one way speech on analogue lines 1

[kokolino23]

Hello,
A really odd fault and I'm running out of ideas... I don't know what else can I try.
It's a 3300 v8 anly with IP users. Never had problems on internal calls but external ones (both incoming and outgoing) are one way speech every now and then. It's very difficult to figure out what's causing this as the fault is not permanent. The lines are analogue and there is one way speech only for about 5-10 seconds of the call.
I've tried to reboot a few phones and I get "L2 & BootLoader Error" but then goes pass that and eventually works fine. The DHCP is external and there is a VLAN in place as well. I suspect it's something dodgy on the network but with such an intermittent fault it's almost impossible to catch something with wireshark.

 

[AH64Armament]

This sounds like a Trunking issue - possibly in the Class of Service.
How are the Analog trunks connected to the system? AMB/AOB? ASU-II? U-ASU? LS/GS trunk?

If you are using AMB/AOB or ASU-II, run the Line Quality Test to help ensure you have the trunk settings set correctly.

Are you getting Caller-ID? Try turning off the settings regarding this and make test calls.

Have you actually timed this while looking at a clock? Is the time consistent? If so, look at timers and find one that refers to that time... change it and repeat the test... did the problem time change? If so, there's your problem. If not, put it back.

3300 Rel 8.0 isn't supported any longer. Consider upgrading the system software and possibly hardware. Make sure it's covered under Software Assurance.

Mark

 

[kwbMitel]

Audio Inhibit Until Answer Supervision

This option is used to prevent a DTMF receiver, in line split mode, from being connected between the originating party and the outgoing trunk during the digit outpulsing phase of the call processing cycle. It is used for sites that are tied to very sensitive central offices or remote PBXs. When enabled, the audio path is fully connected [highlight #FCE94F]and the two parties will be able to talk only after the answer supervision has taken place[/highlight]. Select "Yes" if a DTMF receiver, in linesplit mode, is NOT to be connected between the originating party and the outgoing trunk during digit outpulsing. Select "No" if a DTMF receiver, in linesplit mode, IS to be connected between the originating party and the outgoing trunk during digit outpulsing.



**********************************************
What's most important is that you realise ... There is no spoon.

 

[jrgood0852]

Your best troubleshooting for this is with a butt set before the system. Take one line at a time and test. It could be only one with an issue and when this one is hit, bazinga!

Always look out for the next guy because it may be you!

 

[kokolino23]

I did test already the lines with a butt set but couldn't replicate the fault. The most annoying thing is that I could't replicate the fault when I went there. (3 times!) There is no analogue phone available as the phone system is an LX and the trunks are through ASU II.
I've never ran a Line Quality Test but I had a look in the help files and I hopefully can figure out how to do it.
I don't think caller ID is that important as the fault apparently is for both incoming and outgoing. If I rely on the customer though as I can't replicate it at all. It's not time consistent and is the fault I love the most: " every now and then ". Even if I ask them to keep a record, I'm not sure if this helps at all when I looked through SMDR.

Never looked at Audio Inhibit Until Answer Supervision so I need to investigate this.

The DHCP is not provided by the controller though and every single time when I reboot a phone (ANY!), I get "L2 & BootLoader Error" but after a few seconds, goes pass that and trips to the right VLAN then the phoen comes up eventually.

 

[wireman50]

Sounds like a network problem to me or I suppose it could be the TDM interface going faulty...?

 

[LoopyLou]

L2 error is something in their network setup. Few seconds of one way sounds like a trunk issue. Not a lot for you to work with.

The beatings will continue until morale improves.

 

[kokolino23]

I know, that's the most annoying thing! We reported the trunk issue to the telco but obviously it's not their fault. I need to have a word with the IT dept as the DHCP is external. I'm pretty sure it's nothing wrong with the cct descriptor because we've got the same settings in other places and there are no issues at all.

 

[LoopyLou]

More likely to be the trunks themselves. System grounded?

The beatings will continue until morale improves.

 

[kwbMitel]

Could we narrow down the "every now and then" part of your description

My answer was based on several assumptions that have not yet been eliminated. Specifically, is the problem always at the beginning of the call and clears up after 20 seconds or so?



**********************************************
What's most important is that you realise ... There is no spoon.

 

[kokolino23]

Every now and then means that the fault doesn't happen on every call. I was there a few hours and I just couldn't replicated. However, the customer says that sometimes it's OK but they get it quite often ?! As I said, never been able to replicate it myself.

It's not always a problem and could well be just one of the 4th lines. But I did select them one by one and the with a test butt straight into the main DP and couldn't get it to fail. Obviously, the provider says it's not them.

The system is grounded to the main cab.

 

[kwbMitel]

Have you turned off the option I mentioned?

**********************************************
What's most important is that you realise ... There is no spoon.

 

[LoopyLou]

Have you ever selected them individually through the system with direct trunk select access code rather then using a butt set?

The beatings will continue until morale improves.

 

[jrgood0852]

Loopy is on the money!

Always look out for the next guy because it may be you!

 

[MitelInMyBlood]

At the risk of throwing a curve ball into this discussion & looking at it from a Cisco perspective (shoot me), I have seen 1-way audio on analog lines when a thought to be static-assigned address was still inside the DHCP scope and the scope was handing that address out. I've seen this with Cisco ATA-187's as well as on VG202 analog gateways. You swear you have a device problem when it's actually a configuration issue in DHCP.

Original MUG/NAMU Charter Member

 

[kokolino23]

Finally I've got an update...

kwbMitel, I did try to change Audio Inhibit Until Answer Supervision but this made no difference.

Thanks to LoopyLou, as usual:
[link]

http://www.tek-tips.com/viewthread.cfm?qid=1707711

[/url]
Changed Ignore Answer Supervison from Yes to No and bingo! No more delay... I should've found that thread earlier

"L2 & BootLoader Error" - what a tricky one! The controller was on v8 and the handsets were brand new. They had a newest SW and every single time when they booted up were looking for that on the controller. Because the phones are 5312, they are not supported on v8 !!! Even in IP inventory, they came up as 5212... I did a sw upgrade to v9 (where 5312's are supported) and no more errors on boot up.

 

[wireman50]

5312 are backwards compatable to 5212. So you would just set the device to 5212 and bosh it would have worked. I have quite a few 5312 running as 5212 on v8.