eG Monitoring
 

Measures reported by XChgAssistantsTest

The Microsoft Exchange Mailbox Asssistant service provides functionality for Calendar Attendant, Resource Booking Attendant, Out of Office Assistant, and Managed Folder Mailbox Assistant.

The Exchange Assistants can be either event-based Assistants or time-based Assistants. The event-based Assistants start to process mailboxes on the occurrence of an event, such as on a change of Out-of-Office (OOF) information in one or more mailboxes. The time-based Assistants process the mailboxes periodically. Each time-based Assistant deploys an Assistants Driver that periodically checks whether the current time is within in a specified time window. When the current time reaches the specified time window, the Assistants Driver invokes the corresponding time-based Assistant. The time-based Assistant then obtains a list of mailboxes from the database and starts to process them.

To determine whether there are any mailbox assistant-related issues, use the XchgAssistants test.

 The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Events_in_queue Indicates the current number of events in the in-memory queue waiting to be processed by the assistants. Number Ideally, the value of this measure should be low at all times. High values may indicate a performance bottleneck.
Events_processed_rate Indicates the number of events processed per second. Events/Sec While sporadic dips in this value can be ignored, a consistent decrease could be a cause for concern, and may warrant a thorough investigation.
Events_polled_rate Indicates the number of events polled per second. Events/Sec  
Polling_delay Indicates the current latency between when the most recent MAPI event was polled and when the event was created. Secs  
Average_event_queue_time Indicates the average time (in seconds) that the event lives in the dispatcher queue. Secs Ideally, the value of this measure should be low. A very high value indicates that there are many events in queue with long waiting times. This in turn indicates a processing bottleneck.
Avg_event_processing_time Indicates the average time (in seconds) that the assistants took for processing events. Secs Ideally, the value of this measure should be low. A very high value indicates that there are many events in queue with long waiting times. This in turn indicates a processing bottleneck.
Avg_queue_size_of_event_dispatchers Indicates the average queue size of event dispatchers. Number A gradual but steady increase in the value of this measure over time, could indicate a problem with the dispatcher queue.
Avg_mailbox_processing_time Indicates the average processing time of mailboxes for time-based assistants. Secs A low value is typically desired.
Failed_event_dispatchers Indicates the percentage of Event Dispatchers that are in failure mode, currently. Percent  
Mailboxes_processed_rate Indicates the rate at which time-based assistants processed mailboxes. Operations/Sec  
Number_of_threads_used Indicates the current number of Threads used from the CLR thread pool. Number