sly geode
Hi everyone,
I need a bit of a steer as I'm not sure where even to start with this one.
I've got a particular system which appears to be reporting double stats.
The dashboard is fine, the graph is fine, it displays fine for other systems. It's literally just this one.
Any thoughts?
Thanks,
Dan
fierce wharf
@sly geode If you replace this panel query with below, Does it fix this issue?
label_replace(node_avg_processor_busy{datacenter=~"$Datacenter",cluster=~"$Cluster",node=~"$Node"},"instance", "", "instance", ".*")
@fierce wharf that caused the graph to error out; execution: vector cannot contain metrics with the same labelset
fierce wharf
@sly geode This error could indicate that there are duplicate metrics for the same timestamp in Prometheus. Can you check if you are not running the same poller in Harvest twice?
sly geode
According to the system page within the nabox there is only a single instance of the system in question.
The same can be said for the harvest.yml file.
Only a single nabox-harvest2 container running.
Anywhere else specifically I'd need to check?
fierce wharf
Okay. Could you share output command mentioned here? https://netapp.github.io/harvest/24.02/help/config-collection/#nabox
Also double check that cat /opt/harvest2-conf/harvest.yml file has no duplicate poller listed.
Also if you run below prom query in prometheus. Do you get any results?
count(node_avg_processor_busy) by (node) > 1