Sorry if this gets posted twice, it didn't seem to make it so I put the file on
a web server instead of attaching.
What's under devices/network parameters for NTP?
1.pool.ntp.org Enabled 1.pool.ntp.org NTP
2.pool.ntp.org Enabled 2.pool.ntp.org NTP
Finally found a lead. We had been waiting to
Do us all a favor and consider changing your terminology here.
dropped and is in the middle of a call and dropped the call, and lost
(or lose) registration as in I was registered and the registrar seemed to
determine that my registration was no longer active.
ensure your cron jobs have nothing
Just to clarify a bit, Tony, I think you mean that the below command is
relevant in this case:
And of course, I set the hardware clock using the system time;
hwclock --systohc
It could be that on Mike's system the hardware clock is set every hour or
so by the hwclock command, basically pushing
On 2010-07-07 0:26, m...@grounded.net wrote:
It happened again about 5 minutes ago. I grabbed a snippet after all of the
registered phones had dropped. At the time I took this, there were no phones
registered.
You provided approximately 33 seconds of log data, and did not tell us
the time
Hi Ranga,
I am positive that RTP is being buffered (in loss for better word). I clearly
see rtp coming into port 30508 (as seen in my example) and being buffered.
This happens consequently until 180 Ringing comes x seconds later. Then and
only then are the buffered rtp packets relayed to the
I have a customer who needs the ability to record calls (he is a lawyer). It
does not appear that Sipx has this capability, so he was going to use the call
recording feature that is available on some Polycom phones which have a USB
port (IP 650/670). I believe the USB port connects to a PC
Er... Just keep in mind that if this running as a virtual somewhere/somehow
(EC2 included), that latency (storage method, etc.) and media might result
in what you are seeing.
You might find different results by using 4.2 in the same basic
configuration.
Tony Graziano,
Use it all the time. You just have to sometimes try/fail on a usb stick that
works.
Tony Graziano, Manager
Telephone: 434.984.8430
Fax: 434.984.8431
Email: tgrazi...@myitdepartment.net
LAN/Telephony/Security and Control Systems Helpdesk:
Telephone: 434.984.8426
Fax:
On Wed, Jul 7, 2010 at 9:35 AM, Jeff Gilmore j...@thegilmores.net wrote:
I wonder if anyone has tried this? In particular, are there any gotchas with
either firmware versions needed or with managing the profile through sipx
while using this feature?
every settings should be available to
The recording functionality works very well. Polycom offers a 90 day (I think)
trial to test out the capabilities - available from their website
(http://www.polycom.com/global/downloads/products/voice/desktop_solutions/soundpoint/eval_license_for_prod_suite.zip).
Don't want to muddy the water here... I don't find any issue running bootrom
4.2.x with 3.1.3RevC. If you are running 4.1 or later firmware you should
have no problem running 3.1.3 though.
I think the polycom apps requires fw 3.0 or above. So whatever you are on is
likely ok if that is the case.
You provided approximately 33 seconds of log data, and did not tell us
the time window during which the registrations disappeared.
Didn't think you would want it because the relevant part seems to be when the
problem occurs. It appears to be a timer related issue.
I found some mentions of what
How about fixing the timing on your blade center? I think lots of people
will be happy to have that working properly!
Once that is fixed let's wait to hear if you have anymore lost
registrations.
On Wed, Jul 7, 2010 at 10:23 AM, m...@grounded.net m...@grounded.netwrote:
You provided
On 2010-07-07 10:23, m...@grounded.net wrote:
Just prior to the problem being repeated, everything looked normal, we saw no
errors anywhere. This is why I didn't post the entire log and only a few
seconds prior. When we were able to duplicate the problem, we were watching
wireshark,
I read somewhere (?) where sipx was being used to record phone calls. Is
that possible?
On Wed, Jul 7, 2010 at 9:45 AM, Tony Graziano
tgrazi...@myitdepartment.netwrote:
Don't want to muddy the water here... I don't find any issue running
bootrom
4.2.x with 3.1.3RevC. If you are running 4.1
That should be a different thread...
Tony Graziano, Manager
Telephone: 434.984.8430
Fax: 434.984.8431
Email: tgrazi...@myitdepartment.net
LAN/Telephony/Security and Control Systems Helpdesk:
Telephone: 434.984.8426
Fax: 434.984.8427
Helpdesk Contract Customers:
Maybe running a script like this will shed some light on your time issues.
for i in {1..5000}
do
date /tmp/timecheck
sleep 1
done
This should output the time every second to a file for 5000 seconds. It
may give you an idea of when your clock is going haywire.
On 7/7/2010 9:28 AM, Tony
When, but not why.
He is using a blade center, so the chassis is managing an aspect and the
two are fighting each other.
I think we know where the problem lies. Let him get the chassis to play nice
and see what happens from there.
Tony Graziano, Manager
Telephone:
I know. I figured the when may possibly shed some light on the why and
at least prove it is happening.
On 7/7/2010 10:03 AM, Tony Graziano wrote:
When, but not why.
He is using a blade center, so the chassis is managing an aspect and the
two are fighting each other.
I think we know where
On Wed, 7 Jul 2010 05:32:26 -0400, Tony Graziano wrote:
Do us all a favor and consider changing your terminology here.
Sure, so what should I call it then? I've called it dropped, but always said
from the list in the Active Registration screen.
as in I was registered and the registrar seemed
I placed the info on windows and timesync with a blade chassis/blade center
only because ibm made it more readily available. I don't know what the
equivalent with linux would be, but IBM should.
Tony Graziano, Manager
Telephone: 434.984.8430
Fax: 434.984.8431
Email:
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
?From Dale's observation, you have another cron job somewhere that's
?resetting the clock every hour.
Fair enough
Thanks Tony, two of my cases are running EC2, the third one is on a high
spec'ed server, so I am sure it is not down to that. The buffered
media is released when 180 Ringing is received, dead certain. :)
Sven
-Original Message-
From: Tony Graziano [mailto:tgrazi...@myitdepartment.net]
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
Here it is, the complete run. I deleted the old merged.xml file just to be
100% sure.
# merge-logs
# ls -la merged.xml
-rw-r--r-- 1 root root 79
I thought I read somewhere (?) that a sipx user can record selected calls
from their extension.
Josh Patten provided this info when system wide recording is needed:
http://oreka.sourceforge.net/ can be used as well but it records everything.
However, is it possible for an individual user to
Content-Type: text/plain;
charset=utf-8
Content-Transfer-Encoding: 8bit
Organization: SipXecs Forum
In-Reply-To: be12.4c302...@forum.sipfoundry.org
X-FUDforum: 08063afcdd00a6e76393c5b9527381e8 48798
Message-ID: be9e.4c34a...@forum.sipfoundry.org
I confirm that there is definately a problem
On Wed, 7 Jul 2010 10:28:42 -0400, Tony Graziano wrote:
How about fixing the timing on your blade center? I think lots of people
will be happy to have that working properly!
Once that is fixed let's wait to hear if you have anymore lost
registrations.
Fair enough, I've asked in the IBM
SipXrelay will not just relay whatever media is sent to it. It will
only relay media after it knows where to send the media. Please
examine the signaling carefully using sipviewer and take note of what
point in the processing this can be asserted. Look at the signaling
for the sip call setup.
No.
maybe if you have conference recording you can record from a conference.
or you can use oreka and record everything.
or you can use a softphone that has that enabled (bria) or hardphone
(polycom with app license).
On Wed, Jul 7, 2010 at 11:50 AM, Mike Haun mh...@mwpo.net wrote:
I
To Tony's point - perhaps you can setup a conference that doesn't have the 'you
are the only person on this conference - followed by the music'. If that is
possible, then you could do a phone based 'conference' with the actual
conference room while the person is on there with you. That might
But those lines were not in what you sent before.
My last post from this morning at 2am contained the error lines mentioned. It's
become a crazy long thread so hard to keep up.
If this is not what you're talking about, sorry, I'm working on lack of sleep
and a great deal of stress so easy to
1) Set your logging level to INFO (or DEBUG if you like that better).
It is set to info or, all default at this time.
2) Wait for the problem to be manifest, that is, a phone is first seen
and then later not seen on the Active Registrations screen without any
explanation.
As suggested in
Thanks for the tip Ranga! I will continue using my SIP microscope!
Sven
-Original Message-
From: M. Ranganathan [mailto:mra...@gmail.com]
Sent: 07 July 2010 16:54
To: Sven Evensen
Cc: Tony Graziano; sipx-users@list.sipfoundry.org
Subject: Re: [sipx-users] RTP being buffer in sipX,
I am following everyone's suggestions. I am providing what ever output is
required of me.
Would be happy to send my snapshot to someone who may be able to check it as
the merge-logs problem is one of the original issues.
___
sipx-users mailing list
Please check the sipregistrar.log for the microseconds fired message and
tell us if you again see that message.
Tony Graziano, Manager
Telephone: 434.984.8430
Fax: 434.984.8431
Email: tgrazi...@myitdepartment.net
LAN/Telephony/Security and Control Systems Helpdesk:
for i in {1..5000}
do
date/tmp/timecheck
sleep 1
done
This completed but of course, it's a large file and hard to know if any timing
was missed.
Anyone know of a command or little to check this output file now?
The output looks like;
Wed Jul 7 10:23:37 CDT 2010
Wed Jul 7 10:23:38 CDT
Zip the timecheck file and send it on. It should compress significantly.
On 7/7/2010 11:55 AM, m...@grounded.net wrote:
for i in {1..5000}
do
date/tmp/timecheck
sleep 1
done
This completed but of course, it's a large file and hard to know if any timing
was missed.
Anyone
On 2010-07-07 12:27, m...@grounded.net wrote:
1) Set your logging level to INFO (or DEBUG if you like that better).
It is set to info or, all default at this time.
2) Wait for the problem to be manifest, that is, a phone is first seen
and then later not seen on the Active
On 2010-07-07 12:55, m...@grounded.net wrote:
for i in {1..5000}
do
date/tmp/timecheck
sleep 1
done
This completed but of course, it's a large file and hard to know if any
timing was missed.
Anyone know of a command or little to check this output file now?
The output looks
On Wed, 7 Jul 2010 12:53:47 -0400, Tony Graziano wrote:
Please check the sipregistrar.log for the microseconds fired message and
tell us if you again see that message.
Yes, it is in there. I had rotated the logs when I started the test and wasn't
able to get past the merge-logs problem again.
On 2010-07-07 13:30, m...@grounded.net wrote:
What is interesting is that it happened around 11:01 or so and looking at the
sipregistry.log file, the times are totally weird as was identified yesterday.
I have ntpd turned off at the moment.
Looking at the above file, we see when we roll
I believe this is evidence of your problem It jumped ahead 1 hour.
Something is still resetting your time. Sipx doesn't change the time on
your OS.
I would expect that to cause a lot of problems.
From the logfile you attached:
Wed Jul 7 11:00:58 CDT 2010
Wed Jul 7 11:00:59 CDT 2010
Wed
Thanks guys. I appreciate it very much.
I will experiment with the conference approach and the softphone.
I am also going to try the Polycom 670 and/or the 1500.
___
sipx-users mailing list sipx-users@list.sipfoundry.org
List Archive:
After looking a little more at the logfile, I noticed it then dropped
back an hour. You must have a cron job or something else interacting
with the system in some way that is changing the time. With you being
the one that has access to the system, I don't know how anyone here can
help you
I think the ibm chassis is doing this and he needs ibm guidance to solve it.
If he deleted his ntp entry in cron and reloaded crontab it won't make any
difference.
This is like a virtual guest syncimg time to the hardware kernel of the
host. You won't see it, only that it happens.
Moving it to
On Wed, 07 Jul 2010 12:41:05 -0500, Matthew Kitchin (public/usenet) wrote:
I believe this is evidence of your problem It jumped ahead 1 hour.
Something is still resetting your time. Sipx doesn't change the time on
your OS.
Maybe it's a bug because I had not seen this before it happened yet
See my next email. It is going backwards and forwards. Something has to
have rights, a driver, a cron job, etc to change your clock. Your system
time is being changed by something.
On 7/7/2010 12:47 PM, m...@grounded.net wrote:
On Wed, 07 Jul 2010 12:41:05 -0500, Matthew Kitchin
See my next email. It is going backwards and forwards. Something has to
have rights, a driver, a cron job, etc to change your clock. Your system
time is being changed by something.
Do you mean the following message that you posted?
On Wed, 07 Jul 2010 12:44:33 -0500, Matthew Kitchin
Yes.
-Original Message-
From: m...@grounded.net m...@grounded.net
Sender: sipx-users-boun...@list.sipfoundry.org
Date: Wed, 7 Jul 2010 13:01:40
To: sipx-userssipx-users@list.sipfoundry.org
Reply-To: m...@grounded.net
Subject: Re: [sipx-users] Random Dropped Registrations
See my next
Yes.
Ok, wanted to make sure I didn't miss something.
I made note that the last time the problem happened was around 11:01.
Notice that I started your test script to at 10:23, and it ended at 11:47.
The output from that test shows that system time never went backwards.
However, the
The system time is going backwards and forwards. That is a fact. See
below. Sipx is not changing your time. You need to figure out what is
changing your time. That will solve many or all of your problems.
Wed Jul 7 11:00:55 CDT 2010
Wed Jul 7 11:00:56 CDT 2010
Wed Jul 7 11:00:57 CDT 2010
Wed
Is there a setting similar to:
http://www-03.ibm.com/systems/i/advantages/integratedserver/faqinstall.html#timesyc
This message and any files transmitted with it are intended only for the
individual(s) or entity named. If you are not the intended individual(s) or
entity named you are hereby
On 2010-07-07 14:19, m...@grounded.net wrote:
Yes.
Ok, wanted to make sure I didn't miss something.
I made note that the last time the problem happened was around 11:01.
Notice that I started your test script to at 10:23, and it ended at 11:47.
The output from that test shows that
On Wed, 07 Jul 2010 13:24:05 -0500, Matthew Kitchin (public/usenet) wrote:
The system time is going backwards and forwards. That is a fact. See
below. Sipx is not changing your time. You need to figure out what is
changing your time. That will solve many or all of your problems.
Yes, you're
There is no doubt whatsoever that your clock is being changed by
something, and that is the root of your major problem. Quit wasting
time wondering if it is strange and find what's messing up your clock.
Scott, I'm not trying to waste any time, I'm trying to make sure I have input
for the
I'm going to install another server so I can see if this happens on a stand
alone.
In getting things ready, I noticed that I was still using the 32Bit ISO while
the blades are 64Bit.
I did so because people have said this should not be an issue but is it
possible that it is related?
I'm
Content-Type: text/plain;
charset=utf-8
Content-Transfer-Encoding: 8bit
Organization: SipXecs Forum
In-Reply-To: bac9.4c17f...@forum.sipfoundry.org
X-FUDforum: 08063afcdd00a6e76393c5b9527381e8 48831
Message-ID: bebf.4c34f...@forum.sipfoundry.org
I just found your posts, after doing a search.
I'm guessing sipx 4.0.4? I've had no problems with 5.4.
On 7/7/2010 5:17 PM, David wrote:
Content-Type: text/plain;
charset=utf-8
Content-Transfer-Encoding: 8bit
Organization: SipXecs Forum
In-Reply-To:bac9.4c17f...@forum.sipfoundry.org
X-FUDforum: 08063afcdd00a6e76393c5b9527381e848831
You should reset to defaults and login and set the IP/mask/gateway manually.
Once you jave done that you should config in sipx and upload the config to
the device (don't forget to burn thr config when saving changes).
Anything else you do, or try to do, will not matter until you get into the
Hi, no idea if someone else met the issue of registration expire
discrepancy. Please help if you have same experience.
I made 100 subs to initiate registration from SIPP to SIPX with expires:
300 inside SIP Invite. Whereas, I observed in SIPX the expire timers of all
these 100 subs varies from
Hmmm... maybe a new call record service. Program a button to initiate a
three party call to it... or have a user selectable 'record all calls'
option and have every call to / from the extension be recorded / deposited
in VM or e-mailed.
On Wed, Jul 7, 2010 at 1:41 PM, Mike Haun
endpoints vary... when an endpoint or ua is going to expire there is
generally a setting that begin to retry when xx seconds are left on
the registration.
On Wed, Jul 7, 2010 at 8:54 PM, Wen Jun jun.wen.s...@gmail.com wrote:
Hi, no idea if someone else met the issue of registration expire
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of Wen Jun
[jun.wen.s...@gmail.com]
I made 100 subs to initiate registration from SIPP to SIPX with expires:
300 inside SIP Invite. Whereas, I observed in
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
?6) Do merge-logs --ft=[extension] --include-method=register, where
?[extension] is the extension of the phone in question. ?This extracts
?the
Well, that protection mechanism to avoid surf registration load to SIPX is
understood. That seems a pretty wise fence.
My another question is howwhen SIPX knows to give phones by random
registration timers ?
-Original Message-
From: WORLEY, Dale R (Dale) [mailto:dwor...@avaya.com]
If you have observed a phone disappearing from the Active
Registrations screen (as instructed in step (2)), you'd know which
phone was involved:
There were about a dozen phones registered in the list and when the problem
happened, it would take them all out.
If you are having registration
66 matches
Mail list logo
