here are the results (tested with the new plugin): 1. on Firewalls where both Versions (v1 + v2c) are enabled.
the check without version specification does work: r...@opsview:~# ./check_snmp_linkstatus -H 192.168.10.254 -I port1 -i -o WARNING: Running from command line. This could change statistics for the next run from Nagios OK - port1 (lan.meuse) is up, throughput (in/out) 9.55 Kbps/11.9 Kbps, 0%/0.01%|throughput_in=9784b throughput_out=12192b check with version v 1 specified works but does not read throughput?: r...@opsview:~# ./check_snmp_linkstatus -H 192.168.10.254 -I port1 -i -o -v 1 WARNING: Running from command line. This could change statistics for the next run from Nagios OK - port1 (lan.meuse) is up, throughput (in/out) 0 bps/0 bps, 0%/0%|throughput_in=0b throughput_out=0b check with version 2c specified works but throughput is a mess: r...@opsview:~# ./check_snmp_linkstatus -H 192.168.10.254 -I port1 -i -o -v 2c WARNING: Running from command line. This could change statistics for the next run from Nagios OK - port1 (lan.meuse) is up, throughput (in/out) 10.58 Gbps/19.52 Gbps, 10842.35%/19998.12%|throughput_in=11369037392b throughput_out=20969555048b test with disabling wheather v1 or v2c on the firewall result in correct data for v2c and no throughput data for v1. 2. on switches with only v2c enabled (no option for v1): the check without version specification does not work: r...@opsview:~# ./check_snmp_linkstatus -H 192.168.10.200 -I swp00 -i -o WARNING: Running from command line. This could change statistics for the next run from Nagios UNKNOWN - Interface swp00 not found! check with v1 does work, although there is no option on the switch for version 1. furthermore the troughput result is somewhat unrealistic for a gigabit uplink: r...@opsview:~# ./check_snmp_linkstatus -H 192.168.10.200 -I swp00 -i -o -v 1 WARNING: Running from command line. This could change statistics for the next run from Nagios OK - swp00 (UP1) is up, throughput (in/out) 16 bps/0 bps, 0%/0%|throughput_in=16b throughput_out=0b check with v2c does work, although the result here for throughput/out is not possible (1000Mbit interface): r...@opsview:~# ./check_snmp_linkstatus -H 192.168.10.200 -I swp00 -i -o -v 2c WARNING: Running from command line. This could change statistics for the next run from Nagios OK - swp00 (UP1) is up, throughput (in/out) 29.66 Kbps/80.51 Gbps, 0%/8645.37%|throughput_in=30376b throughput_out=86453772600b here is the same check a few moments later with a more realistic result: r...@opsview:~# ./check_snmp_linkstatus -H 192.168.10.200 -I swp00 -i -o -v 2c WARNING: Running from command line. This could change statistics for the next run from Nagios OK - swp00 (UP1) is up, throughput (in/out) 20.21 Kbps/19.67 Kbps, 0%/0%|throughput_in=20704b throughput_out=20144b so long... juergen On 2009-12-31 15:38, Ton Voon wrote: > > OK, I think I know what is happening. Can you try the updated file > attached. > > My guess is that you have a device which is configured as SNMPv2 in > Opsview, but is actually setup as an SNMPv1 device. > > The check_snmp_linkstatus has some fallback code so that if SNMPv2 > fails, it re-connects with SNMPv1. However, the max_msg_size call is > done before this SNMPv1 reconnection. > > The file attached should set the size correctly. > > Can you try this, and also specify on the command line the exact > version (-v 1 for SNMPv1 and -v 2c for SNMPv2c)? > > Ton > > > > > _______________________________________________ Opsview-users mailing list [email protected] http://lists.opsview.org/lists/listinfo/opsview-users
