can't get data from device, but snmpwalk does.

Exporters and Metrics snmp_exporter
1

this is my snmp.yml，I have used a new template, sysuptime can be obtained, and ifnumber can also be obtained normally, but sysname has been invalidated. What is the reason for this? My snmp_ Exporter startup did not see relevant error logs

image
image1386×76 10.2 KB

image
image1878×1394 197 KB

image
image1272×1532 178 KB

2

bro，I have the same problem. Have you solved it?

3

I don’t have this problem. My problem is described at github .com/prometheus/snmp_exporter/discussions/1202#discussioncomment-10021820

I will repeat that information here:

We seem to be having a similar problem with about half of our Ethernet switches in one datacenter. But the problem seems randomly distributed between Cisco and Juniper switches. I’ve set the scrape_interval to be 300s and the scrape_timeout to be 200s, but almost all of the failures have a very suspicious looking scan_duration of right around 20s. We are only querying if_mib.

Is there any other information that I can provide?

This appears to be related to github. com/prometheus/snmp_exporter/issues/367 and github. com/prometheus/snmp_exporter/issues/850 but both of those issues were closed because the problem was supposedly fixed.

I’ve also posted about this issue at old.reddit. com/r/PrometheusMonitoring/comments/1e0qjsz/differences_between_snmp_exporter_and_snmpwalk/? and in the CNCF Slack channel Prometheus at cloud-native.slack. com/archives/C167KFM6C/p1720710093899439

Are there any other places that people would suggest that could be good resources for this question?

Or any other suggestions in general?

Thanks!

PS: Sorry that I had to break all the links, but apparently new users aren’t allowed to post links here.

1 Like