Topic Last Modified: 2013-02-21

In Microsoft Exchange Server 2013, messages that can't be successfully delivered are subject to various retry, resubmit, and expiration deadlines based on the message's source and destination. Retry is a renewed connection attempt with the destination. Resubmit is the act of sending messages back to the Submission queue for the categorizer to reprocess. The message timeoutexpires after all delivery efforts have failed over a specified period of time. After a message expires, the sender is notified of the delivery failure. Then the message is deleted from the queue.

In all three cases of retry, resubmit, or expire, you can manually intervene before the automatic actions are performed on the messages.

For instructions on how to configure these intervals, see Configure Message Retry, Resubmit, and Expiration Intervals.

Configuration options for message retry

When a transport server can't connect to the next hop, the queue is put in a status of Retry. Connection attempts continue until the queue expires or a connection is made.

Configuration Options for Automatic Message Retry

The configuration options that are available for message retry intervals are described in the following table.

Configuration options that are available for message retry intervals

Parameter or key name Default value Where to configure Description

QueueGlitchRetryCount

4

EdgeTransport.exe.config

This key specifies the number of connection attempts that are immediately tried when a transport server has trouble connecting with the destination server. Such connection problems are typically caused by very brief network outages.

Valid input for this key is an integer from 0 through 15.

Typically, you don't have to modify this key unless the network is unreliable and continues to experience many accidentally dropped connections.

QueueGlitchRetryInterval

00:01:00 or 1 minute

EdgeTransport.exe.config

This key controls the connection interval between each connection attempt that's specified by the QueueGlitchRetryCount key.

Typically, you don't have to modify this parameter unless the network is unreliable and continues to experience many accidentally dropped connections.

TransientFailureRetryCount

6

Set-TransportServicecmdlet or server properties in the Exchange Administration Center (EAC)

This parameter specifies the number of connection attempts that are tried after the connection attempts that are controlled by the QueueGlitchRetryCount and QueueGlitchRetryInterval keys have failed. Connection problems that exhaust the QueueGlitchRetryCount and QueueGlitchRetryInterval keys can be caused by server restarts or cached DNS lookup failures.

Valid input for this parameter is an integer from 0 through 15. If you set this parameter to 0, the next connection attempt is controlled by the OutboundConnectionFailureRetryInterval parameter.

TransientFailureRetryInterval

  • Transport service on Mailbox servers: 00:05:00 or 5 minutes

  • Edge Transport servers: 00:01:00 or 10 minutes

Set-TransportServicecmdlet or server properties in the EAC

This parameter controls the connection interval between each connection attempt that's specified by the TransientFailureRetryCount parameter.

To specify a value, enter it as a time span: dd.hh:mm:ss where d = days, h = hours, m = minutes, and s = seconds.

OutboundConnectionFailureRetryInterval

  • Transport service on Mailbox servers: 00:10:00 or 10 minutes

  • Edge Transport Servers: 00:30:00 or 30 minutes

Set-TransportService cmdlet or server properties in the EAC

This parameter specifies the retry interval for outbound connection attempts that have previously failed. The previously failed connection attempts are controlled by the TransientFailureRetryCount and TransientFailureRetryInterval parameters.

To specify a value, enter it as a time span: dd.hh:mm:ss where d = days, h = hours, m = minutes, and s = seconds.

MessageRetryInterval

00:01:00 or 1 minute

Set-TransportService cmdlet

This parameter specifies the retry interval for individual messages that have a status of Retry. We recommend that you don't modify the default value unless Microsoft Customer Service and Support advises you to do this.

MailboxDeliveryQueueRetryInterval

00:05:00 or 5 minutes

EdgeTransport.exe.config

This key specifies how frequently the queues try to connect to the Mailbox Transport Delivery service for a destination mailbox database that can't be successfully reached.

To specify a value, enter it as a time span: dd.hh:mm:ss where d = days, h = hours, m = minutes, and s = seconds.

Valid input for this key is from 00:00:01 through 1.00:00:00.

Return to top

Configuration options for manual message retry

When a delivery queue is in the status of Retry, you can manually force an immediate connection attempt by using Queue Viewer in the Exchange toolbox or the Retry-Queue cmdlet in the Shell. The manual retry attempt overrides the next scheduled retry time. If the connection isn't successful, the retry interval timer is reset. The delivery queue must be in a status of Retry for this action to have any effect.

For more information, see the "Retry queues" section in Manage Queues.

Return to top

Configuration options for delay DSN messages

After each message delivery failure, the Edge Transport server or the Transport service on the Mailbox server generates a delay delivery status notification (DSN) message and queues it for delivery to the sender of the undeliverable message. This delay DSN message is sent only after a specified delay notification timeout interval, and only if the failed message wasn't successfully delivered during that time. By default, the delay notification timeout interval is 4 hours. This delay prevents the sending of unnecessary delay DSN messages that may be caused by temporary message transmission failures. The sending of delay DSN notification messages can be selectively enabled or disabled for messages that originate inside or outside the Exchange organization.

The configuration options that are available for delay DSN notification messages are described in the following table.

Configuration options that are available for delay DSN notification messages

Parameter name Default value Location Description

DelayNotificationTimeOut

4:00:004 hours

Set-TransportService or server properties in the EAC

This parameter specifies how long the server waits before it sends a delay DSN message to the sender. The value of this parameter should always be greater than the value of the TransientFailureRetryCount parameter multiplied by the value of the TransientFailureRetryInterval parameter.

To specify a value, enter it as a time span: dd.hh:mm:ss where d = days, h = hours, m = minutes, and s = seconds.

ExternalDelayDSNEnabled

$true

Set-TransportConfig

This parameter specifies whether delay DSN messages can be sent to message senders who are outside the Exchange organization.

Valid input for this parameter is $true or $false.

InternalDelayDSNEnabled

$true

Set-TransportConfig

This parameter specifies whether delay DSN messages can be sent to message senders who are inside the Exchange organization.

Valid input for this parameter is $true or $false.

Note:
On Exchange 2007 Hub Transport servers, all ExternalDSN* and InternalDSN* parameters are available on the Set-TransportServer cmdlet, not the Set-TransportConfig cmdlet. If you have any Exchange 2007 Hub Transport servers in your organization, you need to make changes to these values using the Set-TransportServer cmdlet on each Exchange 2007 Hub Transport server.

Return to top

Configuration options for message resubmission

Message resubmission sends undelivered messages back to the Submission queue to be reprocessed by the categorizer.

Automatic message resubmission

Undelivered messages are automatically resubmitted if the delivery queue is in the status of Retry and has been unable to successfully deliver any messages for a specified period of time. That period of time is controlled by the MaxIdleTimeBeforeResubmit key in the EdgeTransport.exe.config application configuration file. Only messages in delivery queues are candidates for automatic resubmission.

To specify a value, enter it as a time span: dd.hh:mm:ss where d = days, h = hours, m = minutes, and s = seconds.

The default value is 12:00:00 or 12 hours.

Manual Message Resubmission

You can manually resubmit messages that have the following status on a Hub Transport server or an Edge Transport server:

  • Delivery queues that have the status of Retry. The messages in the queues must not be in the Suspended state.

  • Messages that are in the Unreachable queue and aren't in the Suspended state.

  • Messages that are in the poison message queue.

For more information about the poison message queue and the Unreachable queue, see "About the Poison Message Queue and the Unreachable Queue" in the topic Queues.

If you want to manually resubmit messages that are located in delivery queues or the Unreachable queue without waiting for the time that's specified by the MaxIdleTimeBeforeResubmit parameter to pass, you need to use the Retry-Queue cmdlet with the Resubmit parameter. To manually resubmit messages that are located in the poison message queue, you can use Queue Viewer or the Resume-Message cmdlet to resume the message. For more information, see the "Resubmit messages in queues" section in Manage Queues.

Another way that you can manually resubmit messages is to suspend the messages, export the messages to text files that have the .eml file name extension, and then copy the .eml files to the Replay directory on any Mailbox server or Edge Transport server. This resubmission method works for messages that are located in delivery queues or the Unreachable queue. Messages that are located in the poison message queue are already in the Suspended state. Messages that are located in the Submission queue can't be suspended or exported.

Note:
When you export messages from a queue, you don't remove the messages from the queue. After you export the messages and successfully resubmit them by using the Replay directory, you should remove the suspended messages to avoid duplicate message delivery.

For more information, see Export Messages from Queues.

Return to top

Configuration options for message expiration

The message expiration timeout interval specifies the maximum length of time that an Edge Transport server or a Hub Transport server tries to deliver a failed message. If the message can't be successfully delivered before the expiration timeout interval has passed, an NDR that contains the original message or the message headers is delivered to the sender.

Automatic message expiration

The message expiration timeout interval is controlled by the MessageExpirationTimeOut parameter in the Set-TransportService cmdlet or in the server properties in the EAC.

To specify a value, enter it as a time span: dd.hh:mm:ss where d = days, h = hours, m = minutes, and s = seconds.

The default value is 2.00:00:00 or 2 days. The valid input range for this parameter is from 00:00:05 through 90.00:00:00.

Manual Message Expiration

Although you can't manually force messages to expire, you can manually remove messages from any queue, except the Submission queue, with or without an NDR.

For more information, see the "Remove messages from queues" section in Manage Messages in Queues.

Return to top