Re: [Asterisk-Users] Prices of g729 codec
On Mon, 5 Jun 2006, Mike Fedyk wrote: How hard is it to use a removable ethernet card for this type of usage? Also a USB ethernet if with Linux drivers should be usable for the 1U rackmount use case where all internal slots are in use. Depending on the type of server chassis, it may not be an option. Also, the g729 security code looks at all ethernet interfaces recognized by the kernel...so if you went the "only use removable ethernet card" route, you'd probably have to avoid using the same chipset as any of the built-in interfaces. ------ Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ --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: Fwd: [Asterisk-Users] Prices of g729 codec
On 6/3/06, Kevin P. Fleming <[EMAIL PROTECTED]> wrote: - Sahil Gupta <[EMAIL PROTECTED]> wrote: We recently had around 60-80 licenses become useless because Digium refused to renew the keys on that. That was a bit of money kissed goodbye. Unless you had been clearly abusing the key licensing system, our support department will never refuse to enable a new registration on your license key(s). There is no 'renew the keys', though, since they don't expire. I hope that's the actual official policy now. There seems to have been some internal conflict or communications failure at Digium a few months ago as to whether or how many times a g729 license key can be reset. As a service provider (you could call us an Asterisk ASP), we regularly build & host systems for customers, retire/upgrade systems, swap out hardware, add interfaces, etc. which causes problems with the g729 licensing. In one attempt a few months ago to get a license reset, I was initially told it was now policy that Digium would only reset the registration count once, and after that, you were SOL (or forced to play MAC address changing games or as someone else posted, try hacking around the license key code). In that particular case, the customer's server had suffered a 2 disk RAID failure, and to get them back online, I moved them to a lower end system (what was readily available) while we waited for parts to get their dual xeon server back online. Both motherboards had built-in dual ethernets. IMO, locking the licensing to a piece of system thats often built-in, has been very annoying. I think I'd be happier if it was locked to some sort of dongle (parallel, or more likely today, USB). At least that way, we could easily move the key anytime we needed to. It would be a bit of a pain any time a system needed to quickly be transfered to hardware already at another location. The TRX idea sounds appealing, but I wonder how they'll handle servers that don't have internet access. Not all VOIP servers are on the internet. I've actually wondered if we could legally use Intel's code in cases where we have licenses bought from Digium, but they're not re-registerable because Digium wouldn't reset the use count. -- Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ --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] odd transfer behavior
Using Asterisk 1.0.x, I'm seeing the following when trying to allow the called party to transfer calls: sip phone --> asteriskC --IAX2--> asteriskA --ZAP--> PSTN --> called party i.e. a sip user on server asteriskC places a call, which is sent to the PSTN via IAX2 to asteriskA. asteriskC includes the t flag in its Dial(IAX2/...). This case works, though I found that the called party can only transfer calls to extensions asteriskC makes accessible to asteriskA via context= in the iax.conf entry...which I suppose makes sense. However, if the call flow is: asteriskB --IAX2--> asteriskC --IAX2--> asteriskA --ZAP--> PSTN -> called party i.e. a call comes into asteriskC via IAX2, and is either: a) answered by a SIP client who transfers the call to an exten that dials out (with t flag) via IAX2 to asteriskA and the PSTN b) is Answered in the dialplan and then sent via Goto to an exten that dials out (with t flag) to the PSTN via IAX2 to asteriskA. In these cases, the called party cannot do a # transfer. I found asteriskC could only recoginze a #transfer attempt by the called party if I forced a mismatch in codecs between the incoming and outgoing IAX. Even with that, though, I can't get #transfer to work as the called party seems to have no context. i.e. -- Operating with different codecs, can't native bridge... -- Playing 'pbx-transfer' (language 'en') -- Unable to find extension '4' in context '' As soon as I dial any digit after #, asteriskC errors saying the extension is not found in a null context. Is what I'm trying to setup possible on a server with no direct PSTN access? All PSTN origination and termination is done via IAX2. I've just tested this with the recently released version 1.2 of Asterisk on asteriskC, and the behavior is largely unchanged. 1.2 seems to do a better job of noticing the # even while it says it's natively bridged the calls, but I still get an Unable to find extension '' in context '' suggesting that the called party trying to do a #transfer has no context. -- Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ --Bandwidth and Colocation sponsored by Easynews.com -- Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] Ring requested on channel already in use?
We've had a recurring issue on one of our servers where one or more zap channels from a PRI will get "stuck" causing incoming calls to fail. I happened to have a CLI session when it happened this morning. The start of the problem looked like: -- Channel 0/3, span 1 got hangup -- Channel 0/3, span 1 got hangup Jul 18 08:34:54 WARNING[20343]: chan_zap.c:7585 pri_dchannel: Ring requested on channel 0/3 already in use on span 1. Hanging up owner. -- B-channel 0/3 restarted on span 1 Jul 18 08:36:31 WARNING[20343]: chan_zap.c:7585 pri_dchannel: Ring requested on channel 0/3 already in use on span 1. Hanging up owner. Jul 18 08:36:31 WARNING[20343]: chan_zap.c:7585 pri_dchannel: Ring requested on channel 0/3 already in use on span 1. Hanging up owner. These messages (Ring requested...) then repeated until we killed and restarted asterisk. I don't normally see a channel get hungup twice in rapid succession. Is it possible that asterisk tried to deal with hanging up the channel from a single call twice and left it in some stuck state? Asterisk CVS-v1-0-07/05/05-10:54:39 zaptel and libpri were checked out/built/and reloaded when we upgraded to this cvs snapshot of asterisk. This system has 2 T100P cards, though only one is in use. ------ Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] Phantom problem authenticating IAX2 with RSA
I'm getting exactly the same behavior as was posted about in http://lists.digium.com/pipermail/asterisk-users/2004-March/040380.html I've upgraded (both ends) to CVS stable (CVS-v1-0-06/17/05-13:15:49). Jun 17 13:46:17 NOTICE[15942]: chan_iax2.c:4053 authenticate: No way to send secret to peer 'a.b.c.d' (their methods: 4) Immediately after that, I'll see frames go by with Tx-Frame Retry[000] Subclass: NEW Rx-Frame Retry[ No] Subclass: AUTHREQ Tx-Frame Retry[000] Subclass: AUTHREP Rx-Frame Retry[ No] Subclass: ACCEPT that make it look very much like rsa authentication is being done, and the call is accepted. I noticed this while cleaning up my IAX config...moving away from type=friend entries to a type=user and a type=peer entry for each system I send/receive calls to/from. i.e. on the remote end, I have: [my.system.name] username=my.system.name type=user auth=rsa inkeys=my.system.name context=my.system.name-iax qualify=no disallow=all allow=g729 allow=gsm deny=0.0.0.0/0.0.0.0 permit=[IP of my.system.name] On the end I'm calling from: [remote.system.name] type=peer username=my.system.name auth=rsa outkey=my.system.name qualify=no disallow=all allow=g729 allow=gsm host=remote.system.name The test call is dialed as IAX2/remote.system.name/${EXTEN} Is there a problem with my config, or is this just an iax2 cosmetic bug? Each end does have appropriate rsa keys (readable by asterisk) in /var/lib/asterisk/keys. BTW, if I'm reading the docs correctly, there are multiple errors in the wiki: http://www.voip-info.org/tiki-index.php?page=Asterisk%20IAX%20authentication#comments where "allow" is incorrectly used [in the context of allowing an IP] where "permit" was meant. ------ Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
RE: [Asterisk-Users] Want to use Asterisk instead of existing MeridianNorstar system ... need some help
On Mon, 18 Apr 2005, Joe Dennick wrote: > You can use the same six lines for both inbound and outbound calling > just like you do now. The 'roll-over' will start on line 1 and move up. > You'll have to configure your outbound calls to start on line 6 and move > down. If you ever get to the point where all six are consumed, you'll > want to expand. What's the reason for configuring the outbound zap group in reverse order of the incoming rollover? > You probably can NOT use the existing Meridian phones because they are > digital phone sets, not standard analog ones. You can purchase 5 > TDM400P cards (assuming you have 5 available PCI slots in your Asterisk > Server), and configure two with FXO ports (making 8 total) for the POTS > lines and the other three with FXS ports (making 12 total) to connect to > regular analog phone lines. I'd be a little worried about the 5000 interrupts/s those boards would generate. Are people actually running systems with that many wcfxs cards? ------ Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users]Unable to register license for G729 codec
On Wed, 13 Apr 2005, Mohammed Firdosh Nasim wrote: > Hi, > > I bought the license for codec g.729a from digium and am now facing some > problem registering the codec with them. > i got the following message. > > Connecting to Digium License Server (216.207.245.3:5646)...FAILED(2)! Perhaps you have a firewall/packet filter that's stopping you from connected to Digium's key server? It's working from here. $ telnet 216.207.245.3 5646 Trying 216.207.245.3... Connected to 216.207.245.3. Escape character is '^]'. 220 Welcome to cpsignd ---------- Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] multiple enum results
On Fri, 11 Mar 2005, Duane wrote: > > Before I hack this into enumlookup.agi or write a new one, I'm just > > curious, have others done this, or are there other better ways to do what > > I'm looking to do? > > There was talk on the dev list on fixing this, not sure how far things > went. I got tired of not having proper enum routing in asterisk I hacked > up a php script ages ago to handle it... > http://www.e164.org/enum.phps I liked most of the way enumlookup.agi worked, just not how equal cost records were handled, and was somewhat sickened by the number of invocations of sed & awk (looks like about 64 seds)...so I basically rewrote it in perl, ripping out some functionality I don't need, fixing the handling of equal cost records while only sorting them based on order/priority, and not iax hostname, and only using one external program (dig). Now that I've got this working the way I wanted, I'm wondering about a suggestion a friend made. Rather than equal cost NAPTRs, would it make sense to replace those with single NAPTRs using iax names that point to hostnames with multiple A records (for the various servers)? How are things like IAX2 trunking handled when an IAX peer entry has a host=NAME line where NAME has multiple A records for different hosts? ---------- Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] multiple enum results
I'm setting up a private enum zone to simplify/centralize dialplans for a number of Asterisk servers. In several dialed number situations, there are a handful of possible destinations for the call, and I'd like to have * try ENUMENTRY1, ENUMENTRY2, .., ENUMENTRYN just in case the first result is temporarily unable to handle the call. In at least some cases, I'd also like the order in which the servers in the enum results are tried to be random. The wiki pointed me at enumlookup.agi for enum support of multiple NAPTR results per lookup. By default, enumlookup.agi will either return equal order/priority NAPTRs as a single &'d result, or individual ones in order according to NAPTR defined order and priority. What I'd prefer is for equal order/priority results to be returned separately in the order the DNS server gave them, while still sorting by order/priority if they differ among results. Before I hack this into enumlookup.agi or write a new one, I'm just curious, have others done this, or are there other better ways to do what I'm looking to do? ------ Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
Re: [Asterisk-Users] Agent login state saving?
On Fri, 31 Dec 2004 [EMAIL PROTECTED] wrote: > >From configs/queues.conf.sample: > > [general] > ... > ; Persistent Members > ;Store each dynamic agent in each queue in the astdb so that > ;when asterisk is restarted, each agent will be automatically > ;readded into their recorded queues. Default is 'yes'. Looks like this is only in cvs-head. Are you using that in production? AFAIK, there have been some serious changes to the ways queues work in cvs-head. ---------- Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
[Asterisk-Users] Agent login state saving?
Has there been any consideration of having asterisk save to a file the state of which agents are logged in such that after a restart (or crash) all agents don't have to manually re-login (after eventually realizing they're no longer logged in and not receiving calls :) ? ------ Jon Lewis | I route Senior Network Engineer | therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_ ___ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users