[Resolved 8/1/10] 5 - 90 Min into the call and drops out

[Postmeon]

If i am allowed to post links to help hundreds and thousands on here i will start ( just give me the ok )

[volenin]

I don't think you need anyone's permission for this - why should you?

[FONGO_kris]

I assure you if it is anything that you feel would help the other users with their troubles using Freephoneline.ca you are of course obliged to post such links - There is of course this big red "Warn User" button that I may use in the instance your link-posting gets superfluous and unnecessarily frequent

[TSellers]

Just my two bits on the router theory. I've been having lots of router problems in the last little while, resulting over the past month in my using an Asus 15N, Linksys 120, and Linksys 320, and as the 320 is not playing nice with my wireless devices it will also be going back to Costco soon (seems to have the best QOS so far though). I've had the dropped call problem on all of them. There is no set time, tonight a call received at 7:12 PM MST dropped at 7:16, then when he called back it lasted for probably half an hour without dropping. During that time I had no downloading torrents and maybe 20kB/s of uPload torrent traffic. QOS enabled. Voice quality was perfect while it lasted.

I was wondering if uTorrent could somehow affect this problem.

Router: Linksys WRT320N
Modem: Thompson ST 516
ATA: Linksys PAP2
DSL: Telus/Teksavvy, static IP

[bridonca]

Just my two bits on the router theory. I've been having lots of router problems in the last little while, resulting over the past month in my using an Asus 15N, Linksys 120, and Linksys 320, and as the 320 is not playing nice with my wireless devices it will also be going back to Costco soon (seems to have the best QOS so far though). I've had the dropped call problem on all of them. There is no set time, tonight a call received at 7:12 PM MST dropped at 7:16, then when he called back it lasted for probably half an hour without dropping. During that time I had no downloading torrents and maybe 20kB/s of uPload torrent traffic. QOS enabled. Voice quality was perfect while it lasted.

I was wondering if uTorrent could somehow affect this problem.

Router: Linksys WRT320N
Modem: Thompson ST 516
ATA: Linksys PAP2
DSL: Telus/Teksavvy, static IP

See what happens when you limit the amount of simultaneous bit torrent users to something like 50 users. I suspect your problems will go away. Some DSL providers put in a limit of around 200 simultaneous connections, and then disconnects anything extra. It does not discriminate either. Usually last in first out. Bit torrent does not need any more than 50 connections to be effective, but as far as bit torrent is concerned, the more the merrier, except when the DSL provider puts in arbitrary limits.

[FONGO_steve]

Just my two bits on the router theory. I've been having lots of router problems in the last little while, resulting over the past month in my using an Asus 15N, Linksys 120, and Linksys 320, and as the 320 is not playing nice with my wireless devices it will also be going back to Costco soon (seems to have the best QOS so far though). I've had the dropped call problem on all of them. There is no set time, tonight a call received at 7:12 PM MST dropped at 7:16, then when he called back it lasted for probably half an hour without dropping. During that time I had no downloading torrents and maybe 20kB/s of uPload torrent traffic. QOS enabled. Voice quality was perfect while it lasted.

I was wondering if uTorrent could somehow affect this problem.

Router: Linksys WRT320N
Modem: Thompson ST 516
ATA: Linksys PAP2
DSL: Telus/Teksavvy, static IP

See what happens when you limit the amount of simultaneous bit torrent users to something like 50 users. I suspect your problems will go away. Some DSL providers put in a limit of around 200 simultaneous connections, and then disconnects anything extra. It does not discriminate either. Usually last in first out. Bit torrent does not need any more than 50 connections to be effective, but as far as bit torrent is concerned, the more the merrier, except when the DSL provider puts in arbitrary limits.

Most routers will put a limit even smaller than that.
http://www.smallnetbuilder.com/componen ... chart,124/

As you can see, most "cheap" consumer routers may be WELL under 75 max simultaneous connections

[Bloodsong]

The Windows TCP/IP Stack also has some fairly strict connection limitations, so also remember to not have the Softphone on for fear of disconnect requests sent from Windows.

[smartoakville]

And the problem persists.

I find it interesting that we are discussing possible problems with our routers, when FPL already admitted the problem is on their end. I had FPL last year and never had the issues, and when a ton of people started having these issues at the same time, we now question if the problem is our routers? Does not make sense.

[Bloodsong]

The problem isn't where a known issue lies, it's ensuring that we have done all in our power to ensure that's the only issue remaining. This issue has not been affecting everyone, and so to properly troubleshoot who it is affecting, it is necessary to reduce the number of false positives.

[FONGO_steve]

And the problem persists.

I find it interesting that we are discussing possible problems with our routers, when FPL already admitted the problem is on their end. I had FPL last year and never had the issues, and when a ton of people started having these issues at the same time, we now question if the problem is our routers? Does not make sense.

The thing is, looking at our overall userbase (which is larger than most can fathom), compared to the very very small number of reports we've heard of this, we're very unsure if the problem is actually us or not. Other than those mentioning it on the forum here, I have yet to talk to anyone on the phone (who isn't one of you guys) who has this issue (occasionally I poke a question about overall feedback in just to get a feel for things).
However, regardless of that, there are a few upgrades planned, but no one has told me an ETA yet

[beverly]

Update

On PAP2T (NAT Support Parameters)
Turned these on (YES).
Handle VIA received: Handle VIA rport:
Insert VIA received: Insert VIA rport:
Substitute VIA Addr: Send Resp To Src Port:

Changed Router's WRT54GS v1.1 to Airlink AR670W
Enabled QoS to make the ATA a premium and the rest standard

Ran into a problem while testing MagicJack and FreePhoneLine (they talked about feedback). Rogers updated their system the on the 17th of March around 3pm. don't know what their doing (could be updating their system to docsis 3.0) but I got an extra .1 Mb download. (10.7) The bandwidth just dropped to 0 and the modem restarted a few times. Later that day i saw two big roger truck's go by.. their doing construction/renovating at the local cable office (Most Reliable my A$$)

Second problem while testing MagicJack and FreePhoneLine (they talked about feedback) around 3pm on the 19th of March, the battery died on MJ. And sometime during the day MJ and FPL failed their registration (couldn't call either or, answer machine) don't know what happened but i should of waited 180 sec before reseting the ATA.

no problems of the connection dropping yet, but can't be sure since it took 5 day to get the networking back in place (tweaking). Now everything stable except the voip's not registering. plus the girl doesn't talk on the phone 3 to 6, i'm guessing worried the call would drop...


What are the peak time's for voiping?

I really can't see the server being busy during 3 to 6, when everyone is on the road.

[bazza]

Hi,
I've experienced this problem also for several weeks. This moning it happens at least 3 times and the problem is that now the call droped out in les than 1 min. Really frustrating. I don't think that it is the problem with the modem or the ISP, that was not changed.

Thanks for your help

[smartoakville]

Interesting...the post made earlier by one of the staff saying they were aware of the issues and were working on it....I cant find it.

[FONGO_kris]

Which post are you referring to?.. Keep in mind there are 3 pages

[Bloodsong]

Which post are you referring to?.. Keep in mind there are 3 pages

Post 8, page 2, made by you, but with the number and speed of posts you make I'm not surprised if it's in response to another thread haha!

"Isolated the problem will check on it in morning"

At the same time however, as I keep reminding people, there are a lot of factors outside of anything at FPL servers that could cause this issue, many of our loyal thread followers have by now done the suggested testing to narrow the scope of probable issues I'm sure.

[smartoakville]

Again we have isolated the problem and are working our hardest to resolve it, I will check into it this morning and see how the solution is coming along.

The thing is, looking at our overall userbase (which is larger than most can fathom), compared to the very very small number of reports we've heard of this, we're very unsure if the problem is actually us or not. Other than those mentioning it on the forum here, I have yet to talk to anyone on the phone (who isn't one of you guys) who has this issue (occasionally I poke a question about overall feedback in just to get a feel for things).
However, regardless of that, there are a few upgrades planned, but no one has told me an ETA yet

Im confused. Problem has been identified and isolated, yet FPL does not even know if there is a problem !?

[FONGO_steve]

Again we have isolated the problem and are working our hardest to resolve it, I will check into it this morning and see how the solution is coming along.

The thing is, looking at our overall userbase (which is larger than most can fathom), compared to the very very small number of reports we've heard of this, we're very unsure if the problem is actually us or not. Other than those mentioning it on the forum here, I have yet to talk to anyone on the phone (who isn't one of you guys) who has this issue (occasionally I poke a question about overall feedback in just to get a feel for things).
However, regardless of that, there are a few upgrades planned, but no one has told me an ETA yet

Im confused. Problem has been identified and isolated, yet FPL does not even know if there is a problem !?

Keep in mind this thread has been going on for many months.
I think there's been several problems discovered and restored during that time. Most recent update is mine.

[beverly]

no news to update, as no one really use's the phone during 3 - 6 or even long enough.
haven't been dropped yet.
also no complaints from the phone master.

[smartoakville]

no news to update, as no one really use's the phone during 3 - 6 or even long enough.
haven't been dropped yet.
also no complaints from the phone master.

Lucky you. I got dropped 3-5 times today only. Yesterday I was in a middle of a really important call, after finally being transferred to the right department, and the call dropped. Boy I was pissed. My Magicjack is now my main telephone source and I think I will have my FPL as a backup only, as the service has been very unreliable lately.

[FONGO_steve]

I'm going to throw a link to TSellers' one post in here. I'm not saying this is the answer to everyone's problems, but it goes to show just how much the smallest network settings can really affect VOIP:

http://forum.freephoneline.ca/viewtopic.php?f=10&t=1011

[FONGO_steve]

no news to update, as no one really use's the phone during 3 - 6 or even long enough.
haven't been dropped yet.
also no complaints from the phone master.

I looked into our call volume charts, we don't really experience any "peak hours". The majority of calls are made between the hours of 1000 and 2000, but it's pretty steady during that time (while most people are awake, basically). Some hours will jump slightly higher, at random, but not enough to have any affect on this.

[mlaposta]

This has been happening to me as well, ans just happened tonight at 8:04 pm.

Was talking with my wife who was at home, phone cut out after 8 minutes.
Tried to call back but got ring busy.
Tried again and she answered and could hear me, but I couldn't hear her.
Tried again, worked this time for about 12 seconds.
Tried again and rang 3 times and then got busy signal.


This is getting kind of annoying now, was working fine for me like most people for months, but lately been getting these dropped calls.
I realize service is "free", but if you can't rely on the service then it's kind of pointless.

I hope you guys are getting closer to finding something, cause I doubt the issue is on my end, and my ATA is connected straight to my modem.

The call logs seem fine, I'm not showing any failed calls.
Below is the call stamp of the call that just cut off for me:

From To Duration Connect time Disconnect time
514789XXXX 1514552XXXX 0:12 2010-03-25 20:05:05.0 2010-03-25 20:05:17.0
514789XXXX 1514552XXXX 08:14 2010-03-25 19:56:25.0 2010-03-25 20:04:39.0

[volenin]

No, the problem HAS been identified and admitted by FPL (Kris) as a load balancing issue: http://forum.freephoneline.ca/viewtopic ... cing#p4133

I'm not really sure why there are so many discussions around the router settings, when most people who DO experience this issue explicitly mention that they've been using FPL over significant period of time and, with nothing else being changed, the calls just started being dropped...

We really need to distinguish between 'new' subscribers who might not have got the settings right yet (and in this case it can be anything really - router, ATA settings, QoS, etc) AND those ones who had been using the service over long time, tuned all required parameters over time and already HAD experience reliable uninterrupted service some time in the past.......

So, I wonder if the load balancing issue identified by Kris has been already dealt with or not......

[kan]

If I may also add that I am very disappointed with this service. I am also experiencing dropped calls on every call that is made. I guess its called growing pain, I have seen the same with Acanac.

[volenin]

Again,as I wrote on this or the other thread, to be absolutely honest, I personally do NOT _seem_ to experience the dropped calls issue anymore. The only thing I changed in my ATA settings for that was setting 'NAT Keep Alive Msg:' from the default value of '$NOTIFY' to '$PING', as well as changing the NAT keep-alive time settings to 1 min if I'm not mistaken... And, again, the sequence of events was

- everything worked, no dropped calls
- without any changes on my side, I started experiencing dropped calls 2-3 months ago for any calls longer than 10-15 min. No call went uninterrupted for over an hour (ie, the call would drop some time 10 - 60 min into the call)
--- other people reported the same issue on the FPL board within day or two
- issue got escalated to FPL, Kris mentioned that the problem is on FPL side, has smth to do with Load Balancing (which does make sense to me, since the issue is intermittent and quite hard to reproduce)
- while waiting on the fix from FPL, I searched the net and the 'NAT Keep Alive' setting was smth that was worthwhile to try providing the potential root of the problem
--- the default $NOTIFY setting is not going through properly - if you analyze the log, you'd see the socket connection is not established.
--- The documented $REGISTER setting does more harm than good - seems to cause SIP server to reregister line instead of keeping the connection alive
--- the $PING is an 'undocumented' setting (at least I couldn't find it in any Linksys / CISCO manuals) referenced to me by one of the users on this forum - search the threads...
- Around the time I set 'NAT Keep Alive Msg' setting to '$PING' the dropped calls issue seemed to stop happening
--- right after setting NAT Keepalive settings, I did a few tests at night time making calls to and from my cell. Disconnected myself after the connection was alive for over 1.5 hours straight

So...... I can't reliably tell whether $PING solves the issue or helps to solve it (for that, I would need to turn it off again and run the tests - don't have time for this now). It might have been the setting or it might have been the changes FPL did on their side around the same time. In any case, some change or change in load on the FPL side caused this issue in the first place, as we seem to have reliably eliminated the ISP from the chain at that time, as people using other DSL & Cable ISPs (Bell, Acanac, Rogers, etc) started experiencing the same issue around the same time.

Hopefully this would straighten out the historical trace on this one . And as a general recommendation, try playing with NAT Keepalive settings, especially if you are behind a router - some of those settings actually make sure to overwrite the SIP header with proper values... (see attached screen shots).