Hi David,
Long story, sorry..

I’m always installed FS on dedicated hardware (single node, compiled from source) and this issue really never happened in the past.
“show calls” command actually matched PostgreSQL "channels" table record count.

Now I changed the architecture this way.
1. THIRD PARTY SIP PROXY /  on same server, central PostgreSQL database with FreeSWITCH Core db and CDRs database. Proxy is balancing traffic over
2. Number TWO freeswitches (mirrored configuration, scripts and gateways) with a custom Lua script that query central database to get customer data and routing, based on that will bridge calls
3. Delayed write CDRs with Master.csv import files


My issue is that the calls count grow too much, and I start to get on console following error “NO RTP audio available”, so only way to restore services is to restart FS on nodes.

I’ve even created a bash script to kill “DOWN” calls with uuid_kill looping on channels tables, but this doesn’t save me on RTP error.

Maybe you have some clue on how to solve this issue. I know my FS installation is old, but migration to newest version is not so seamless
Thank you so much
Max




Il giorno 24 ott 2019, alle ore 00:53, David Villasmil <david.villasmil.work@gmail.com> ha scritto:

Any idea at all where those calls are coming from? Are there really calls going on?
In my experience (from the beginning of the project) show calls (SQLite) is not perfect and will show calls that are not really there, but doing a quick investigation will show it’s just a bogus call. You can’t even kill them with uuid_kill.

This doesn’t happen with MySQL/Postgres in the core, for some reason.

Is your setup by any chance on a VM?


On Wed, 23 Oct 2019 at 23:29, Max Solaris <s0l6r1s@gmail.com> wrote:
Hi,
I’m a long time user with Freeswitch, however I’m getting a strange issue I never found over past years.


If I take a look at fs_cli console and look at “show calls” result I can see a great list of calls with state:

CS_INIT,,,,,,,,DOWN


This call count is increasing and is eating hardware resources until CPU / MEMORY leave the server unresponsive…

How to clear these “zombie calls?”
Is there an explaination for calls being frozen with this state?


Just to clarify I’m using FS 1.6.20 git 43a9fe with Debian 8 ( I know it’s old, but can’t update for policy reason.. )


Thank you for your help, FS rocks!
Max
_________________________________________________________________________

The FreeSWITCH project is sponsored by SignalWire https://signalwire.com
Enhance your FreeSWITCH install with disruptive priced SMS and PSTN services.
Build your next product on our scalable cloud platform.

Join our online community to chat in real time https://signalwire.community

Professional FreeSWITCH Services
sales@freeswitch.com
https://freeswitch.com

Official FreeSWITCH Sites
https://freeswitch.com/oss
https://freeswitch.org/confluence
https://cluecon.com

FreeSWITCH-users mailing list
FreeSWITCH-users@lists.freeswitch.org
http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
https://freeswitch.com
--
Regards,

David Villasmil
phone: +34669448337
_________________________________________________________________________

The FreeSWITCH project is sponsored by SignalWire https://signalwire.com
Enhance your FreeSWITCH install with disruptive priced SMS and PSTN services.
Build your next product on our scalable cloud platform.

Join our online community to chat in real time https://signalwire.community

Professional FreeSWITCH Services
sales@freeswitch.com
https://freeswitch.com

Official FreeSWITCH Sites
https://freeswitch.com/oss
https://freeswitch.org/confluence
https://cluecon.com

FreeSWITCH-users mailing list
FreeSWITCH-users@lists.freeswitch.org
http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
https://freeswitch.com