jlumby
-
Posts
290 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by jlumby
-
-
I was wondering if anyone has ever used a PRI gateway backwards. What I mean by this is instead of having a PRI from the phone company come into the gateway, and connect to the PBXnSIP as a trunk, is it possible to have the gateway register as an extension, and use a crossover cable to connect the gateway to the customer's existing phone system.
The reason for this is I have a customer that is really interested in utilizing the PBXnSIP for the conference bridge, as well as the monthly savings on their bill by being able to use an ITSP, however they are happy with their current phone system, and I would be hard pressed to talking them into replacing all of the phones that they currently have, as well as adding new cable since not all phones have cat 5 running anywhere near them.
If you have had some experience doing something similar to this, can you post the manufacturer/model number you have tested out.
-
In my experience I had done the TGZ WEB UPDATE PROCESS 2 times, and it did not upgrade the firmware. I had to manually get/load the file from PBXnSIP, and then it ran fine.
-
Some customers demand failover to pots if their internet connection goes down, as well as it is much safer to route 911 over a pots line. We can also save a significant amount of money by being able to purchase minutes for them in bulk through a single account we use for all of our hosting clients. While it does add some cost, most customers are willing to pay for the increased reliability. If the customer is more cost cost driven, and willing to waive the extra protection, we will gladly offer it without a 410 on site.
-
Now that I look at the packet capture more carefully, I agree that it is not responding. I have poured over it for almost a day now, and I cannot figure out why. The thing that really confuses me is I can dial extensions on the same switch, and they work without a problem, or if I register a second line to a different PBXnSIP switch, the problem does not exist. I have used these phones in 5 installs now, and this one customer is the only one that has the problem. I am at a complete loss.
As for firmware, I was only 1 version behind. I have since upgraded to 8.4.1 and it has not made a difference. 8.4.1 is about a month old. As for thinking the latest was 8.6, I think you might be thinking of the 7960. The 7960's version numbers are ahead of the 79x1 and newer models (phones that use the xml config file) The current version for the 7960, and 7940 phones is 8.9.0
-
I had this problem a month ago. It ended up being the FXO gateway firmware version. From a SSH prompt type
cat /etc/sipfxo-release
If the version is not
sipfxo 0.84, MSP m828_v2_04, VAPI Library Release 2.4.2, API Version 4.00
Then that is probably your problem. I was only 1 version behind, and I could not make it a half a day without it rebooting
-
Just for clairification, I would have 1 extension (on the main server) for each different outbound caller ID number I wanted to send. The problem I see with this is then I would need need as many trunks on the 410 which has a low limit on the number of trunks that it supports. Isn't there a variable string you could use in the ANI field of the extension on the main server so you would only need 1 trunk from the 410 to the main server's 1 extension, and it would pull the caller ID off of the SIP invite message from the 410? Seems much simpler, as well as it is how most ITSPs handle it.
-
I have one installation that is having a strange problem calling out, and it is intermitent. The problem only exists when calling across a trunk, internal calling works fine. When you dial the phone number, the phone just sits there, despite the dialplan telling it to dial, or pushing the dial button. In the background it authenticates with the PBX, however the PBX never sends the final sip packet containing the 200 OK. Then when I hang the phone up, and it cancels the call with the PBX, the pbx goes on to send the 200 OK and place the call through the itsp. This customer is running 3.0.0.2998 I have noticed that the pbx logs the error
Via is empty, cannot send reply
Despite the via field having content in it.
I am attaching the pcap, and logfile to this post.
The pcap was done from a remote phone, however the same problem exists within the office where there is no NAT
-
It would be running at all times. All of the phones in that specific office would register to it. It would then register it's trunk as an extension to my Large PBXnSIP server at my data center. If for some reason their internet connection went down, the settings on the 410's trunk to my server would be set to allow failover. The next thing in the dialplan of the 410 would be its PSTN gateway. The dialplan would also have an entry for 911 that would go directly to the 410's pstn gateway and not the trunk back to my datacenter.
-
I think the easiest way to solve that problem is to use the setting "Explicitly specify park orbit preference". If someone put a "*" in there it will mean "ask". This way we nicely stay backward compatible with that we have now and we can even specific the behavior on per-extension basis.
That sounds like a good way to do it. I will look forward to seeing it in a future code set. Do you have any ideas on if you can do something similar with retrieve?
-
I don't want to be too optimistic here. Presence has the problem that you might see the "presence" (whatever that is), but the PBX is not able to tell the phone what to do when the button is being pushed.
And there are a lot of x-cisco-xxx headers in the packets. Look for "extended-refer", if you want to have some fun. When the draft was written Rohan Mahy was still with Cisco, that is a long time ago. I would not build opon that...
I am not concerned about what to do when the button is pushed. That can easily be taken care of in the phone's config file. I was just wondering if the capture made it any easier to make it light up when the extension it is subscribing to the presence of is in use. (the capture was the packets that got sent to an idle phone from the call manager when the BLF was turned on, because the extension it was monitoring the presence of went into use)
-
With the latest firmware on my 7961 and a newly written config file, I was able to get my 7961 to subscribe to the presence of another extension. The following is what shows up in my extension's registration field:
presence 7208 sip:211@192.168.3.130:16785;transport=TCP Cisco-CP7961G-GE/8.4.0 170
I was wondering now that I have gotten this far, if it is possible to make the BLF work. I am attaching a packet capture of the same model phone registered to a Cisco Callmanager running SIP. I hope it is of some use to see if it is possible to light the BLFs through PBXnSIP
-
I have been running the numbers and thinking about offering hosted phone service to my customers utilizing the PBXnSIP product line. I was thinking that in some cases, where the customer would be interested in failover of having a CS410 at the remote office, and having one of it's trunks registered to my main high powered PBXnSIP server back at my office, and seting the 410 to fail over to it's built in PSTN gateway if the network connection to my server goes down, or 911 is dialed. It all seems quite easy until I start to think about Caller ID. Under normal operating circumstances, I would like to be able to send caller ID on a per extension basis. I can do this without a problem until the call hits the main server where the 410 would be registered. Hou would I use the caller ID that is coming in from the 410 across the trunk, instead of the value that would be entered in the ANI field of the extension on the main server that the 410 would be registered to. Are there variables that can be entered in the ANI field to pull the caller ID off of the incoming invite, instead of hardcoding a number in the field?
-
It seems to be working perfectly! THANKS!!!
-
I realize that it does not require an extension number behind the star code, however when you get to a larger install it is very problematic that it automatically assigns it to an orbit that is the same number as the extension number that is doing the parking. If there was an extended prompting radio button option next to the star code, and when it was turned on, the system would ask for the orbit number. This would allow the operator to type in any orbit number that they wanted. They could then page, and say you have a call parked on orbit XXX After that the receipient would press the retrieve button, and if the extended prompting radio button was turned on it would ask for the orbit (which they could type in), and then they would be connected.
While something similar to this could currently be accomplished if you had different buttons for different park zones, this is problematic when the average Cisco phone has 6 buttons. The first 3 are usually lost to the extension, intercom, and general message. Leaving park, retrieve, and one other. It also does not help much to assign zones in the PBX since the operator would need to be a member of all, and the person picking up the phone could be picking up from anywhere.
-
I have successfully done this with a manual/static registration. The process is documented at http://wiki.pbxnsip.com/index.php/Manual_Registration Be aware that it does not support trunks that require registration. If your gateway, or ITSP can allow the outbound call based on IP, then it will work fine.
-
Just a clairification, this would be usefull for any feature that would normally require more than just the * code. I would love to see it in parking/retrieving, pickup, and so on.
-
I think I am experiencing the same problem with some newer releases of PBXnSIP The following is what my mail server logs
Tue 2008-09-09 14:32:40: Accepting SMTP connection from [75.146.173.69:1303]
Tue 2008-09-09 14:32:40: --> 220-twincitytelephone.com ESMTP MDaemon 9.6.6; Tue, 09 Sep 2008 14:32:40 -0500
Tue 2008-09-09 14:32:40: --> 220-You are connecting to the Twin City Telephone Mail Server
Tue 2008-09-09 14:32:40: --> 220 Unauthorized use is prohibited!!!
Tue 2008-09-09 14:32:40: <-- EHLO localhost
Tue 2008-09-09 14:32:40: --> 250-twincitytelephone.com Hello localhost, pleased to meet you
Tue 2008-09-09 14:32:40: --> 250-ETRN
Tue 2008-09-09 14:32:40: --> 250-AUTH=LOGIN
Tue 2008-09-09 14:32:40: --> 250-AUTH LOGIN CRAM-MD5
Tue 2008-09-09 14:32:40: --> 250-8BITMIME
Tue 2008-09-09 14:32:40: --> 250-STARTTLS
Tue 2008-09-09 14:32:40: --> 250 SIZE 0
Tue 2008-09-09 14:32:41: <-- STARTTLS
Tue 2008-09-09 14:32:41: --> 220 Begin TLS negotiation
Tue 2008-09-09 14:32:41: SSL negotiation successful (TLS 1.0, 1024 bit key exchange, 128 bit RC4 encryption)
Tue 2008-09-09 14:32:41: <-- EHLO localhost
Tue 2008-09-09 14:32:41: --> 250-twincitytelephone.com Hello localhost, pleased to meet you
Tue 2008-09-09 14:32:41: --> 250-ETRN
Tue 2008-09-09 14:32:41: --> 250-AUTH=LOGIN
Tue 2008-09-09 14:32:41: --> 250-AUTH LOGIN CRAM-MD5
Tue 2008-09-09 14:32:41: --> 250-8BITMIME
Tue 2008-09-09 14:32:41: --> 250 SIZE 0
Tue 2008-09-09 14:32:41: <-- MAIL FROM: <PBXnSIP@twincitytelephone.com>
Tue 2008-09-09 14:32:41: Performing SPF lookup (twincitytelephone.com / 75.146.173.69)
Tue 2008-09-09 14:32:41: * twincitytelephone.com 75.146.173.69; matched to SPF cache
Tue 2008-09-09 14:32:41: * Result: softfail
Tue 2008-09-09 14:32:41: ---- End SPF results
Tue 2008-09-09 14:32:41: --> 250 <PBXnSIP@twincitytelephone.com>, Sender ok
Tue 2008-09-09 14:32:41: <-- RCPT TO: <jlumby@twincitytelephone.com>
Tue 2008-09-09 14:32:41: --> 530 Authentication required
Tue 2008-09-09 14:32:41: Connection closed
Tue 2008-09-09 14:32:41: SMTP session terminated (Bytes in/out: 440/1631)
I am sure the password is correct. If you are still interested in testing through someone else's email system, I would be happy to send the email account logon info to you via PM. Just send me a PM if you are interested
-
Well, I tested your domain name theorey, by fully implimenting DNS, and using the name instead of the IP in the phone's config file. It did the trick.
I did compare packets, and the only changes I saw were the public IPs changed to privates, and vice versa in all locations within the sip packets (even within the payload).
-
When connecting from the outside, 2 more items come into the mix that could possibly be modifying the packets. There is a Cisco 2621 router that is doing 1:1 nat to the softswitch, and a Cisco ASA5505 firewall at the remote office side. I doubt it is the ASA5505 since I have used it in other installs without a problem, however the 1:1 nat on the 2621 router is not one of my standards. The one thing that really throws me is if I dial an internal number (another extension on the softswitch) everything works fine, however if I dial a number that would be on a trunk, then the problem exists. I would not think the routers would know the difference. Does the softwsitch do anything different based on the desitnation that the routers might be picking up on?
-
I have a strange issue with one specific implementation of PBXnSIP 3.0.0.2992. When the cisco 7961 is in the office, it works perfectly. When the phone is out of the office it can dial extensions within the office without a problem, and can receive calls from anywhere, however when it goes to place a call across one of the sip trunks, the pbx loggs the following password mismatch error.
SIP/2.0 401 Authentication Required
Via: SIP/2.0/UDP 75.146.173.70:10120;branch=z9hG4bK45f54842
From: "Mark" <sip:129@75.146.163.187>;tag=001c5879d3e7000854dd80e8-d219ec45
To: <sip:4193922384@192.168.169.64;user=phone>;tag=ff92e2f250
Call-ID: 001c5879-d3e70008-6d6a8836-a414b057@192.168.1.2
CSeq: 102 INVITE
User-Agent: pbxnsip-PBX/3.0.0.2992
Warning: 399 75.146.163.187 Password does not match
Content-Length: 0
I am trying to figure out what is going on since the same config file is used both inside, as outside of the office, so the password is the same, plus the extension successfully registers, and can even make calls to other users on the same system. To make things even more complicated 7060 (slightly older model) phones work fine from both the inside, as well as the outside of the office. It appears to me something specific to this install since I can stay at the same offsite location, and register the phone to a different PBXnSIP machine, and I do not have a problem.
-
I am running 3.0.0.2998 on a CS410. The customer has 2 phones in an agent group. One of the 2 phones is almost always logged out. They claim that there have been 3 times now where the phone has logged itself in. When I look at the extension in the web interface, the phone status shows it is logged in, however the agent group sometimes will not show the time range that the phone was available for, and when it does, it will often show a start time when no one was in the building. Is there a better way to track this?
-
I have found that in 9 out of 10 cases the domain address book does not insert or overwrite the name when the call comes into a hunt/agent group, however I have found it works fine if the DID is pointed directly at an extension.
-
I have attached some packet captuers, as well as found some more info on the problem. It seems like when doing an attended transfer, it does not always fail. It seems lile there is a race condition, (or the pbx will only work on one of the 2 extensions) where it may work if you happen to pick up the call on one of the 2 extensions as opposed to the other. When it fails, you pick up, and you just get silence, and the other registered extension just continues to ring, and then roll to voicemail.
The latest thing I have found its it also happens in some cases where there is only a single extension registered, and you attempt to answer the call as it is forked to a cell phone. You answer on the cell, and get nothing but silence. The desk phone continues to ring, and then rolls to voicemail. I have attached a packet capture of this as well.
-
I have seen it happen with both the 7960, as well as the 7961, and 7941. I will get you some packet captures on Monday
Product Enhancement Request - Extended Prompting
in Product/Site Comments
Posted
I saw you posted a link in another part of the forum, so i downloaded, and tried 3.0.1.3017 to see if the * character worked to have the system prompt for the park orbit number. It does prompt now, however if I enter an orbit number that is anything other than my extension number, it says call parking failed, and the log file shows
[5] 2008/09/23 10:42:08: Could not find call of user 201 that should be parked
Is there a permission I need to set somewhere?