I encounter an issue that i am not able to resolve. The following problem is found on one of my environment but not on another one, and i can't find where the difference lies.
I am trying to have more than one SNMP server in my environment to simulate many components to supervise. For each component a specific conf file is generated without specifying the agentAddress. The agentAddress is specified in the following command that launch my snmpd process:
/usr/local/sbin/snmpd -Lsd -Lf /home/[...]/snmp/snmp.log -g users -I -smux,mteTrigger,mteTriggerConf -p /home/[...]/snmp/snmpd.pid -c /home/[...]/snmp/snmpd.conf --agentAddress=199.13.1.1:161
What I await and what is present on the working environment is having only one agentAddress to access my snmp base:
udp 0 0 199.103.1.1:161 0.0.0.0:* 71638/snmpd
On my second environment, it also open a localhost access (like below); then while I try to launch another process, a conflict appear due to 127.0.0.1:161 being already used.
udp 0 0 199.103.1.1:161 0.0.0.0:* 71638/snmpd
udp 0 0 127.0.0.1:161 0.0.0.0:* 71638/snmpd
Note: configuration file and command line are generated by code which is identitic. So I bet my issue comes from a network binding mistake that I would have done (or forbade to undone).
Hope someone can see what i'm missing. Any question to help you help me, i will answer ;) Thanks in advance.
K