Re: [asterisk-users] Cisco remote reboot

2007-05-28 Thread Daryl Jurbala

I've fired a script from an AGI-BIN to accomplish that.

Try this one:

#!/usr/bin/perl
# mk 2004  feel free to distribute
#  [EMAIL PROTECTED], _Vile
#  perl script to reboot phones
#  try telnetting to your phone, first.
#
use Net::Telnet ();

$phone_ip = shift;

# Your Cisco 79xx prompt
$prompt = "Enter Your Prompt Here";

# Your Password
$password = "xx";

# Reset Command
$command = "reset";

if ($phone_ip eq "all")
{
reboot("xxx.xx.x.xx",$password,$command,$prompt);
reboot("xxx.xx.x.xx",$password,$command,$prompt);
reboot("xxx.xx.x.xx",$password,$command,$prompt);
reboot("xxx.xx.x.xx",$password,$command,$prompt);
} elsif ($phone_ip eq "") {
print "Enter an IP or 'all' for All.";
} else {
reboot($phone_ip,$password,$command,$prompt);
}

exit;

sub reboot{

my ($ip,$password,$command,$prompt) = @_;

$t = new Net::Telnet;
$t->open($ip);

$t->waitfor('/Password :.*$/');
$t->print($password);

$t->waitfor('/'.$prompt.'>.*$/');
$t->print($command);

}


___
--Bandwidth and Colocation provided by Easynews.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] Asterisk High-Capacity Stability

2007-05-15 Thread Daryl Jurbala


On May 14, 2007, at 11:27 PM, Atlanticnynex wrote:

I'm curious what kind of configuration/features/modules you could  
recommend for my setup. Can you explain further what you mean by  
OpenSER to Asterisk?


If you want to go Open Source, I think OpenSER is a good choice.  You  
won't need to do any hacking to make it work..I'd suggest making  
1 or 2 openser boxes to act as registrars for your user agents, and  
use the openser dispatcher module to point at one or more openser  
boxes that do LCR for calls that go directly out, and at one or more  
asterisk boxes for feature servers if you need them.


Using Asterisk realtime and the database extensions for OpenSER you  
can share the user database between them and things should just  
work.  Write your CDRs to a separate database (as to separate  
business data and call flow datajust in case someone does a  
complex CDR query you don't want your PDD to go through the roof) and  
come up with some kind of CDR remediataion for billing.


___
--Bandwidth and Colocation provided by Easynews.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] Asterisk High-Capacity Stability

2007-05-14 Thread Daryl Jurbala


On May 14, 2007, at 1:29 PM, Zoa wrote:



Several people do use it for handling > 50k minutes a day. (I'm one  
of them).
Yes, you need to know what you are doing, and have a nice design,  
but it is possible.Our code is only slightly altered. (mainly for  
billing purposes).


That's great if you're good enough/have the time to make that  
happen.  But when I have issues and call/pay Digium and don't get  
timely or meaningful answers, it's doesn't make for a good business  
decision to continue using it for that purpose when I can toss in a  
Nextone or Sansay and have it "just work".  All the time.  No  
babysitting.  Full professional and timely problem resolution from  
the vendor, etc, etc, etc.  Don't even get me started on Digium not  
being able to get TC400Bs to properly negotiate g.723.1 5.3k when a  
client requests 6.3k first (thank god for Cantata).


I guess it all comes down to whether you want things to just work and  
be able to have tier 1/2 support capable of actually doing anything  
meaningful, or if you want to have the engineering level people  
forced to do all the work.  From my standpoint, the smart business  
decision is quite clear.


But, as I said, Asterisk is still driving the feature servers, and  
works well for it.  As mentioned by someone else previously in the  
thread, it makes a great endpoint.


If you're having good success with it, that's fantastic.  I would  
hope that you contribute back to the list how you set things up to  
make this a possibility.


___
--Bandwidth and Colocation provided by Easynews.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] Asterisk High-Capacity Stability

2007-05-14 Thread Daryl Jurbala

On May 12, 2007, at 4:11 PM, Atlanticnynex wrote:


Thanks Alex, some great ideas.
I think, however, I'm leaning towards Asterisk at this point- since  
I have quite a bit of experience there, and very little with SER.  
At this point, I'm wondering from a dimensioning standpoint, what  
kind of capacity my machine will have (Dual Core Xeon 2.4GHz 4GB  
RAM). As I said, I don't plan to do any transcoding. I read the  
voip-info page on dimensioning and it seems theres some mixed  
feelings about Asterisk in high-capacity environments. I guess I'm  
looking for input as to whether Asterisk could handle roughly one  
DS3's worth of calls (672 calls) just doing the LCR (I've seen some  
pre-built LCR apps, looks like they all do on-the-fly MySQL  
queries- I think I'd write my own AGI that would use a cache).



With my hardware, could Asterisk run stable for this amount of  
traffic?

What stability issues does Asterisk have at this scale?



Simply put, NO.  I am on a project now where a client had an OpenSER  
box acting as an SBC and registrar passing traffic to several  
asterisk boxes which are doing LCR lookups on the fly as well as  
writing custom CDRs all through PHP AGI scripts to a Postgres DB.   
The Asterisk boxes do not scale, and randomly start swallowing calls  
or, more often, restart the process (safe_asterisk is handling  
this).  There is some light IVR type usage for reporting account  
balances and the like.  With anything more than 80 or 90 calls on the  
box, the IVR prompts start to break up.  Ben through replacing  
hardware, more memory, different Asterisk builds, etc.


I've had an open issue with Digium support on this for at least a  
couple of weeks, and the best advice so far was "try using the SVN  
build".  That makes things better, but it's still not anywhere close  
to fixed..


It's absolutely incredible that Asterisk works at all for some of the  
situations its been put in - major kudos to the developers.  But I  
don't think using it for what you're talking about is a long-term  
business strategy.  When the highlight of the 1.6 release is bridging  
channels, you know high volume sip to sip usage in a carrier class  
call routing environment is NOT what development is focused on.  And  
that's fine.  If you use a wrench to do the job of a screwdriver, you  
shouldn't complain when you bust your knuckles


That being said, I don't meant to trash Asterisk at all.  It's a  
fantastic feature server, and a great PBX, both of which things I use  
it for very successfully.  I just don't think it's ready to handle  
50k plus minutes a day SIP to SIP with LCR and billing data, no  
matter what you do with it.  I'm 100% positive there are people out  
there doing it successfully, but those are the exception, not the  
rule.  And I doubt they are running unmodified code.


___
--Bandwidth and Colocation provided by Easynews.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] Packet2Packet Bridging Questions

2007-03-08 Thread Daryl Jurbala
OK...that makes much more sense.  So here's my follow-up question:  
what's the easiest way to check if I'm native bridging a call.  I'm  
trying to offload as much RTP traffic as possible, and want to have a  
way to check quickly (there are well over 50 calls on each of these  
boxes at any given time).  I've been going the ethereal route, which  
is great for debugging, but not so good for a quick look.


Thanks again,
Daryl

___
--Bandwidth and Colocation provided by Easynews.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
  http://lists.digium.com/mailman/listinfo/asterisk-users


[asterisk-users] Packet2Packet Bridging Questions

2007-03-08 Thread Daryl Jurbala
I'm just starting to upgrade some boxes from 1.2.x to 1.4.1 as well  
as trying to get some of the RTP traffic offloaded from the network.   
I think I'm misunderstanding what the console messages mean when it  
says "Packet2Packet Bridding SIP/blah to SIP/blah".  I though that  
meant that it had successfully (re)INVITED and the media was no  
longer going through my Asterisk box, but ethereal says different.


I'm not having much luck finding any information on this on the wiki  
or google.  Can someone point me in the right direction?


Thanks,
Daryl



___
--Bandwidth and Colocation provided by Easynews.com --

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users