I have a couple of heavily loaded VMs, which I cannot reboot ideally.
My TX/RX packet count in ifconfig far exceeds a 32 bit number, meaning my snmp data is very sporadic.
How can I make use of the MIBs at 1.3.6.1.2.1.31.1.1.1 so I can use the Counter64 values instead of the Counter32 values for monitoring eth0 in/out packets?
Thanks
I have managed to fix this using an SNMPd config on my Linux VMs.
I basically exclude the counter32 values for the interface MIBs.
Example from snmpd.conf
## LibreNMS MIBs (restricted for faster polling)
## ---------------------------------------------------------
# Inc. System Info (name/loc) OIDs
view libre-mibs included .1.3.6.1.2.1.1
# Include Interface MIBs
view libre-mibs included .1.3.6.1.2.1.2
# Include Interface MIBs w/ 64 bit counters for high traffic boxes
view libre-mibs included 1.3.6.1.2.1.31
# Exclude 32 bit counters as high traffic boxes overflow
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.10
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.11
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.12
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.13
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.14
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.15
view libre-mibs. excluded .1.3.6.1.2.1.2.2.1.16
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.17
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.18
view libre-mibs excluded .1.3.6.1.2.1.2.2.1.19
This seems to force the usage of the 64 bit counters instead of 32 bit counters.
2 Likes
For ref, also fixes miscalculated port speeds in TP-Link Jetstream T1700G-28TQ
I also added as an include: 1.3.6.1.4.1.11863.6.4.1 specifically for TPLInk CPU and RAM stats