> by far the most frequent poster), and almost all of yours are completely
> lacking anything like sufficient information or clarity that anyone can
I provide what information I can based on the input I get. I provide everything
I'm asked for and don't hold back. As you know, I also have some pro
On Tue, 06 Jul 2010 22:51:14 -0500, Josh Patten wrote:
> Does merge-logs need to be run within the /var/log/sipxpbx directory? I
> seem to recall that being necessary.
From what I understand, it does, and I do.
___
sipx-users mailing list sipx-users@li
Does merge-logs need to be run within the /var/log/sipxpbx directory? I
seem to recall that being necessary.
On 07/06/2010 10:09 PM, m...@grounded.net wrote:
Fix your clock - make sure that all timestamps in the logs are moving
forward.
This is the task at hand for sure.
your
> Fix your clock - make sure that all timestamps in the logs are moving
> forward.
This is the task at hand for sure.
> your ntp server sync stays stable over at least several hours.
So far, both hardware and OS are perfectly on time.
> Until you do this, _NOTHING_ else is worth spending time o
> No - all sipXecs logs are deliberately recorded only in UTC. That
> allows us to correlate events that occur in servers regardless of what
> timezone the servers are in.
Ok, so then it's normal for the logs to have that time. Guess I never noticed
that before. Never searched by time I guess, o
> You are using NTP incorrectly and it is messing up your clock, which is
> then messing up registrations. Actually, all kinds of things will be
> fowled up by this - it's amazing you have not had other problems.
Ok, I have no issues with being wrong about something but there's only one way
to l
On 2010-07-06 22:17, m...@grounded.net wrote:
> I've not cleared my logs in a day or two now and decided to double check that
> merge-logs in indeed installed, as I'm sure I ran it.
> This time, it took a rather long time to run so figured perhaps it didn't
> have enough logs when I've run it b
> I don't see the command line that executed merge-logs. Did it report a
> "not found" error as you say it has done in the past? Show us what
> happened!
What is there to show??? It doesn't output anything, it just creates the
merged.xml file.
As I said, I want to start a different thread for
On 2010-07-06 22:35, m...@grounded.net wrote:
> I've never seen this on any other server I've got so this is beyond my
> knowledge.
> I'm central US time, I have an offset of -360 set in the GMT Offset in sipx.
> This is what I have set in ntp.conf.
>
> So, I have a UTC setting somewhere which i
On 2010-07-06 22:36, m...@grounded.net wrote:
> So I have a UTC and GMT conflict somewhere?
No - timezones have nothing to do with it.
You are using NTP incorrectly and it is messing up your clock, which is
then messing up registrations. Actually, all kinds of things will be
fowled up by this
Are you losing your mind or something and just picking me to blame?
> I have just asked you to run certain commands and report the output. You
> have not, in fact, then run the commands and reported the output, you have
> simply said that you have previously done so in some place that you don't
>
Wow Dale, that's is one very weird reply. Why even help if all it does it piss
you off?
Man that's just too weird.
> I have no idea. But I have no interest in solving your NTP problems.
> Presumably you are responsible for doing so. When you get the time problem
> fixed, see if you still get
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
I've not cleared my logs in a day or two now and decided to double check that
merge-logs in indeed installed, as I'm
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
So I have a UTC and GMT conflict somewhere?
I have no idea. But I have no i
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
>?In this instance you've said "The merge-logs command and the regtimes
>?command don't [exist] on the ISO", but you h
I removed the script I had in there which was setting the time every hour so as
far as I know, there is nothing else.
And the time servers are just ntp.org, known good time servers.
On Wed, 7 Jul 2010 02:34:57 +, Matthew Kitchin (Public) wrote:
> I'm not a sipx internals expert. I'm still g
So I have a UTC and GMT conflict somewhere?
On Tue, 6 Jul 2010 22:28:33 -0400, WORLEY, Dale R (Dale) wrote:
>
>
> From: sipx-users-boun...@list.sipfoundry.org [sipx-users-
> boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
> [m...@grounded.net]
I've never seen this on any other server I've got so this is beyond my
knowledge.
I'm central US time, I have an offset of -360 set in the GMT Offset in sipx.
This is what I have set in ntp.conf.
So, I have a UTC setting somewhere which is overriding?
On Tue, 06 Jul 2010 22:30:38 -0400, keccle
I'm not a sipx internals expert. I'm still guessing that huge time jumps are
causing this. I would make sure all your time synch mechanisms besides default
ntp daemon are disabled and then reboot.
-Original Message-
From: "m...@grounded.net"
Sender: sipx-users-boun...@list.sipfoundry.or
On Tue, 2010-07-06 at 21:25 -0500, m...@grounded.net wrote:
> I'm sort of back to something I noticed a while back in this thread, when I
> mentioned that the server has the right time but the live logs don't. I'm
> watching the sipregistrar.log going by, the system has the proper time but
> the
We've not seen a pattern yet but I'm watching to see if that is related. For
example, maybe as someone mentioned before, one of the phones is taking them
all down.
On Wed, 7 Jul 2010 02:28:46 +, Matthew Kitchin (Public) wrote:
> Did they countdown to 0 and then disappear or simply disappear
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
>?Get the sipregistrar.log from one of your failure windows and post it
>?somewhere to which you can provide a URL, th
Did they countdown to 0 and then disappear or simply disappear no matter how
much time they had left before they were supposed to expire/re-register?
-Original Message-
From: "m...@grounded.net"
Sender: sipx-users-boun...@list.sipfoundry.org
Date: Tue, 6 Jul 2010 21:21:11
To: sipx-users
I'm sort of back to something I noticed a while back in this thread, when I
mentioned that the server has the right time but the live logs don't. I'm
watching the sipregistrar.log going by, the system has the proper time but the
logs don't!
# date
Tue Jul 6 21:24:59 CDT 2010
Then tail the log
Same as explained before, dropped, vanished, from the list of Active
Registrations.
Slowly but surely, they have all come back and I'm watching to see when it
happens again.
I'm also monitoring the sipregistrar.log file at the same time.
The OS and hardware time are still correct.
On Wed, 7 J
I've not cleared my logs in a day or two now and decided to double check that
merge-logs in indeed installed, as I'm sure I ran it.
This time, it took a rather long time to run so figured perhaps it didn't have
enough logs when I've run it before.
Thing is, even though it definitely took longer t
What does dropped mean in this specific case?
-Original Message-
From: "m...@grounded.net"
Sender: sipx-users-boun...@list.sipfoundry.org
Date: Tue, 6 Jul 2010 21:12:43
To: sipx-users
Reply-To: m...@grounded.net
Subject: Re: [sipx-users] Random Dropped Registrations
The problem is not s
The problem is not solved. The time is correct and all phones just dropped
again.
Happy to send a snapshot privately.
> In this instance you've said "The merge-logs command and the regtimes
> command don't [exist] on the ISO", but you haven't *copied down the details
> of the failure*. What is t
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of dan [...@dm3.us]
The logs are in UTC aren't they? Central Daylight time is 5 hours behind UTC.
___
Yes, the log
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
One thing I forgot to note is that someone was on the phone when all of the
registered devices went away but she neve
On Tue, Jul 6, 2010 at 10:21 AM, Sven Evensen wrote:
> Our product, which sits kind of on top of sipX, generates calls, redirects
> calls etc.
>
>
>
>
> So my question is : Why does RTP get buffered in sipX?
The answer is - it is not.
There is no buffer. If it is being delayed, that is for a
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
>Use the ntp.conf configuration created for you by the setup script -
>someone send him a copy
Copy of what? The ntp servers are in the ntp file.
> network - your firewall may need to be configured). A quick google
> search will tell you how to check whether or not it is working (and/or
> others
My bad. I'm on a bberry wand was doing it from memory.
-Original Message-
From: "m...@grounded.net"
Sender: sipx-users-boun...@list.sipfoundry.org
Date: Tue, 6 Jul 2010 20:12:10
To: sipx-users
Reply-To: m...@grounded.net
Subject: Re: [sipx-users] Random Dropped Registrations
On Tue, 6 J
On 2010-07-06 21:40, m...@grounded.net wrote:
> On Tue, 6 Jul 2010 23:47:48 +, Matthew Kitchin (Public) wrote:
>> Can you paste output from crontab -l and the file /etc/ntpd.conf
> Sure, nothing in crontab;
>
> # crontab -l
> no crontab for root
>
> And there is no ntpd.conf in /etc/
>
> I
On Tue, 6 Jul 2010 23:47:48 +, Matthew Kitchin (Public) wrote:
> Can you paste output from crontab -l and the file /etc/ntpd.conf
Sorry, it's ntp.conf;
# more ntp.conf
# ntpd configuration
# --
#
# Permit time synchronization with our time source, but do not
# permit the sourc
On Wed, 7 Jul 2010 00:49:57 +, Matthew Kitchin (Public) wrote:
> I would set the clock to the correct time and disable any and all time
> synch mechanisms for your server for now. That is just my opinion. I don't
> think any amount of clock drift could be worse than the problems you are
> havin
> At least there is no reason to think there is a problem with the
> sipxregistrar!
I'm keeping an eye on that since... um... as I was typing this, every
registration got knocked out again.
It's definitely something with the server, just as it happened, I checked the
time.
# hwclock --show
Tue
On Tue, 06 Jul 2010 19:42:46 -0500, Josh Patten wrote:
> Also, if you're in Central time then something is wrong where you are.
> All your emails are off by an hour.
Thanks for pointing that out. I used to have a central time server but have
started using ntp.org and we're out of sync on some mac
I do that on some systems, but leave the run level at 3 and only issue
startx when needed to walk someone through something, at which point logoff
unloads "x".
At least there is no reason to think there is a problem with the
sipxregistrar!
Remember that moving the hardware means the NIC will have
On Tue, 06 Jul 2010 19:41:56 -0500, Josh Patten wrote:
> GUI
> Are you running X on your sipX server?
No, the sipx GUI, the admin screens via web.
> If so, turn it off. It's generally a bad idea to run X and any telephony
> application on the same server.
I don't run anything else with
I would set the clock to the correct time and disable any and all time synch
mechanisms for your server for now. That is just my opinion. I don't think any
amount of clock drift could be worse than the problems you are having.
-Original Message-
From: "m...@grounded.net"
Sender: sipx-us
Also, if you're in Central time then something is wrong where you are.
All your emails are off by an hour.
On 07/06/2010 07:41 PM, Josh Patten wrote:
> GUI
>
> Are you running X on your sipX server?
>
> If so, turn it off. It's generally a bad idea to run X and any telephony
> application
GUI
Are you running X on your sipX server?
If so, turn it off. It's generally a bad idea to run X and any telephony
application on the same server.
On 07/06/2010 08:40 PM, m...@grounded.net wrote:
> On Tue, 6 Jul 2010 23:47:48 +, Matthew Kitchin (Public) wrote:
>
>> Can you pa
On Tue, 6 Jul 2010 23:47:48 +, Matthew Kitchin (Public) wrote:
> Can you paste output from crontab -l and the file /etc/ntpd.conf
Sure, nothing in crontab;
# crontab -l
no crontab for root
And there is no ntpd.conf in /etc/
I have a cron job sitting in /etc/cron.daily/ which I call timeset
Can you paste output from crontab -l and the file /etc/ntpd.conf
-Original Message-
From: Tony Graziano
Sender: sipx-users-boun...@list.sipfoundry.org
Date: Tue, 6 Jul 2010 19:20:27
To:
Cc: sipx-users
Subject: Re: [sipx-users] Random Dropped Registrations
___
> You should look at your clock source and your sync methods. I do not think
> there is a problem with your sipxecs system, other than the clock in this
> regard.
Yup, definitely something going on with time. Usually, when there are hardware
problems which are obvious, such as memory, CPU, these
I believe you have a clock problem of some sort. Which is why everything
expires suddenly, causing the registrations to "drop" then "re-register"
suddenly. The registrar logs pick it up too. Hence my original question.
You should look at your clock source and your sync methods. I do not think
ther
Nope, it's not a virtual machine else I would post that it was.
Linux uc.mydomain.com 2.6.18-194.8.1.el5 #1 SMP Thu Jul 1 19:07:06 EDT 2010
i686 athlon i386 GNU/Linux
On Tue, 06 Jul 2010 18:57:00 -0400, Matt White wrote:
"m...@grounded.net" 07/06/10 5:31 PM >>>
Physical server
>>> "m...@grounded.net" 07/06/10 5:31 PM >>>
>>>Physical server, IBM LS20 blade in a BladeCenter chassis. It would be dual
>>>duo-2.2Ghz with 4G of ram.
>>>We've had some heat issues in the server room so you might see some things
>>>related to that.
You say this is a physical server and its n
I'm in Central time zone so have
ZONE="America/Chicago"
UTC=true
ARC=false
... in my clock settings. I also have -360 (-6hrs) GMT offset in the GUI
settings. I have DST enabled as well with an offset of 60.
> The logs are in UTC aren't they? Central Daylight time is 5 hours behind
> UTC.
On Tue 06.Jul.10 17:59, m...@grounded.net wrote:
>Again, I have snapshots of the exact moments :). I would be happy to send it
>to one of the regulars on the list but prefer not to post it.
>
>On Tue, 06 Jul 2010 17:29:24 -0400, Matt White wrote:
>>?Can you provide details on this log? ?At what ti
By the way, it appears to be a 5Hrs forward difference.
On Tue, 6 Jul 2010 17:59:48 -0500, m...@grounded.net wrote:
> Again, I have snapshots of the exact moments :). I would be happy to send
> it to one of the regulars on the list but prefer not to post it.
>
> On Tue, 06 Jul 2010 17:29:24 -0400
Again, I have snapshots of the exact moments :). I would be happy to send it to
one of the regulars on the list but prefer not to post it.
On Tue, 06 Jul 2010 17:29:24 -0400, Matt White wrote:
> Can you provide details on this log? At what time did the phones
> "disconnect" and what is an extens
Physical server, IBM LS20 blade in a BladeCenter chassis. It would be dual
duo-2.2Ghz with 4G of ram.
We've had some heat issues in the server room so you might see some things
related to that.
On Tue, 6 Jul 2010 17:25:52 -0400, Tony Graziano wrote:
> "2010-07-
> 06T10:10:02.591488Z":1162049:K
Can you provide details on this log? At what time did the phones "disconnect"
and what is an extension/ip address we can look for.
The more details the better.
-M
>>> On 7/6/2010 at 06:23 PM, in message <20107617237.457...@leena>,
>>> "m...@grounded.net" wrote:
>No. The ISO install _is
"2010-07-06T10:10:02.591488Z":1162049:KERNEL:WARNING:uc.mydomain.com:OsTimer-12:B6E75B90:SipRegistrar:"OsTimerTask::insertTimer
timer to fire 3313262275 microseconds in the past, queue length = 0"
Is this on a virtual server?
On Tue, Jul 6, 2010 at 6:23 PM, m...@grounded.net wrote:
> >No. Th
>No. The ISO install _is_ an RPM install - changing from one to the
>other will accomplish nothing at all but consume time.
Figure I've got nothing to lose by installing another system to at least find
out if the condition goes away, especially if I can't find the solution on the
current server
Just posting the facts, as asked.
> Being registered has _nothing_at_all_ to do with making a call.
>
> ___
> sipx-users mailing list sipx-users@list.sipfoundry.org
> List Archive: http://list.sipfoundry.org/archive/sipx-users
> Unsubscribe: http://list
On 2010-07-06 16:47, m...@grounded.net wrote:
> The tools which were suggested to me are both where I'm stuck at this time.
> The merge-logs command and the regtimes command don't exit on the ISO so I've
> gotten sidetracked in resolving those issues. This has lead to other problems
> while th
On 2010-07-06 17:15, m...@grounded.net wrote:
> One thing I forgot to note is that someone was on the phone when all of the
> registered devices went away but she never lost her call.
> After she hung up, she used her cell phone to call her sipx phone which is
> internal and it went directly to
One thing I forgot to note is that someone was on the phone when all of the
registered devices went away but she never lost her call.
After she hung up, she used her cell phone to call her sipx phone which is
internal and it went directly to VM.
Without being in the list of registered phones, she
I think the bottom line is it can't hurt to see what the difference might be.
I've only ever used the ISO installs and like I said, nothing special about my
installs. I'm fully updated and the last update I saw was a couple days ago,
sipregistrar, haven't updated since.
I guess I'll know shortl
All of my installation are based on ISO. All. I've never seen these issues
you are seeing. Although that might help your situation, it seems it is
because it is replacing something that is broken with a new known working
component.
-Original Message-
From: sipx-users-boun...@list.sipfou
Just got a snapshot of all but one registration being dropped.
I was in info mode at the time, working on the other system but caught it out
the corner of my eye. I'll hang on to it but can't do much with it til the new
system is installed and perhaps have better luck with the tools. I am dying t
> 1) Set your logging level to INFO (or DEBUG if you like that better).
That's where they are at.
> 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.
We've done this.
> 3) Take a snaps
From: sipx-users-boun...@list.sipfoundry.org
[sipx-users-boun...@list.sipfoundry.org] On Behalf Of m...@grounded.net
[m...@grounded.net]
For a couple of weeks or more now, we have started noticing a behavior which is
that of phones dropping registration.
On Tue, 6 Jul 2010 14:43:15 -0400, Tony Graziano wrote:
> What I would do...
> Install webmin on it and let webmin handle the CPAN and fetch the
> prerequisites for you. Unless you feel like learning CPAN.
I've always used cpan but I also use webmin.
Either way, at this point, I've decided it's
I'll second the opinion on Voip.ms -- I initially got them just for
quick testing, but it quickly became obvious that they have their stuff
together. My only complaint is that they get involved in the media
stream. However, it's been extremely reliable. I use their Dallas POP
which consistently
A good alternative to VMware is KVM.
And if you download the Proxmox VE ISO, you get centralized web
management of KVM and OpenVZ virtualization containers.
-- Robert
On 7/3/2010 12:50 PM, Nathaniel Watkins wrote:
> I need to do some testing on a non-production sipx install. I was going to
>
On Tue, Jul 6, 2010 at 2:48 PM, m...@grounded.net wrote:
> I don't know of any other way to add perl modules?
you'd be running the most common platform and setup for production
machines. many developers use fedora, some F10, some F12 but not for
production and cpan may have an easier time instal
> ISO is meant to get you a running system. The idea is running yum to
> install things after that, that are not critical to the operation of
> the system is not difficult for users. Trying to keep the size <650MB
> is getting harder and harder to do as sipXecs grows.
Right, that makes sense. Th
On Tue, Jul 6, 2010 at 1:12 PM, m...@grounded.net wrote:
> I guess I don't understand why so many things seem to be missing in the ISO
> installs I have. Every time I try to run a program that most seem to think
> should be on the system, it's not there. Maybe I need to switch over to using
> a
I guess I don't understand why so many things seem to be missing in the ISO
installs I have. Every time I try to run a program that most seem to think
should be on the system, it's not there. Maybe I need to switch over to using a
source based system? Would that make a difference?
I very much w
Our product, which sits kind of on top of sipX, generates calls,
redirects calls etc.
This issue is for sipX 4.0.4 using sipXBridge and the fault is seen on
several
different ITSPs
In this scenario we have an external call redirected out to an other
external party.
We then want to add a p
In the diagram I sent you it explained how NAT needs to same port in both
directions in order to properly function.
A call trace and viewing it in sipviewer will tell you if the port is
identical on both legs of the call. If not, its an iptables issue.
Tony Graziano, M
Hi, Tony.
I really appreciate you help and your support in sipx.
I found the problem. The problem was that in ITSP trunk settings this
option was checked:
"User part of INVITE SIP URI is a phone number". As i see this is
checked by default and i didn't realized previously that this can be the
pr
I received an email from the Voice Operator Panel developer that there
has been a new version of voice operator panel that incorporates new
features that will be helpful in sipXecs installations:
quote from email:
"""
We just released an update to fix a bug with sipXecs and also to bring
a new
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
KVM on Fedora works great and is completely free (as in beer, and
speech). The KVM tools have also matured nicely. We use it quite a bit. :)
joe
On 07/03/2010 01:50 PM, Nathaniel Watkins wrote:
> I need to do some testing on a non-production sipx ins
79 matches
Mail list logo
