error "*** was collected before with the same name and label values"

[marcolan018]

An error has occurred during metrics collection:

300 error(s) occurred:
* collected metric collectd_PCHMonitor__User_BSOS1_SweepFramework_SweepQueue_AbandonedContentDeletionSweep_gauge gauge:<value:0 >  was collected before with the same name and label values
* collected metric collectd_PCHMonitor__CPU_gauge gauge:<value:150 >  was collected before with the same name and label values

When using collectd exporter v0.4.0, we run into above error. By searching in google, seems this error is due to duplicate metrics. We are aware the best option to resolve this problem is to remove the duplicate ones. But compared with this, with similar configuration, the metrics works well for Graphite Writer. Seems that means this exporter is not flexible enough.

We found a possible workaround here: prometheus/node_exporter#377 (comment). Not sure whether collectd exporter support this.

[josephlim75]

Hey bluezoe,

Do you get any solution for this duplicate metrics issue in collectd exporter ?

[ainiml]

getting the same problem

* collected metric "atNetAddress" { label:<name:"atIfIndex" value:"58" > label:<name:"atNetAddress" value:"1.10.254.254" > gauge:<value:1 > } was collected before with the same name and label values

[eero-t]

@octo is this related to #78 ?

[octo]

This does sound like a duplicate metric. There is a small chance this is caused by the name mangling mapping to distinct collectd metrics to one Prometheus metric. In my experience it is much more likely that there are legitimately two metrics with the same identity, as the file system metrics mentioned by OP.

#78 is talking about a different issue, where there is a mismatch in the number of values/data sources expected and present.