= snmp issue in upgrading from sarge to etch =
After upgrading from Debian sarge to etch, there was a problem with snmp on one of my servers. No snmp information could be obtained. Note: this problem only occurred when I upgrade to etch on 10th of April, 2007. A previously upgraded server to etch, a few months previous did not have this problem.
Research and Debug
snmpwalk -v 1 -OS -c public localhost
netstat -a //check for snmp
The output of netstat showed:
udp 0 0 localhost:snmp *:*
So snmp had bound to 127.0.0.1 for some reason. Looking in /etc/snmp/* did not reveal anything.
Problem and Solution
ps -eaf | grep snmp snmp 8852 1 2 17:01 ? 00:00:00 /usr/sbin/snmpd -Lsd -Lf /dev/null -u snmp -I -smux -p /var/run/snmpd.pid 127.0.0.1
The snmpd had specifically been given a 127.0.0.1 starting parameter.
//change the following: SNMPDOPTS='-Lsd -Lf /dev/null -u snmp -I -smux -p /var/run/snmpd.pid 127.0.0.1'
//to: SNMPDOPTS='-Lsd -Lf /dev/null -u snmp -I -smux -p /var/run/snmpd.pid'
Done. Hope it works. The above config to /etc/default/snmpd was done on the original Snmpd & mrtg setup, however on upgrading, etch must have put a new default/snmpd into /etc/. Check with “snmpwalk -v 1 -OS -c public localhost” from localhost and/or from your monitoring server.