eG Monitoring
 

Measures reported by HoldQueueTest

Whenever a Postfix mail server detects spam messages/malware, then those messages are automatically placed in the hold queue. By default, the administrator defines smtpd access policies, or cleanup header/body checks that may cause messages to be automatically diverted from normal processing and be placed indefinitely in the hold queue. The messages remain in the hold queue until the administrator intervenes. The messages in the hold queue are barred from periodic delivery attempts. Messages can stay in the hold queue longer than the maximum lifetime specified for the hold queue. The old messages in the hold queue can be released only by moving those messages into the maildrop queue. Once the messages are moved into the maildrop queue, the messages will get a new timestamp and will be attempted delivery. The messages that are relatively new in the hold queue can be moved directly to the deferred queue. If the hold queue consists of too many messages, then the hold queue may grow indefinitely and at some point, the queue may not be able to accommodate new messages resulting in a bottleneck of the queue. To avoid such bottlenecks and figure out the exact time duration for which messages are in the hold queue, it is necessary for the administrator to periodically monitor the hold queue. The HoldQueueTest helps administrators to identify the time duration from which the hold queue was bombarded with messages.

This test periodically monitors the hold queue of the target Postfix mail server and reports the total size of the queue as well as the split up of the message count in terms of time duration i.e., the number of messages that were in the queue for a specified time duration.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Queue_size Indicates the total size of the queue. Number An unusually high number of messages in the queue is indicative of any problem with the corresponding queue or its end points.

The Detailed diagnosis of this measure lists the message count for each domain.
Less_than_5_mins Indicates the number of messages that were in the queue for less than 5 minutes. Number  
Between_5_to_10_mins Indicates the number of messages that were in the queue for a time duration between 5 to 10 minutes. Number  
Between_10_to_20_mins Indicates the number of messages that were in the queue for a time duration between 10 to 20 minutes. Number  
Between_20_to_40_mins Indicates the number of messages that were in the queue for a time duration between 20 to 40 minutes. Number  
Between_40_to_80_mins Indicates the number of messages that were in the queue for a time duration between 40 to 80 minutes. Number  
Between_80_to_160_mins Indicates the number of messages that were in the queue for a time duration between 80 to 160 minutes. Number  
Between_160_to_320_mins Indicates the number of messages that were in the queue for a time duration between 160 to 320 minutes. Number  
Between_320_to_640_mins Indicates the number of messages that were in the queue for a time duration between 320 to 640 minutes. Number  
Between_640_to_1280_mins Indicates the number of messages that were in the queue for a time duration between 640 to 1280 minutes. Number  
More_than_1280_mins Indicates the number of messages that were in the queue for more than 1280 minutes. Number A high value for this measure is a cause of concern as this may be an indication of spam messages and malware which may have been filtered out.