"Slave"-Bridge not registering to Master when using Tunnel

[bishop2k]

HI all,

my secondary "slave" 3cx is not registering to the Master anymore
when i check to use the Tunnel.

Unchecking Tunnel works fine.

Tunnel was working since it was configured. Master was rebooted yesterday.
All tunnelclients register just fine to the master.
Hosted. VM Debian9 Static IPv4. PBX Version 16.0.493

25.01.2020 14:44:04 - [CM504005]: Registration failed for: Lc:10003(@rz[<sip:10003@IP-OF-MASTER:5090/UDP>]); Cause: Cause: 503 Remote end of tunnel is not connected/REGISTER from 127.0.0.1:5080

25.01.2020 14:44:03 - Sent registration forLc:10003(@rz[<sip:10003@IP-OF-MASTER:5090/UDP>]) :ClientReg Send Req REGISTER from 0.0.0.0:0 tid=04300011ae986961 Call-ID=gCPg17p5LJbAtywzFxUEGQ..: REGISTER sip:IP-OF-MASTER:5090 SIP/2.0 Via: SIP/2.0/ ;branch=z9hG4bK-524287-1---04300011ae986961;rport Max-Forwards: 70 Contact: <sip:10003@IP-OF-SLAVE:5060;rinstance=51daaf64035d6881> To: <sip:10003@IP-OF-MASTER:5090> From: <sip:10003@IP-OF-MASTER:5090>;tag=248f935f Call-ID: gCPg17p5LJbAtywzFxUEGQ.. CSeq: 1 REGISTER Expires: 60 Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REGISTER, SUBSCRIBE, NOTIFY, REFER, INFO, MESSAGE, UPDATE Supported: replaces, timer X-bridge-selector: 10003 Content-Length: 0

Master Firewallrules:


any suggestions?
Reboot Master again? wait?

Thanks @ all !
Bish

 

[Evolute IT]

Your "secondary slave"? Have you connected 2 slaves to 1 master using the same bridge?

 

[bishop2k]

Hi Frederic,

Thank you for your kind reply.
It´s 2 PBXes in total so just one Slave connecting to one Master.

On the Master side "SIP Trunks" i have 1 SBC and 1 Webmeeting bridge registered


"Bridge Master-Tunnel" is not connected since Serverreboot yesterday.

Do i need some deeper Log Info?

Best, Bish

 

[Evolute IT]

Check the IP Blacklist on both systems. If the IP is banned it won't work.

 

[bishop2k]

I did already.
The IP of Master was listed in the Slave, so i deleted the Entry.
Both Systems show no Blocks concerning the relevant IP.
Maybe i forgot to check somewhere else?

I found this in the Log:
3CX Clients / Softphones ID: 12290

The IP IP_of_Master has been blacklisted for 186400 sec. (Expires at: 2020-01-27 17:50:53). Reason: 3CX Clients / Softphones blocked - 3CX API Component

deleted the entry in Blacklist and Setup the Bridge again without success.
Is there some block to find elsewhere or additionally reset "blacklisted for 186400 sec. " ?

 

[Evolute IT]

I did already.
The IP of Master was listed in the Slave, so i deleted the Entry.
Both Systems show no Blocks concerning the relevant IP.
Maybe i forgot to check somewhere else?

I found this in the Log:
3CX Clients / Softphones ID: 12290

The IP IP_of_Master has been blacklisted for 186400 sec. (Expires at: 2020-01-27 17:50:53). Reason: 3CX Clients / Softphones blocked - 3CX API Component

deleted the entry in Blacklist and Setup the Bridge again without success.
Is there some block to find elsewhere or additionally reset "blacklisted for 186400 sec. " ?

You can simply whitelist the IPs on both systems then click "Refresh Registration" on the slave bridge. Try it and see if it works.

 

[AWS2P]

Same ID 10003 used on Slave and Master? as it's slave who connects to master, could you see what's on logs on slave side?

Password is the good one? no change?
Have you tried to reboot Slave PBX ?

 

[bishop2k]

Thanks for the nice help here!

Behaviour persists
28.01.2020 00:18:34 - [CM504005]: Registration failed for: Lc:10003(@rz[<sip:10003@IP_OF_MASTER:5090/UDP>]); Cause: Cause: 503 Remote end of tunnel is not connected/REGISTER from 127.0.0.1:5080

28.01.2020 00:18:34 - Sent registration forLc:10003(@rz[<sip:10003@IP_OF_MASTER:5090/UDP>]) :ClientReg Send Req REGISTER from 0.0.0.0:0 tid=5a2ae64bffa4353f Call-ID=EzOX2Xi3iwNOdLw34_dWPA..:
REGISTER sip:IP_OF_MASTER:5090 SIP/2.0 Via: SIP/2.0/ ;branch=z9hG4bK-524287-1---5a2ae64bffa4353f;rport Max-Forwards: 70 Contact: <sip:[email protected]:5060;rinstance=e8436e1c174e33ca>
To: <sip:10003@IP_OF_MASTER:5090>
From: <sip:10003@IP_OF_MASTER:5090>;tag=9c93a93c Call-ID: EzOX2Xi3iwNOdLw34_dWPA.. CSeq: 1 REGISTER Expires: 60 Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REGISTER, SUBSCRIBE, NOTIFY, REFER, INFO, MESSAGE, UPDATE Supported: replaces, timer X-bridge-selector: 10003 Content-Length: 0

ends up in this:


IP and FQDN is reachable, SBC to Master works perfect.
Slave is dynamic IP(cable) does not change often but Master is static hosted.
Same ID 10003 on both Systems
Password checked but the Log shows no Passwd request since it looks like a "DNS" Problem
while DNS works for all other applications.

Is there any suggestion how to debug this Behaviour since it did occur after rebooting the master.

Will reboot the Slave and report if this helps.

Best @ all,

Bish

 

[conversant]

This is a bug, but 3CX support won't even take the time to confirm that this issue has been around since update 3. We had update 2 instances that were bridged successfully for months, but once updated to 3 and then 4, bridging no longer functions. We have confirmed on numerous update 3 and 4 instances that bridging throws the same error that you report and fails, but 3CX wants us to do their debugging for them! Fortunately, bridging is not a critical requirement for these customers, so it isn't a big issue. If you search the forum, you will find other reports of the same issue, so it is a known issue, but no action from 3CX.

 

[AWS2P]

tonight have tried to make a new test bridge from a cloud lightsail PBX and one W10 on premise
This seems to register on both side and calls works on both side too

 

[bishop2k]

This is a bug

I can confirm the following:

hours before way more entries:


The Bridge came back during the night with no changes at all.

change use tunnel "IP-Address and Port ext. 3CX-PBX " on the master from IP to FQDN
Bridge breaks again and does not come back when checking tunnel.

Its a throw of dice wether or not the bridge comes back "immediately"
"IP-Address and Port ext. 3CX-PBX " on the master was set to FQDN before and the bridge worked.

now wait if the bridge will rebuild again

can someone confirm that checkbox use tunnel "IP-Address and Port ext. 3CX-PBX " on the master works with FQDN and a dynamic Slave (even if the IP did not change since its cable...)?

happy debugging ...

if anyone is interested in the caputure, let me know:
SupportInfo_toQ8W0lYGbWK3ScGE0XzhI8o

Bish

 

[AWS2P]

can someone confirm that checkbox use tunnel "IP-Address and Port ext. 3CX-PBX " on the master works with FQDN and a dynamic Slave

on my side the 2 PBXs have public Static IP, and for first setting I've used FQDN on both side of the bridge, it works but like you sometimes randomly 3 or 4 time a day tunnel is lost and come back up after 2 minutes.
Have changed settings this night to use IPs instead of FQDN and see if behavior change, wait and see

some example log messages for bridge set with FQDN then IP

 

[bishop2k]

Bridge is back ... reliable? ...



is there any option to show "uptime" of the bridge? like SBC?