We would like to get metrics on how many index failures we are getting. The only thing that seems close looking at the /system/metrics api route is the org.graylog2.periodical.IndexFailuresPeriodical.queueSize.count metric. However the data that I see in there does not seem to match with what I see on the /system/indices/failures page that shows a list of index failures? Is there a metric that can be used to have a count of every index failure that occurs?
If you want to to check number of unprocessed messages in node check REST API ( api/cluster/NODE_ID/journal ) which can return value:
"uncommitted_journal_entries":,
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.