-
Notifications
You must be signed in to change notification settings - Fork 220
Description
Upon trying to deploy the KPI dashboard yesterday via cid-cmd, all kpi Athena views and datasets are created, except for kpi_tracker which fails to create the Athena view due to the missing "tags_json" column in "summary_view."
Currently, I have the following resource_tags independently called out in the summary_view:
resource_tags['user_team'] "tag_team"
, resource_tags['user_cost_center'] "tag_cost_center"
, resource_tags['user_service'] "tag_service"
, resource_tags['user_component'] "tag_component"
, resource_tags['user_cluster'] "tag_cluster"
, resource_tags['user_owner'] "tag_owner"
, resource_tags['user_clustername'] "tag_clustername"
, resource_tags['user_name'] "tag_name"
, resource_tags['user_tenant'] "tag_tenant"
, resource_tags['user_environment'] "tag_environment"
Because kpi_tracker is looking for the map<string,string> format of the resource_tags column as tags_json, the view fails to create. How should I either change my summary_view to accommodate the kpi_tracker query, or change the kpi_tracker query to accomodate the resource_tags format I am already using? Looking for a suggestion that may reduce dataset size, and improve efficiency, but if changes need to be made to the summary_view query, I need to account for the changes required on production dashboards.