Hi.
I'm supervisor of network services in my office.
There is about 8 months that i'm working on migrating from our commercial voip product to opensource fusionpbx.
My team and I did all tasks and test them but there are two times that when we switch to fusionpbx as voip server and when work time starts, all extensions(endpoints) start to unreg and reg and different types of errors occur. I realy don't know where is the problem.
My system and configuration specifications are:
- Debian 9.1 X64
-Server: HP
- Ram: 16G
- CPUs: 8
Extensions: 2914
Phones: Yealink T19-T20-T22-T27-T28-T53 , ...
- Provisioning phones through DHCP and tftp (setting their sip server and expire , ...)
Features:
- IVRs
-Queues
- SRTP (rtp_secure_media=optional and SRTP=compolsury in phone web interface)
- codec= alaw,ulaw
- BLF
- aut-acl per extension to limit ip address of endpoint and extension number(ex: 172.21.6.6/32)
- 4 gateway (2 E1 gw, 2 FXO gw)
- setting prefix and gateway per extension based on it's number (scripting in dialplan)
- implementing Boss/Secretary feature and valid lists to call bosses (scripting in dialplan and defining boss and secretary number and valid list to call boss per boss)
- Inbound routes (one rule for all extensions DID and multiple lines for FXOs)
- connecting fusion and gateways via same network adapter which clients register (172.20.1.8)
- All gateways and phones and fusion server have local ip address
- Disabled Fail2ban adn iptables
- Changing Max sessions to 150000 and Max concurrent per sec to 30000 (switch.conf.xml)
- No Nat and UDP in phone web interface
- Expire: 60 in phone web interface
- But in Registeration page of fusion interface I see expression "UDP-NAT (unknown)" in front of each extension row
PROBLEM: I switch to fusion server via DNS at middle night and all extensions after about 1 hour register to this server. But tomorrow morning (7 am) when work time starts, after about 1 hour at 8 am, number of registered extensions reduce and increase frequently and calls drop with different cause errors (forbiden, user_not_register, ...). and there are many IPPhone@172.a.b.c logs instead of EXT@IP.
whould you help me please to solve this problem. I really confused that what is the problem.
I'm supervisor of network services in my office.
There is about 8 months that i'm working on migrating from our commercial voip product to opensource fusionpbx.
My team and I did all tasks and test them but there are two times that when we switch to fusionpbx as voip server and when work time starts, all extensions(endpoints) start to unreg and reg and different types of errors occur. I realy don't know where is the problem.
My system and configuration specifications are:
- Debian 9.1 X64
-Server: HP
- Ram: 16G
- CPUs: 8
Extensions: 2914
Phones: Yealink T19-T20-T22-T27-T28-T53 , ...
- Provisioning phones through DHCP and tftp (setting their sip server and expire , ...)
Features:
- IVRs
-Queues
- SRTP (rtp_secure_media=optional and SRTP=compolsury in phone web interface)
- codec= alaw,ulaw
- BLF
- aut-acl per extension to limit ip address of endpoint and extension number(ex: 172.21.6.6/32)
- 4 gateway (2 E1 gw, 2 FXO gw)
- setting prefix and gateway per extension based on it's number (scripting in dialplan)
- implementing Boss/Secretary feature and valid lists to call bosses (scripting in dialplan and defining boss and secretary number and valid list to call boss per boss)
- Inbound routes (one rule for all extensions DID and multiple lines for FXOs)
- connecting fusion and gateways via same network adapter which clients register (172.20.1.8)
- All gateways and phones and fusion server have local ip address
- Disabled Fail2ban adn iptables
- Changing Max sessions to 150000 and Max concurrent per sec to 30000 (switch.conf.xml)
- No Nat and UDP in phone web interface
- Expire: 60 in phone web interface
- But in Registeration page of fusion interface I see expression "UDP-NAT (unknown)" in front of each extension row
PROBLEM: I switch to fusion server via DNS at middle night and all extensions after about 1 hour register to this server. But tomorrow morning (7 am) when work time starts, after about 1 hour at 8 am, number of registered extensions reduce and increase frequently and calls drop with different cause errors (forbiden, user_not_register, ...). and there are many IPPhone@172.a.b.c logs instead of EXT@IP.
whould you help me please to solve this problem. I really confused that what is the problem.