Recently we have seens bucket corruption issues in our cluster. Upon chekcing with support they concluded that the root cause might be disk latency issues. Is it possible to monitor the Disk Health parameters of indexer servers in Splunk ?
Take a look at the Distributed Management Console, it should help you detect problems on your indexers, including poor disk performance (located on the Resource dashboards). You can configure a single DMC to collect data from all of your indexers and search heads.