My asterisk server to handle calls from sip randomly, three times a week, asterisk does not crash, no error appears on the console and debug log, not only handles calls sip, when the error occurs every GW lose the registry with the server.
I keep getting calls dahdi trunks, but are not delivered because the extensions are all disconnected.
All commands are executed, with the exception of Core restart now “and” Core stop now ", nothing happens when these commands are executed. To restart asterisk, I have to kill the process by linux console.
The server is Intel Xeon with 4GB RAM, the OS is Debian 6.0 64 bit version, there are seven D-Link DVG-2032 FXS GW connected and some Polycom phones, the total numer of connected peers is 245, we have at most 10 simultaneous calls.
The 64 bit version of Asterisk is already stable, someone has gone through a similar problem?
I got the same problem with Asterisk 1.4 a while ago when I used both DAHDI and SIP.
When I lost the internet connection it tried to reconnect to my SIP provider.
For some reason I also lost the connection to my other SIP devices I have on my own network.
When the internet got up again the problem was solved, but if it did not the problem persisted.
What I did was adding these two register lines in [general] context of sip.conf.
registertimeout=4
; Number of SIP REGISTER messages to send to a SIP Registrar before
; giving up. Default 0 (no limit)
registerattempts=3
; Number of seconds to wait for a response from a SIP Registrar before
; classifying the SIP REGISTER has timed out. Default 20 sec
With this it will try to register with your SIP provider 4 times and each time during 3sec each time.
After the 4:th time it will stop trying and reconnect your local SIP extensions.
Now when I only uses SIP I don’t uses them anymore!
Thanks, i will try to put these parametres and see if can solve, but i dont think so…
Im using 1.6.2.18 asterisk version and only have local SIP extensions.
External calls are made only by dahdi and by one IAX Trunk for another Asterisk Server on a branch office.
These two servers has the same configuration, with the same number of peers and simultaneous calls.
The erro i describe occurs on two servers, never at the same time.