Retrieving ILM metrics - pod fails

[m4sb5r]

I have tried to use ILM metrics feature added in PR #513. This setup is running on kubernetes and I have the pod running with the flag:
Command: elasticsearch_exporter --es.ilm
When I start the pod it fails with the following error message:

panic: inconsistent label cardinality: expected 3 label values but got 1 in []string{"STOPPED"}goroutine 380 [running]:github.com/prometheus/client_golang/prometheus.MustNewConstMetric(...)        /go/pkg/mod/github.com/prometheus/client_golang@v1.12.2/prometheus/value.go:107github.com/prometheus-community/elasticsearch_exporter/collector.(*IlmStatusCollector).Collect(0xc000305180, 0xc000590120)        /app/collector/ilm_status.go:159 +0x3c8github.com/prometheus/client_golang/prometheus.(*Registry).Gather.func1()        /go/pkg/mod/github.com/prometheus/client_golang@v1.12.2/prometheus/registry.go:446 +0x10dcreated by github.com/prometheus/client_golang/prometheus.(*Registry).Gather        /go/pkg/mod/github.com/prometheus/client_golang@v1.12.2/prometheus/registry.go:538 +0xb0c

The pod image being used is quay.io/prometheuscommunity/elasticsearch-exporter:master. I wouldn't know where to start troubleshooting this. Sorry if this issue doesn't meet the standard in some way, since this is the first time I have ever opened an issue. Let me know how if this is the case and I will try my best to improve.

[tortuegenialez]

Same issue over here

Exporter builded from master branch, go version go1.19.6 linux/amd64 on a debian host.
Command line: ./elasticsearch_exporter --es.uri='http://my.shiny.cluster:9200' --web.listen-address=':9122' --log.output="stdout" --log.level="debug" --es.ilm
As soon as I query the /metics endpoint the exporter panic

The issue seems to come from an inconsistency in the function NewIlmStatus were three statuses are used:

elasticsearch_exporter/collector/ilm_status.go

Line 82 in bc05ee5

ilmStatuses, nil,

and the collector were only one status is provided:

elasticsearch_exporter/collector/ilm_status.go

Line 163 in bc05ee5

status,

[sysadmind]

Thanks for tracking that down @tortuegenialez. That made understanding the fix rather trivial. PR is open to resolve.