In Consideration SNMP Monitoring - IN CONSIDERATION

[[email protected]]

It would be nice if you implement SNMP Monitoring. This will allow to monitor things like:
-Active Calls:
-Between Extensions
-Between Bridges
-PSTN Calls
-Incoming Calls
-Outgoing Calls
-Online Extensions
-Offline Extensions
-CPU usage per 3CX Process
-Memory usage per 3CX Process
-Logging attempts.
-3CX Service Status

And so on.

Thanks in advance.

 

[Nick Shortal]

+1 please for SNMP monitoring feature.

 

[Brad Cann]

+1 from me too

 

[Andrzej Pierscionek]

This is an absolute must have for hosted. We already have great monitoring system and many others are avalible.
Currently there is no way to extract monitoring information from 3CX (well unless we would attach with .net API or parse logs). Most universal would be SNMP as this can be used by most monitoring systems.

 

[Paul Omans]

I agree with integrating SNMP, mostly for the simultaneous call graphing possibility it introduces.

 

[Gavin Hill]

I know this is an old thread, but I'm new to 3CX and surprised there is no SNMP-based data available. We use Nagios (check_mk) extensively and the ability to see call, trunk and phone registration data would be a big help.

Given Yealink removed SNMP, what would be a good way to see when phones register/deregister? Even a syslog capability so that I can parse the logs would be a useful addition.

 

[B.Fluehmann]

HTTP / JSON!

We are using Nagios to monitor our environment and although SNMP is the de facto standard, I try to get around it whenever i have other options.

Please take a look at Ubiquiti UniFi. Our first approach for monitoring was SNMP as well, which ended up with frustration. Forums full of questions without answeres, missing MIBs, missing parameters etc.
Then I discovered that the REST API of the controller provides everything we could imagine.
- Status info / Parameters of the entire system
- Events (Approx. last 5000)
- Alerts (Approx. last 5000)

Based on this we have written our own Python Nagios plugin which provides us with very detailled infos in an easy way.

From my point of view, theese are the advantages of JSON over SNMP:
- Human readeable
- Easy to implement
- Easy to access (Web)
- Easy to process

 

[Gavin Hill]

I'd be happy with any suitable interface to monitor 3CX metrics via Nagios.

As an aside - is your UniFi script open-source? I'm running 5 Unifi products and I agree their SNMP implementation is sorely lacking.

 

[B.Fluehmann]

I'd be happy with any suitable interface to monitor 3CX metrics via Nagios.

As an aside - is your UniFi script open-source? I'm running 5 Unifi products and I agree their SNMP implementation is sorely lacking.

The plan is to make it open source. I am in the process to clean it up and add propper comments. As soon as this is done it will be published.

 

[Helmut72]

+1 for SNMP, this is IMO vital e.g. to report KPIs to the management.

 

[Neturius]

+1

 

[TreuOrga - JM]

+1

 

[Mohammadjuve]

+1

 

[Silly English Kniggit]

I don't want, and won't use SNMP trapping. We'd want SNMP polling, but SNMP ceases to be much use for this with 3CX going to the cloud, as no one should be running SNMP over the Internet.

I want to be able to poll the 3CX server (Linux or Windows) via some process (probably a command on the server itself) and read back one line with one sensibly formatted value for each query (command). That value might be an pure integer "5" (number of sim calls), or a string "OK: system operating normally". That we can read into management system and use for alerting etc - that's relatively monitoring system independent, and won't rely on trying to use insecure protocols that should be blocked in a lot of situations anyway.

I was really disappointed to see that in the Linux version, it doesn't log at all to the system log - so even the small amount of event monitoring we have automated for 3CX Windows isn't possible for Linux. It would have been relatively easy to implement this given the calls are already in the Windows code. With syslog forwarding, this could effectively replace SNMP trapping. If you want that added, please upvote https://www.3cx.com/community/threads/linux-version-to-log-to-system-log-like-windows-does.50954/

 

[flemieux]

+1, have to develop an interface to monitor this 3CX: command line, snmp, etc... We need it

 

[qmex-mp]

+1

 

[Trent Hamilton]

+1

 

[Emerson Pereira]

+1

 

[phintec]

+1 we would love SNMP, that would be simple to plug into PRTG and get great stats and alerts!

 

[yanceyworks]

+10 for us, but simple = better.

First we need a simple way to monitor multiple instances of 3cx (hosted Linux in particular) on one dashboard
Next we need to see on that same dashboard alerts of phones/trunks not registered

We need to be able to see the issue quickly and respond appropriately.

Solutions:
1. Internal 3cx tool, fine, but lots of dev time I suspect
2. SNMP, if well implimented will take dev time too, but lots of outside tools can do the fancy stuff
3. Others, rest API and more? I'm open, but don't want to rebuild things or spend a fortune on "more" monitoring tools.

Key is that if we want to sell more we have to deliver great service and for that we have to monitor to help mitigate issues as quick as possibly.

 

[Trent Hamilton]

I was working on a fabricobbled script that would automatically log in to the 3CX admin panel and parse the information displayed in such a way that the various parameters could be picked up by my monitoring platform, as I'm also in the same situation where we want an at-a-glance panel for multiple installations... I didn't end up getting there because I don't have spare time to invest on this, and it really is something that 3CX should provide.