Article ID: 271545
Article Last Modified on 10/28/2006
APPLIES TO
- Microsoft Exchange Server 5.5 Standard Edition
This article was previously published under Q271545
SUMMARY
This KB article discusses the differences between message timeouts, and Open Retry timeouts.
MORE INFORMATION
As an example of how these two issues may be confusing, look at the following two articles in the Microsoft Knowledge Base:
157630 Adjustment of Connection Retry values on Override tab
178539 How to adjust message time-outs for the Exchange MTA
At first glance, these articles can seem contradictory; it is unclear whether mail is returned undeliverable after 24 hours, or after seven days (in the case of Normal priority mail).
Message timeouts define how long a message has to live within the message transfer system before the system returns the message undeliverable. A concept that is very similar to the Time to Live (TTL) value in IP packets. Open Retry intervals, on the other hand, define the length of time during which the message transfer agent (MTA) tries to connect to another MTA to deliver a message before giving up, and then returning the message as undeliverable.
Consider these two different hypothetical scenarios, assuming the simplest configuration of only two servers (MTA1 and MTA2):
Scenario 1
MTA1 attempts to send a message to MTA2, but MTA2 is down. Because MTA1 cannot open a connection to MTA2, MTA1 falls into a state of Open Retry, and the Max Open Retry counter starts to increment. By default, the retry interval (the length of time between connection attempts) is 600 seconds, or 10 minutes. By default, MTA1 tries to make a connection to MTA2 144 times before it giving up, and then returning any messages, which were queued for MTA2, to their originators as undeliverable. 144 retries multiplied by 600 seconds is equal to 86,400 seconds, which is approximately 24 hours. Thus, if MTA1 has mail to send to MTA2, and MTA2 is down for 24 hours, all mail, regardless of the priority, that was bound for MTA2 is returned as undeliverable. For additional information, click the following article number to view the article in the Microsoft Knowledge Base:
157630 Adjustment of Connection Retry values on Override tab
Scenario 2
MTA1 has mail queued for MTA2, but before it completes the delivery of all the mail, MTA1 shuts down; after a while, MTA1 is back online. The mail that is in the MTA1 queue does not timeout until the MTA on MTA1 detects that the default timeout period for the priority of a given message has elapsed. The defaults are two days for High priority, seven days for Normal priority, and ten days for Low priority messages. Thus, if MTA1 has mail queued for MTA2, and MTA1 shuts down completely for five days, when MTA1 is back online, all of the High priority mail that MTA1 had queued for all destinations is returned as undeliverable to the message originators. However, messages with a Low or Normal priority remain in the queue until they are delivered. For additional information, click the following article number to view the article in the Microsoft Knowledge Base:
178539 How to adjust message time-outs for the Exchange MTA
The importance of this issue is determined by which server is down: the server that has the mail queued (the Source), or the server that should receive the mail (the Destination).
If you have more than two MTAs in the site, or more than one route, which a message can use for delivery, then message rerouting factors into the equation. Remember that the MTA can reroute mail that is queued to non-secured queues, such as the X.400 Connector, Site Connector, Dynamic RAS Connector, and other MTAs. The MTA cannot reroute mail that has been queued to secured queues, such as the Internet Mail Service, and the Microsoft Exchange Connectors for MS Mail, cc:Mail, Lotus Notes, GroupWise, SNADS, and PROFS.
Additional query words: XCON
Keywords: kbinfo KB271545
