Metricbeat pod logs: "metricset":common.MapStr{"name":"ec2", "period":300000}, "service":common.MapStr{"type":"aws"}}, Private:interface {}(nil), TimeSeries:true}, Flags:0x0} (status=400): {"type":"mapper_parsing_exception","reason":"failed to parse field [aws.tags.kubernetes] of type [keyword] in document with id 'b1okqG8B_OfmuHL1SKga'. Preview of field's value: '{io/cluster/lakeplacid-eksCluster-0a550bd=owned}'","caused_by":{"type":"illegal_state_exception","reason":"Can't get text on a START_OBJECT at 1:365"}}
Hi @eyesmoker, thanks for posting your question here. This issue is caused by label/annotation dots ( . ) creating hierarchy in Elasticsearch documents. Let me create a github issue to fix this. Thank you!
Thank you very much. Is there a workaround for this? Is this happening only in 7.5.0 and do you recommend 7.5.1?
Please point me to the GitHub issue.
This is an important feature requirement for us to see ec2 observability.
Here is the ticket: https://github.com/elastic/beats/issues/15843
Unfortunately I don't see a workaround besides not using dot in kubernetes labels. Or if you don't mind collecting metrics without the label, you can use cloudwatch metricset with config:
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.