Mailing List Archive: Cisco: VOIP

Phone System Outages

Index | Next | Previous | View Threaded

toddnh65 at gmail

Jul 12, 2008, 5:42 AM

Post #1 of 7 (385 views)
Permalink

Phone System Outages

Environment:

CCM 4.1, Unity version 4 also.
3560s, 2801s. 6 buildings all linked by fiber. Each building has a 2801
and 3560.
Total phone count: 60, mostly 7940s, some 7912s, a few 7960s.

All fiber runs back to main location where it joins the network through
GBICs on 3560s.
At the main location we have a PRI that goes into a 2801.

Problem: 3 times in the last year, we will get a situation that goes like
this:
Pressing 9 to dial out, and dialing your number: Phone shows call is
completed, starts counting the seconds, but you hear nothing.
If we call another extension internally, they will hear us but we cannot
hear them.
If someone calls in from outside and gets the auto attendant, no input is
accepted, the attendant just keeps going. (i.e. Press 2 for this, she just
keeps going, you can press as many buttons as you like, no change).

The solution seems to be to reboot the main 2801 with the PRI. I am not
sure why this would cure internal extension-to-extension calls, but it
does. My facility runs 24 hours a day, so a reboot is never a good thing.

Where should I start looking? Broadcast storms? Is this a sort of known
bug that Cisco might have fixed somewhere? Anything in Unity or CCM I
should look at? Counters on ports?

Would appreciate any direction at all!

Todd

toddnh65 at gmail

Jul 12, 2008, 11:56 AM

Post #2 of 7 (358 views)
Permalink

Re: Phone System Outages [In reply to]

Ed, I wouldn't think my internal calls need to go through the 2801. CCM and
Unity hang off a 3560G (along with a 3560POE and the 2801). So my
assumption has always been the call gets 'built' through the CCM, the
'route' is then set, and then CCM can drop out of the picture pretty much.
But I will need to confirm that.

Thanks. I'm going to go dig through the Cisco site now for that document.
My 2801 is running 12.3(11)T9.

Todd

On Sat, Jul 12, 2008 at 1:11 PM, Countryman, Edward <
Edward.Countryman[at]provena.org> wrote:

> Todd do you have smartnet?
>
>
>
> Do your internal (ext to ext) calls have to pass through the 2801?
>
>
>
> Check the Cisco support site; they have section on diagnosing 1-way audio
> issues.
>
>
>
> I have seen port cards (PRT / T1) fail and stop passing the RTP traffic, if
> a reboot fixes it then is is likely a bug in IOS. I would check with Cisco
> if you have smartnet.
>
>
> ------------------------------
>
> *From:* cisco-voip-bounces[at]puck.nether.net [mailto:
> cisco-voip-bounces[at]puck.nether.net] *On Behalf Of *Todd Franklin
> *Sent:* Saturday, July 12, 2008 7:42 AM
> *To:* cisco-voip[at]puck.nether.net
> *Subject:* [cisco-voip] Phone System Outages
>
>
>
> Environment:
>
> CCM 4.1, Unity version 4 also.
> 3560s, 2801s. 6 buildings all linked by fiber. Each building has a 2801
> and 3560.
> Total phone count: 60, mostly 7940s, some 7912s, a few 7960s.
>
> All fiber runs back to main location where it joins the network through
> GBICs on 3560s.
> At the main location we have a PRI that goes into a 2801.
>
> Problem: 3 times in the last year, we will get a situation that goes like
> this:
> Pressing 9 to dial out, and dialing your number: Phone shows call is
> completed, starts counting the seconds, but you hear nothing.
> If we call another extension internally, they will hear us but we cannot
> hear them.
> If someone calls in from outside and gets the auto attendant, no input is
> accepted, the attendant just keeps going. (i.e. Press 2 for this, she just
> keeps going, you can press as many buttons as you like, no change).
>
> The solution seems to be to reboot the main 2801 with the PRI. I am not
> sure why this would cure internal extension-to-extension calls, but it
> does. My facility runs 24 hours a day, so a reboot is never a good thing.
>
> Where should I start looking? Broadcast storms? Is this a sort of known
> bug that Cisco might have fixed somewhere? Anything in Unity or CCM I
> should look at? Counters on ports?
>
> Would appreciate any direction at all!
>
> Todd
>

jonvoip at gmail

Jul 12, 2008, 10:50 PM

Post #3 of 7 (348 views)
Permalink

Re: Phone System Outages [In reply to]

One-way audio is almost always a routing issue, my guess is that you
have a duplicate route to a subnet or a duplicate IP taking over a key
layer-3 routed interface.

Jonathan

On Sat, Jul 12, 2008 at 7:42 AM, Todd Franklin <toddnh65[at]gmail.com> wrote:
> Environment:
>
> CCM 4.1, Unity version 4 also.
> 3560s, 2801s. 6 buildings all linked by fiber. Each building has a 2801
> and 3560.
> Total phone count: 60, mostly 7940s, some 7912s, a few 7960s.
>
> All fiber runs back to main location where it joins the network through
> GBICs on 3560s.
> At the main location we have a PRI that goes into a 2801.
>
> Problem: 3 times in the last year, we will get a situation that goes like
> this:
> Pressing 9 to dial out, and dialing your number: Phone shows call is
> completed, starts counting the seconds, but you hear nothing.
> If we call another extension internally, they will hear us but we cannot
> hear them.
> If someone calls in from outside and gets the auto attendant, no input is
> accepted, the attendant just keeps going. (i.e. Press 2 for this, she just
> keeps going, you can press as many buttons as you like, no change).
>
> The solution seems to be to reboot the main 2801 with the PRI. I am not
> sure why this would cure internal extension-to-extension calls, but it
> does. My facility runs 24 hours a day, so a reboot is never a good thing.
>
> Where should I start looking? Broadcast storms? Is this a sort of known
> bug that Cisco might have fixed somewhere? Anything in Unity or CCM I
> should look at? Counters on ports?
>
> Would appreciate any direction at all!
>
> Todd
>
> _______________________________________________
> cisco-voip mailing list
> cisco-voip[at]puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>
>
_______________________________________________
cisco-voip mailing list
cisco-voip[at]puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-voip

toddnh65 at gmail

Jul 13, 2008, 8:07 AM

Post #4 of 7 (345 views)
Permalink

Re: Phone System Outages [In reply to]

Interesting thought. The 2801 did have some sort of notice about a
duplicate DHCP entry or some such thing....unfortunately, everything was
happening so fast, and there is that need to JUST FIX IT,that there isn't a
lot of time to write stuff down or gather relevant stats.

Looks like I will be doing further research!!

Todd

I think it might be time to set up my cisco gear so it logs relevant
happenings to a syslogger on one of my linux/doze boxes.

On Sun, Jul 13, 2008 at 12:50 AM, Jonathan Charles <jonvoip[at]gmail.com>
wrote:

> One-way audio is almost always a routing issue, my guess is that you
> have a duplicate route to a subnet or a duplicate IP taking over a key
> layer-3 routed interface.
>
>
>
> Jonathan
>
> On Sat, Jul 12, 2008 at 7:42 AM, Todd Franklin <toddnh65[at]gmail.com> wrote:
> > Environment:
> >
> > CCM 4.1, Unity version 4 also.
> > 3560s, 2801s. 6 buildings all linked by fiber. Each building has a 2801
> > and 3560.
> > Total phone count: 60, mostly 7940s, some 7912s, a few 7960s.
> >
> > All fiber runs back to main location where it joins the network through
> > GBICs on 3560s.
> > At the main location we have a PRI that goes into a 2801.
> >
> > Problem: 3 times in the last year, we will get a situation that goes
> like
> > this:
> > Pressing 9 to dial out, and dialing your number: Phone shows call is
> > completed, starts counting the seconds, but you hear nothing.
> > If we call another extension internally, they will hear us but we cannot
> > hear them.
> > If someone calls in from outside and gets the auto attendant, no input is
> > accepted, the attendant just keeps going. (i.e. Press 2 for this, she
> just
> > keeps going, you can press as many buttons as you like, no change).
> >
> > The solution seems to be to reboot the main 2801 with the PRI. I am not
> > sure why this would cure internal extension-to-extension calls, but it
> > does. My facility runs 24 hours a day, so a reboot is never a good
> thing.
> >
> > Where should I start looking? Broadcast storms? Is this a sort of known
> > bug that Cisco might have fixed somewhere? Anything in Unity or CCM I
> > should look at? Counters on ports?
> >
> > Would appreciate any direction at all!
> >
> > Todd
> >
> > _______________________________________________
> > cisco-voip mailing list
> > cisco-voip[at]puck.nether.net
> > https://puck.nether.net/mailman/listinfo/cisco-voip
> >
> >
>

smgustafson at gmail

Jul 13, 2008, 9:20 AM

Post #5 of 7 (346 views)
Permalink

Re: Phone System Outages [In reply to]

This may be an obvious question, but why is there a 2801 at each location?
is it for SRST/911 purposes? Do all IP Phones have the same default
gateway, or do they look to their respective 2801s to provide routing for
them?

Steve

On Sun, Jul 13, 2008 at 8:07 AM, Todd Franklin <toddnh65[at]gmail.com> wrote:

> Interesting thought. The 2801 did have some sort of notice about a
> duplicate DHCP entry or some such thing....unfortunately, everything was
> happening so fast, and there is that need to JUST FIX IT,that there isn't a
> lot of time to write stuff down or gather relevant stats.
>
> Looks like I will be doing further research!!
>
> Todd
>
> I think it might be time to set up my cisco gear so it logs relevant
> happenings to a syslogger on one of my linux/doze boxes.
>
>
> On Sun, Jul 13, 2008 at 12:50 AM, Jonathan Charles <jonvoip[at]gmail.com>
> wrote:
>
>> One-way audio is almost always a routing issue, my guess is that you
>> have a duplicate route to a subnet or a duplicate IP taking over a key
>> layer-3 routed interface.
>>
>>
>>
>> Jonathan
>>
>> On Sat, Jul 12, 2008 at 7:42 AM, Todd Franklin <toddnh65[at]gmail.com>
>> wrote:
>> > Environment:
>> >
>> > CCM 4.1, Unity version 4 also.
>> > 3560s, 2801s. 6 buildings all linked by fiber. Each building has a
>> 2801
>> > and 3560.
>> > Total phone count: 60, mostly 7940s, some 7912s, a few 7960s.
>> >
>> > All fiber runs back to main location where it joins the network through
>> > GBICs on 3560s.
>> > At the main location we have a PRI that goes into a 2801.
>> >
>> > Problem: 3 times in the last year, we will get a situation that goes
>> like
>> > this:
>> > Pressing 9 to dial out, and dialing your number: Phone shows call is
>> > completed, starts counting the seconds, but you hear nothing.
>> > If we call another extension internally, they will hear us but we cannot
>> > hear them.
>> > If someone calls in from outside and gets the auto attendant, no input
>> is
>> > accepted, the attendant just keeps going. (i.e. Press 2 for this, she
>> just
>> > keeps going, you can press as many buttons as you like, no change).
>> >
>> > The solution seems to be to reboot the main 2801 with the PRI. I am not
>> > sure why this would cure internal extension-to-extension calls, but it
>> > does. My facility runs 24 hours a day, so a reboot is never a good
>> thing.
>> >
>> > Where should I start looking? Broadcast storms? Is this a sort of
>> known
>> > bug that Cisco might have fixed somewhere? Anything in Unity or CCM I
>> > should look at? Counters on ports?
>> >
>> > Would appreciate any direction at all!
>> >
>> > Todd
>> >
>> > _______________________________________________
>> > cisco-voip mailing list
>> > cisco-voip[at]puck.nether.net
>> > https://puck.nether.net/mailman/listinfo/cisco-voip
>> >
>> >
>>
>
>
> _______________________________________________
> cisco-voip mailing list
> cisco-voip[at]puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>
>

toddnh65 at gmail

Jul 13, 2008, 5:30 PM

Post #6 of 7 (333 views)
Permalink

Re: Phone System Outages [In reply to]

Yes, the 2801 at each location is for SRST and 911 issues.
All IP phones have the same default gateway. It might be the 2801 or a
3560, I will have to confirm tomorrow when I am back at the office. I think
the 2801 is doing DHCP server duties, something I need to get off of it.

On Sun, Jul 13, 2008 at 11:20 AM, Steve G <smgustafson[at]gmail.com> wrote:

> This may be an obvious question, but why is there a 2801 at each location?
> is it for SRST/911 purposes? Do all IP Phones have the same default
> gateway, or do they look to their respective 2801s to provide routing for
> them?
>
> Steve
>
> On Sun, Jul 13, 2008 at 8:07 AM, Todd Franklin <toddnh65[at]gmail.com> wrote:
>
>> Interesting thought. The 2801 did have some sort of notice about a
>> duplicate DHCP entry or some such thing....unfortunately, everything was
>> happening so fast, and there is that need to JUST FIX IT,that there isn't a
>> lot of time to write stuff down or gather relevant stats.
>>
>> Looks like I will be doing further research!!
>>
>> Todd
>>
>> I think it might be time to set up my cisco gear so it logs relevant
>> happenings to a syslogger on one of my linux/doze boxes.
>>
>>
>> On Sun, Jul 13, 2008 at 12:50 AM, Jonathan Charles <jonvoip[at]gmail.com>
>> wrote:
>>
>>> One-way audio is almost always a routing issue, my guess is that you
>>> have a duplicate route to a subnet or a duplicate IP taking over a key
>>> layer-3 routed interface.
>>>
>>>
>>>
>>> Jonathan
>>>
>>> On Sat, Jul 12, 2008 at 7:42 AM, Todd Franklin <toddnh65[at]gmail.com>
>>> wrote:
>>> > Environment:
>>> >
>>> > CCM 4.1, Unity version 4 also.
>>> > 3560s, 2801s. 6 buildings all linked by fiber. Each building has a
>>> 2801
>>> > and 3560.
>>> > Total phone count: 60, mostly 7940s, some 7912s, a few 7960s.
>>> >
>>> > All fiber runs back to main location where it joins the network through
>>> > GBICs on 3560s.
>>> > At the main location we have a PRI that goes into a 2801.
>>> >
>>> > Problem: 3 times in the last year, we will get a situation that goes
>>> like
>>> > this:
>>> > Pressing 9 to dial out, and dialing your number: Phone shows call is
>>> > completed, starts counting the seconds, but you hear nothing.
>>> > If we call another extension internally, they will hear us but we
>>> cannot
>>> > hear them.
>>> > If someone calls in from outside and gets the auto attendant, no input
>>> is
>>> > accepted, the attendant just keeps going. (i.e. Press 2 for this, she
>>> just
>>> > keeps going, you can press as many buttons as you like, no change).
>>> >
>>> > The solution seems to be to reboot the main 2801 with the PRI. I am
>>> not
>>> > sure why this would cure internal extension-to-extension calls, but it
>>> > does. My facility runs 24 hours a day, so a reboot is never a good
>>> thing.
>>> >
>>> > Where should I start looking? Broadcast storms? Is this a sort of
>>> known
>>> > bug that Cisco might have fixed somewhere? Anything in Unity or CCM I
>>> > should look at? Counters on ports?
>>> >
>>> > Would appreciate any direction at all!
>>> >
>>> > Todd
>>> >
>>> > _______________________________________________
>>> > cisco-voip mailing list
>>> > cisco-voip[at]puck.nether.net
>>> > https://puck.nether.net/mailman/listinfo/cisco-voip
>>> >
>>> >
>>>
>>
>>
>> _______________________________________________
>> cisco-voip mailing list
>> cisco-voip[at]puck.nether.net
>> https://puck.nether.net/mailman/listinfo/cisco-voip
>>
>>
>

dustin.s.fowler at gmail

Jul 14, 2008, 7:14 PM

Post #7 of 7 (322 views)
Permalink

Re: Phone System Outages [In reply to]

I have had similar issues with one way voice for certain phones or
locations.

The solution has been to excluded network addresses from the DHCP pool. I
have many customers using the Cisco DHCP off of a 28XX with no issues. The
key is to exclude address. It doesn't know how to do it itself and will
actually give out the router ip address.

Semper Fidelis,

Dustin Fowler
Senior Cisco Consultant
cid:image001.gif[at]01C8D09B.0C46BD50

Email: <mailto:dfowler[at]data-corporation.com> dustin.s.fowler[at]gmail.com
"KEEPING YOUR BUSINESS HIGHLY AVAILABLE"

From: cisco-voip-bounces[at]puck.nether.net
[mailto:cisco-voip-bounces[at]puck.nether.net] On Behalf Of Todd Franklin
Sent: Sunday, July 13, 2008 8:30 PM
To: Steve G
Cc: cisco-voip[at]puck.nether.net
Subject: Re: [cisco-voip] Phone System Outages

Yes, the 2801 at each location is for SRST and 911 issues.
All IP phones have the same default gateway. It might be the 2801 or a
3560, I will have to confirm tomorrow when I am back at the office. I think
the 2801 is doing DHCP server duties, something I need to get off of it.

On Sun, Jul 13, 2008 at 11:20 AM, Steve G <smgustafson[at]gmail.com> wrote:

This may be an obvious question, but why is there a 2801 at each location?
is it for SRST/911 purposes? Do all IP Phones have the same default
gateway, or do they look to their respective 2801s to provide routing for
them?

Steve

On Sun, Jul 13, 2008 at 8:07 AM, Todd Franklin <toddnh65[at]gmail.com> wrote:

Interesting thought. The 2801 did have some sort of notice about a
duplicate DHCP entry or some such thing....unfortunately, everything was
happening so fast, and there is that need to JUST FIX IT,that there isn't a
lot of time to write stuff down or gather relevant stats.

Looks like I will be doing further research!!

Todd

I think it might be time to set up my cisco gear so it logs relevant
happenings to a syslogger on one of my linux/doze boxes.

On Sun, Jul 13, 2008 at 12:50 AM, Jonathan Charles <jonvoip[at]gmail.com>
wrote:

One-way audio is almost always a routing issue, my guess is that you
have a duplicate route to a subnet or a duplicate IP taking over a key
layer-3 routed interface.

Jonathan

On Sat, Jul 12, 2008 at 7:42 AM, Todd Franklin <toddnh65[at]gmail.com> wrote:
> Environment:
>
> CCM 4.1, Unity version 4 also.
> 3560s, 2801s. 6 buildings all linked by fiber. Each building has a 2801
> and 3560.
> Total phone count: 60, mostly 7940s, some 7912s, a few 7960s.
>
> All fiber runs back to main location where it joins the network through
> GBICs on 3560s.
> At the main location we have a PRI that goes into a 2801.
>
> Problem: 3 times in the last year, we will get a situation that goes like
> this:
> Pressing 9 to dial out, and dialing your number: Phone shows call is
> completed, starts counting the seconds, but you hear nothing.
> If we call another extension internally, they will hear us but we cannot
> hear them.
> If someone calls in from outside and gets the auto attendant, no input is
> accepted, the attendant just keeps going. (i.e. Press 2 for this, she
just
> keeps going, you can press as many buttons as you like, no change).
>
> The solution seems to be to reboot the main 2801 with the PRI. I am not
> sure why this would cure internal extension-to-extension calls, but it
> does. My facility runs 24 hours a day, so a reboot is never a good thing.
>
> Where should I start looking? Broadcast storms? Is this a sort of known
> bug that Cisco might have fixed somewhere? Anything in Unity or CCM I
> should look at? Counters on ports?
>
> Would appreciate any direction at all!
>
> Todd
>

> _______________________________________________
> cisco-voip mailing list

> cisco-voip[at]puck.nether.net

> https://puck.nether.net/mailman/listinfo/cisco-voip
>
>

_______________________________________________
cisco-voip mailing list
cisco-voip[at]puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-voip

Attachments:

image001.jpg (1.66 KB)

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact lists@gossamer-threads.com