I’m new to tickstack and snmp and trying to understand the data I’m getting collected into influx related to an snmp request being made through telegraf.
From a shell in ubuntu I run snmpwalk with the following results.
snmpwalk -v2c -ccommunity x.x.x.x 1.3.6.1.4.1.14823.2.3.3.1.2.2.1.7
iso.3.6.1.4.1.14823.2.3.3.1.2.2.1.7.240.92.25.202.17.40.0 = INTEGER: 0
iso.3.6.1.4.1.14823.2.3.3.1.2.2.1.7.240.92.25.202.17.40.1 = INTEGER: 13
Note: if run multiple times the first oid almost always returns 0, the second generally ranges between 8-15.
In my telegraf.conf I have:
[agent]
interval = “10s”
round_interval = true
metric_buffer_limit = 1000
flush_buffer_when_full = true
collection_jitter = “0s”
flush_interval = “10s”
flush_jitter = “0s”
debug = false
quiet = false
logfile = “installpath/telegraf/telegraf.log”
[[inputs.snmp]]
agents = [“x.x.x.x”]
version = 2
community = “community”
name = “snmp” # This is used for all the fields, of which there are none (yet).
[[inputs.snmp.table]]
name = “ap_util”
[[inputs.snmp.table.field]]
name = “util4”
oid = “.1.3.6.1.4.1.14823.2.3.3.1.2.2.1.7”
In influx I see a row of data every 10 seconds for util4 with either a value of 0 or ranging between 8-15 (the 8-15 range is what I was expecting to see when I first started looking into why we were getting zeros as a outlier).
So questions:
- Shouldn’t I be seeing two rows of data in influx? One for each snmpgetnext found during the walk?
- It seems like sometimes influx gets the data for the first returned oid in the walk (~25% of the time) and the second oid returned in the walk (the other ~75% of the time). Are both values being written to the same data row in influx and I’m seeing race conditions with one overwriting the other? Some other explanation?
Thanks.
