool to allow us to refer back to it when it happens to us.

We cannot have customers out there getting crap from people for recommending VT over issues like this. This sort of thing should be viewed as a high-priority find at this point. I don't see it getting solved by VT just by poking around here and there with random support calls.

There's nothing worse than sitting right next to the phone waiting for an important call and nothing happening. If you're LUCKY, they leave a message and you call them back. Still looks bad.

[shodge]

Chulo and VIA:Talk,

I started another thread with exactly the same issues. I am having random calls go to voicemail and a second call go right threw and ring. This is an issue. My network has been very stable and I was a user of SunRocket for the past 2 years.

I've been down the path of almost all the solutions listed in this thread. While my IP is dynamic - it has not changed in over 2 months. The SIP port is not the issue as SunRocket did not hove these problems.

I have not tried the time out values posted, prehaps this weekend. I personally would like to know how many others have experienced or are currently experienceing this issue.

-Stan

[MagillaDKE]

I've had similar problems. Random calls go to the "network down" phone number and then we call about 10 min later and all is well. I was with Vonage before w/o these problems, but I might try the DMZ:( suggestion and see what happens.

[Agrajag]

I haven't been able to reach a friend on Vonage in 3 days ("this number is temporarily unavailable.") I'd rather it went to voicemail over that but should do neither.

[bubbanc]

If you have a Unix box on your LAN, you can setup this Perl script in cron to check and see if your pap2t is registered or not. Make sure you change $url to be "http://the_IP_of_your_pap2t" and that the LWP::UserAgent Perl module is installed (perl -MCPAN -eshell then > install LWP::UserAgent). I run it every 10 minutes via cron: */10 * * * * /home/user/bin/pap2tcheck.pl


#!/usr/bin/perl

use LWP::UserAgent;

$url = "http://192.168.1.250";

$request = new HTTP::Request GET => $url;
$ua = LWP::UserAgent->new;
$result = $ua->request($request);
if ($result->is_success) {
$data = $result->content;
if ($data =~ m{Registration State:<td><font color="darkblue">([^<]*)</font>} && (!$reg_state)) {
$reg_state = $1;
}
if ($data =~ m{Last Registration At:<td><font color="darkblue">([^<]*)</font>} && (!$reg_time)) {
$reg_time = $1;
}
if ($data =~ m{Next Registration In:<td><font color="darkblue">([^<]*)</font>} && (!$next_reg)) {
$next_reg = $1;
}
}
if ($reg_state ne "Online") {
print "Device unreachable or unable to register\n";
print "Registration State: $reg_state.\nLast Registration time at $reg_t
ime.\nNext Registration Time in $next_reg\n";
}

[Chulo]

I've been missing calls again today (First time since the 7th).

I called from my cell and wasn't getting through, but I picked up my VT phone and SIMULTANIOUSLY dialed out. The outgoing call connected. Any questions? (or answers?)

As usual Line monitor was good, and All 10 calls I made to the lingo line rang through.
Out of 10 calls to the VT line, only 6 rang through...

Edit- The calls seem to start ringing if the caller waits through 30+ seconds of silence (which no callers will). Check your call logs to see if you were affected.

[KC9FOI]

I was having the same issue. Here are the results of my ticket:
------------------------------------------------------------
ME:
I have been having a number of calls being forwarded to my NETWORK DOWN CALL FORWARDING number when my network is still up. I have tried rebooting all of the components of the network and this still occurs. Some of the instances have occured when I was calling from the same phone that the forwarding is going to. If I were to hang up and try the call again it goes through without problems. The most recent call that did this was a call from "New Brunswic NJ" at 2007-06-19 08:21:35 MST. I called home shortly after this call and I did get the answering machine at my residence so I think it is an issue at Via Talk or with my phone adapter.
Can you check to see if any settings could be adjusted to help avoid this issue. I would like calls to the home phone to be received there. Thank you very much.
Donald.
-------------------------------------------------------------
VT:Hello,

Can you please reply back to this ticket with an updated/current description of how you have your equipment setup? We will need a detailed description of how you have each piece of equipment plugged into each other piece, and in what order. Also, please provide what lights are on solid, blinking, or off, as well as, the color of the lights.

If there is a router in the network, please provide the make, model and firmware version (if available).

How do you have the phone line connected to your phones? Do you have the ViaTalk line connected to the wall jacks in your house? Do you have any splitters on the line? What brand of phones are you using with the service?

Please also include the name of your Internet Service Provider.

Once we receive this information we will be able to further troubleshoot this issue. Thank you in advance for your reply.



ViaTalk Support
-------------------------------------------------------------------
ME:My network is configured as follows.
1. Cable modem is hooked to router.
2. Router is hooked to one computer on port 4 and Phone on port 2. The other ports have not been connected to other devices or in use while this issue has occured. We also were not using the wireless LAN connection and the wireless is password protected.

On the PAP2, The Phone 1, Power, and Internet lights are solid yellow. Phone 2 is off.

The router is a Linksys Wireless-G Broadband Router with SpeedBooster Model WRT54GS with firmware version v4.71.1.

The phone is plugged into 1 splitter which splits the line 3 ways. It is hooked to 2 cordless phones (both V-tech) and 1 wired phone(XACT). The phone adapter is not connected in any way to the existing wiring in my home.

My ISP is Comcast.
-----------------------------------------------
VT: Hello,

To ensure proper communication, a DMZ in the router will need to be set. In order to configure the adapter in DMZ, you will need to follow these instructions:

1) Pick up the phone connected to the ViaTalk adapter, and press **** to access the configuration menu. When prompted for an option number, press 110#. The IP address of the phone adapter will be read back to you. Write this down.

2) Click on the following link: http://192.168.1.1
This will bring you to your Linksys router page. If you are prompted for a username and password and have not changed the administrative password to that device, the default credentials are as follows:

username: (leave this blank)
password: admin

3) Once the page loads, click on Applications and Gaming. Once the sub menu appears, click on DMZ. Click the radio button next to 'Enable', and enter the IP address of the phone adapter that you wrote down earlier into this section.

4) Click on 'Save Settings' at the bottom and your adapter will be configured in DMZ.

Test this out, and if you continue to have issues, let us know. Thank you.

ViaTalk Support

----------------------------------------------------------------

So far, so good. Calls to my VT phone have been ringing more quickly and going to the answering machine. We'll see if this holds up.


EDIT: Problems reoccured shortly after posting this. I have now gone with a static IP and port forwarding in an effort to curb issues. Another thing I noticed, although I don't know if it was related, that my Account in the CP did not show that it was pointing at any server (although my device was still pointing at the Chicago server.) I did point it back at the Chicago server on the account page. I left the ticket open too just to see if they think of another cause.

[Chulo]

EDIT: Problems reoccured shortly after posting this. I have now gone with a static IP and port forwarding in an effort to curb issues. Another thing I noticed, although I don't know if it was related, that my Account in the CP did not show that it was pointing at any server (although my device was still pointing at the Chicago server.) I did point it back at the Chicago server on the account page. I left the ticket open too just to see if they think of another cause.

The CEO of Viatalk posted an update on this problem at http://www.dslreports.com/forum/viatalk.

They found it to be a capacity problem and they're working on it.

[bubbanc]

Shouldn't this be fixed by now?

I had a call go to my network down forward early the other morning (while I was on-call for work). An hour later when they called me back, it went through fine.

Also, I've been having issues where the other party cannot hear me when I make a call, but if I call them back, they can hear me just fine. No, I have not opened a ticket yet because I've disabled provisioning and made some other changes to my pap2t which haven't had any adverse affect until now.

[Agrajag]

A bunch happened yesterday which I think you'll hear more details about shortly. See if anything is better for you now. Things are much improved for me at this point.

[bubbanc]

Nope; wife just called her mother and her mother could not hear her. I'll be opening another ticket... sigh.

[GregM]

There are a lot more details over at DSLreports forum, including remarks from Brendan, than here in the VT forum. I never would have guessed that would be the case.

From what I can tell, SF and SJ servers are offline. Comcast here in Denver jumps over to Level3's network in SF. In the absence of either server in the Bay Area, my preferred server was set to Richmond-1a by VT which drags me all over the country. The server option has been removed from the CP, so I manually changed the PAP2T to Houston-1 since the routing from here to there is much more efficient.

[GregM]

This is happening to me again today. It seems to always be mid-day. I thought it was the Bay Area servers that were the problem, but I'm still pointing my PAP2T to Houston. Last ticket said there was an issue with my SIP port and that they had changed it to resolve issues with Comcast.

Now what's the excuse?:mad:

[Chulo]

This was a capacity problem that was fixed when capacity was upgraded last month.

I haven’t had a problem yet, but I can see why we might see it again with the load of Sunrocket refugees coming to VT.

As long as it’s handled timely and professionally I won’t complain.