Project

General

Profile

Bug #4197

Updated by keith over 4 years ago

I observed on a Debian system that systemd is starting the osmo-sip-connector before the system is fully up and all network interfaces are configured. 
 This results in: 

 <pre> 
 DLGLOBAL NOTICE <0004> telnet_interface.c:104 Available via telnet 127.0.0.1 4256 
 DMNCC NOTICE <0001> mncc.c:1021 Scheduling MNCC connect 
 DSIP NOTICE <0000> :0 su_source_port_create() returns 0x55a3df9dd5d0 
 DSIP NOTICE <0000> :0 nta: bind(172.16.0.1:5050;transport=*): Cannot assign requested address 
 DSIP NOTICE <0000> :0 nua: initializing SIP stack failed 
 DSIP ERROR <0000> main.c:163 Failed to initialize SIP. Running broken 
 </pre> 

 This is not terribly useful.  

 Instead we could: 

 1) exit (in the hopes that the OS will bring us back up again and now the interface I/F is configured. configured 

 2) retry to bind at regular intervals. intervals 

 3) In the case of systemd, add the following (i believe) to the service file: 

 <pre> 
 After=network-online.target 
 Wants=network-online.target 
 </pre> 

 However, I'm no systemd expert. 

Back

Add picture from clipboard (Maximum size: 48.8 MB)