andrewgroup
-
Posts
615 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by andrewgroup
-
-
A new ITSP we are working with says that one of two IP's negotiate the inbound invite and we are to be accepting the call on another IP.. We have the trunk set to accept redirects.. Running 2.1.5..
Cisco is the culprit - ignore this post for now
-
ERE are not so simple... That is why we recommend to use the other "simple" forms - if possible.
To what extend can someone study http://www.opengroup.org/onlinepubs/009695...xbd_chap09.html and use what they learn here? I guess I mean does PBXnSIP truncate the support to a predined subset needed for dial plans, or is the sky is the limit on the use of BRE.
-
Maybe it would make sense to work on the shell scripts that pull out the neccessary information from the file system? Okay, would be only a solution for Linux...
What about the Win systems? let's not force cygwin onto windows just to get reports. What about then saving accounts to an FTP site? We can put an ftp site up and backup all clients to a central location. Easy is better...
-
More than once, I've resorted to copy / paste(special) text only the accounts pages into a spreadsheet to create some system documentation for reference purposes. It would very useful to be able to export the accounts information into a CSV file for reference purposes. XML files just don't compute, and wouldn't it be great to be able to reimport the accounts setting from the very same CSV file too. Yes I understand the full template process, but that doesn't account for first name last name, cell phones, etc...
-
Thank you for pointing me in the direction of best practices. Looking at the checklist I would have to say I am in compliance
So what do you attribute high CPU utilization to? Is it high with no traffic or does it only rise with traffic?
Ulaw vs Alaw, and are you familiar with the default admin codec settings? do they match the ITSP or Gateways?
Diagnosing this problem needs to be split 50/50
is this a CPU utilization issue or is this an ITSP QOS issue or are you using POTS or PRI gateways?
Are you using two NICS internal and Gateways or Internal and ITSP.
Anything you can do to split the problem gets you 50% closer to a solution. (do you have a test ITSP that you can build a test trunk and determine the results?)
Are you familiar with Wireshark? other tool?
The more you can eliminate with each test the closer you get.
That's all for now.
-
there is always the problem of secretaries leaving late and comnig early and then the PBX overwrites their manual state in a way that they did not envision.
I was reading a wiki entry and http://wiki.pbxnsip.com/index.php/Agent_Group. and it talks about having multiple service flags listed. Does the concept of having multiple service flags listed in the night service option for AA's apply. One of the service flags could be a scheduled flag and the other a manual flag. Is this right? Seems so simply if so.
-
Isn't this the reason the caller is in the mailbox in the first place?
-
We have been above 25% Media CPU on older versions without this issue occurring. Our server is Dual CPU 2GB RAM.
Are you running WIN? if so, I've posted in best practices some of the technical specifications and performance measurements of a reasonably busy mid sized 45 extension system. We've never seen 10% utilization on a single AMD 3100 series processor with a full PRI (23 channels in use). The "Don'ts" are many in regards to maximizing realtime performance for such things as PBX systems and video editing systems. Would it be possible one of the "don'ts" have been violated? (Software Raid control vs. real Disk controller) or something like that.
-
Opps,
Seems I need to read about the 8 digit. It's kind of special heh. Is it only special while in Auto attendants?
-
Two companies come to Mind
cbeyond in atlanta I think, they have a national presence and use multiple Class5 softswitches (CiscoBTS10200 rev5) extremely, extremely redundant. My local carrier has same switch on smaller scale but no national presence. (we and clients have been 100% Voip for 4 years) and I have real analog Fax lines on my provided IADs. not many VoIP carriers offer real fax lines.
Another is a Wholesale only provider NGT, that offers VoIP to larger interconnects. I can specify an agent that represents a larger Telecom provider.
something I just recently learned, is that many small ILEC phone companies have no agreements with the national ViOP providers recognizing their area telephone extensions and in more than one case. moving a client to a VoIP telephone number has resulted in calls across the street being long distance. You'll apreciate having a strong agent that know the tarriffs to get fixes for this stuff in hours vs. weeks, months or never. It's a real surprise to clients and you if your on the hook..
-
I was able to add a Jpeg attachment in one forum (best practices) other but I cannot add attachments in the other forums Autoattendants. Is this my ingnorance or a setting in IP Board?
-
I have 2.1.5 with the attached 800 Autoattendent that has dialby name on digit 9 plus 7 single button transfers with nothing assigned to digit 8
when calling the AA and trying to 3 digit dial account 800 thru 807 (same as single digit dialing) the message played says welcome to the mailbox zero.
-
No Gateways, Pure SIP trunks to our Provider. We can easily perform this again in a controlled tested and perform any number of tests. We did change the DTMF in-out of band settings, and start/stop service to no avail. Any tips to perform in a controlled retest away from the panic of the unexpected discovery would be appreciated.
-
We did an inplace upgrade on our internal system and after a day or so we discovered our Main AA wasn't recognizing inband DTMF signalling, thus the caller could not select a department. We were no aware of any changes other than the upgrade. Rolled back to previous all is good again.
Prior to rolling back, though we did some log troubleshooting and were not able to find a log setting that would indicate the DTMF tones were being detected. Does this exists, and I suppose if it did it would come from the same place that the PBX gets it and, thus DTMF either works or it doesn't.
Same config 2.03 works as expected and 2.1.5 worked (spotily)
-
Do you have a comparision number of CPU performance if you turn the real-time scanning on? I think it could be an important CPU performance difference.
An attachment is the last 24hours via WMI System Services a chart of the CPU processor on a 45 extension / with PRI 23channel SIP Gateway. This is an AMD 3100 single processor system,
Also find a 48 hour incoming /outgoing call chart for the e1/t1 for reference. These represent a couple of slow days for the business thats affected by rainy days.
-
Using our configuration of a Windows XP Pro System with 45 Extensions and a regular call count of 15+ calls peaking at 23 (Full PRI) we base the following tips.
A clean XP Pro or Server installation.
1GB RAM (ECC as a standard)
1 - Internal NIC for internal Phones
1 - Public Facing NIC for ITSP phones
1 - Nic for GATEWAY services (PRI or ANALOG)
Set all services to manual and start to determine the minimal running services.
we do use an Anti-Virus tool (but only with scheduled scans late in the day/morning.
We do not use real-time scanning or real time updates as these services can consume CPU and affect active calls.
We do apply the QOS recommended windows Patch.
We do install Wireshark directly on the server and validate COS/TOS bits.
We do use Wireshark to validate latency, SIP invites and SDP packets.
We do use capable switches support both Port and VLAN tagging 802.1PQ
We do use hardware based RAID Disk Controller with alerting.
We do use SNMP and pull stats.
We do use WMI performance monitoring.
the above system when fully loaded with 23 active calls, the CPU peaks at 8% utilization and we can run Wireshark and create a complete trace and not affect call quality.
your Mileage may vary.
-
Linux or Windows?
I'll post my comments in Misc Best Practices...Forum
-
Changing a periodic flag dynamically... hmm worth a try. Maybe it is enough to just call the service flag and change it's state. But I remember there was a lot of trouble when the user and the PBX both try to change the state. If possible it is better to make a clear decision.
So do we assume the last service flag to trigger is the active service flag?
Will a service flag triggering either scheduled or manually, will it reset all other service flags?
you said you remember "Lot's of Problems" how far back are you remembering and were these issues addressed in later versions.
We are in the process of trying to standardize our clients with a standard set of greetings and processes with a 1 page usage manual.
-
Progress yes solution no.
I did discover the CODEC entries in the global settings overide whatever settings you have in the domain trunk settings. Duh...
So I do now have the SDP offering Ulaw first...
I read the Cisco BTS10200 manual and found an update for 4.5.1 release allowing SIP trunks to independently support U and A law. Prior to that release only Ulaw appears to be supported.
Working with the ITSP, we are discovering perhaps some settings that are wrong on the CISCO.
The problem with Codec was our fault as Global settings for a Codec override Domain Settings. So I assume you should have no Codec Entries in Global. The ITSP is having a Cisco BTS10200 problem with REFER statements and Cisco has askcnowledge the problem. Hopefully a patch is forthcoming.
-
The ITSP is using a Cisco BTS10200 Class 5 softswitch. It appears as if they use Ulaw globally on SIP trunks and based upon the packet captures that I have done, altering the codec entries doesn't change the SIP/SDP invite packets. I've compared the packets before and after the codec preference changes and the packets appear the same. If you are a gluten for punishment here is the packet capture. I'll note the area in question.
-
Our experience is that using good managed switched Cisco/Linksys SRW224P with QOS enabled on all ports we've have great luck with phones with PC's attached to the phone switch port SNOM 320's for example. This setup has been 100% functional including when we had a Wireless 54G bridge with 20 computers and phones in a remote building 1/2 mile away. We always use some sort of SNMP management tool to monitor all ports on all switches and we closely monitor PBXnSIP via WMI/SNMP on windows boxes and the Switch port stats...
VLanning is simply not needed, however we are looking into a very large deployment and VLAN may be the only / best way to deploy and we'll advise as this progresses.
Out installation has experienced call counts in excess of 45,000 minutes per month, with average call lengths of 2 minutes meaning we were taking 22,500 calls in a given month M-F 8-5 pm.
Smart - Managed Switches with 802.1P/Q enabled can easily handle the small loads RTP adds to the LAN.
-
With lot's of ways to accomplish this, leaves a large margin for error and the likelyhood that those with little practical experience setting up a configuration as described, what's the likelyhood we or others will accidently choose the most affective way to accomplish this goal. The Best Practices forums would be a wonderful place to have an open discussion and real examples of how best, (based on real experience) to create inter-office trunks.
-
Give the following a try...
Set all services to manual and allow Windows to boot. It should start automatically those that are needed....
Here are the services running on a Windows XP Pro box with 50 extensions... We can shutdown some more of these if necessary...I'll comment on some
net start
These Windows services are started:
3DM (3ware Raid controller Management alerting service)
Automatic Updates (windows could disable, but no auto applies allowed)
AVG7 Alert Manager Server (we only have scheduled scans late in night, disabled realtime scanning for possible issue, may reenable)
AVG7 Update Service (More AVG antivirus, could disable if needed)
COM+ Event System
Cryptographic Services
DCOM Server Process Launcher
DNS Client
Event Log
Fast User Switching Compatibility (could disable)
Network Connections
Network Location Awareness (NLA) not sure
pbxnsip PBX ha-ha
Plug and Play (could disable)
Remote Procedure Call (RPC)
Shell Hardware Detection
Task Scheduler
Terminal Services
Windows Agent
Windows Agent Watchdog
Windows Management Instrumentation
The command completed successfully.
-
A new ITSP says we must use G711U..
They say we are trying 711A first
with Wireshark we see the following in the SDP initial packets... we have set the codec options to be 0 only and 0 8 as described as the defaults in the manual, still this descriptor shows as follows..
Media Description, name and address (m): audio 63720 RTP/AVP 8 0 18 2 3 101
the first RTP stream listed is Media Format: ITU-T G.711 PCMA
the second is Media Format: ITU-T G.711 PCMU
can this be true despite setting 0 in the codec field for the trunl?
The problem is we get the following reply
Status-Line: SIP/2.0 400 Bad Request
this comes immediately after after what appears to be a normally looking request.
inbound call being redirected resulting in error
in Trunk Setup
Posted
No more on this topic "case closed until Cisco Fixes Bug.