Logs recorded for pools and virtual services are categorized as Non-Significant (green) and Significant (red).  At times, the log histogram ribbon might show a yellow question mark, indicating that logs are missing near that time frame.  There are scenarios which can cause logs to be in a missing state.

Purged Logs

By default, both Significant logs (errors) and Non-Significant logs are not automatically forwarded by Service Engines to the Controller Cluster for indexing.  They are instead offloaded from Service Engines and replicated across all Controllers when requested by an administrator. Generally, Controllers are recommended to have larger storage capacity for greater log retention.  Service Engines are usually configured with significantly smaller storage and, so the SE might need to rotate logs by purging the oldest first.  If a Controller requests purged logs, the Service Engine reports them as missing, along with a rough count of the number of purged logs for the time period.

Rate Limited Logs

To prioritize service delivery tasks (such as load balancing), Service Engines can automatically rate limit the number of logs per second that it records.  The number varies depending on the performance and capacity of the underlying hardware and also on the use of the hardware.  For example, SSDs tend to have significantly faster write operations per second than hard disk drives and, so they are capable of supporting greater logging per second.