NEED HELP ASAP- IN HUGE TROUBLE

[Winning]

Hi Everyone below is my server info

Im currently using 2 servers. 1 as an all in 1 server and the other as a telephony server. My DB server is a HP ProLiant Server and has 32 GB of Ram . My telephony server is the same server expect only 24 GB of ram.

I have currently had vicibox 4.0.3 installed professionally by a member of this forum and the installation has worked smoothly. I am just stuck with a problem. I keep have horrible quality issues when I have 40 agents and more. I have the agents split equally on both servers and all calls over 7 seconds are being recorded for only outbound dialling. There is no inbound dialling at all. The current bandwidth we have is 20 Mbit both ways. We are using both x lite, eye beam and zoiper on hp and dell computers. The carrier settings are below. I really just want to understand what my next steps are to determine how to resolve the choppy lines. We also have a load of under 35% on both servers when we reach 50 agents so I can safely rule out the servers being a problem. Can someone please provide a detailed explanation as to what steps I take and how I can resolve this ASAP? I also use ulaw and I am open to using GSM. Please get back to me asap as all the help is necessary ! Thank you

[103691]
disallow=all
allow=ulaw
auth=md5
type=friend
username=******
secret=*******
host=

[williamconley]

1) Welcome to the Party!

2) As you are obviously new here, I have some suggestions to help us all help you:

When you post, please post your entire configuration including (but not limited to) your installation method and vicidial version with build.

This IS a requirement for posting along with reading the stickies (at the top of each forum) and the manager's manual (available on EFLO.net, both free and paid versions)

You should also post: Asterisk version, telephony hardware (model number is helpful here), cluster information if you have one, and whether any other software is installed in the box. If your installation method is "from scratch" you must post your operating system and should also post the .iso version from which you installed your original operating system. If your installation is "Hosted" list the site name of the host.

If this is a "Cloud" or "Virtual" server, please note the technology involved along with the version of that techology (ie: VMware Server Version 2.0.2). If it is not, merely stating the Motherboard model # and CPU would be helpful.

Similar to This:

Vicibox X.X from .iso | Vicidial X.X.X-XXX Build XXXXXX-XXXX | Asterisk X.X.X | Single Server | No Digium/Sangoma Hardware | No Extra Software After Installation | Intel DG35EC | Core2Quad Q6600

(I note that you did mention vicibox 4.0.3 ... which is merely your Installer Version ... this is not your vicidial version, please remember to always post that at the very least)

3) You cannot rule out the servers based on "a load of under 35%". Where did you see "a load of under 35%"? The best way to check server load is at the command line of each server DURING an episode of poor sound quality. Use htop to see: 1,5,10 minute average server load AND immediate cpu usage on each core. If any individual CPUs are pegging out (100%) for more than a second or so, you could be experiencing overload even if your "average" is ok.

4) Try turning off recording during your next episode to see if this reduces the problem (could be an indication of low RAM availability or disk problems moving those recordings instead of deleting them after the call).

5) Seriously scour the htop for running processes that may be present (and using a lot of CPU) when an episode occurs. If you notice a process that is always present during an episode, but rarely present otherwise ...

6) Do you have any customizations? (Another thing you didn't mention in your setup ...)

7) Have you tried a 2nd carrier (after all, if it is indeed NOT your server ...)?

8 ) Have you cataloged the complaints to see if there is any correlation to any hardware/workstation/physical location in your facility? I've actually found an agent with a FOOT on a network wire (sometimes) to be a culprit with a poor connection issue. No way to find that without some tracing. LOL

9) Happy Hunting!

[williamconley]

Oh, and it is worthy of note that "20M in both directions" does not allow for unlimited calling. The number of agents is NOT the limiting factor here. Note that your agents are (likely) on the same subnet as your servers. So the "internet bandwidth" is unrelated to them as they are not using it for vicidial communications.

What DOES matter is communications going out on that 20M of bandwidth.

So: How many CALLS are active and ringing simultaneously when this happens? If you find that this happens at roughly the same level of calls each time ... you may well be hitting your ceiling.

Also, have you TESTED this 20M x 20M service? To be sure it is REALLY 20M in both directions?

Next, I've found (on numerous occasions) that internet bandwidth is in use in areas it should NOT be and that this often manifests itself under heavy load (when Vicidial uses up the remaining bandwidth!). Your system SHOULD be set up so that Vicidial servers have DIRECT (uninterruptible) internet access. No router. It should also NOT be required to share bandwidth with your agents. If this is a requirement, then the bandwidth in use by the agents should able to be TURNED OFF in an emergency. Like a switch. IE: The vicidial servers go to the internet directly (each with its own IP, perhaps through a switch but NOT through a router). The Agents, on the other hand, should go through one (or many) router(s). This allows you to pull the plug (just the internet plug) on those Agent routers in times of poor quality calls. If the call quality clears up in just a few seconds (or instantly), you have a bandwidth waster in your midst.

No twitts, utubers, netflixians, or iheartradioians allowed.

[Winning]

Thank you William,

I will troubleshoot with your advice and repost.

[williamconley]

I will say that in many cases when I've been called with this particular situation and the client is willing to pay Emergency pricing to get this fixed ... I've ALWAYS suggested that before hiring us and paying to have this resolved the first step should always be ... Turn Off Internet for anything other than Vicidial during your next episode. If internet is a requirement of some sort, you can still do without it for FIVE MINUTES to test the theory. If that resolves your sound issue for those five minutes, you now have a place to start from. If it does not, you can begin to look at internal networking and/or overall bandwidth and/or carrier issues (carriers are not perfect by any stretch of the imagination). If possible, changing carriers is my first try for a fix in many cases (having more than one carrier is a requirement for a professional call center).

Happy Hunting!