In Windows Active Directory (in connection with Exchange 2010), I am unsure about the semantic difference between mail: and proxyAddresses: attributes. As far as I can tell, mail: is one-valued whereas proxyAddresses: is multivalued and (apart from the possibility to include non-SMTP addresses) allows one value starting with SMTP as main address and several values starting with smtp as secondary addresses.
I think I recall that in former versions of Windows and/or Exchange the main proxyAddresses was always kept in sync with the mail attribute. But I now noticed that these are no longer automatically kept in sync (depending on how one edits the data). So it may happen that I have a user with
mail: AAA@example.com
proxyAddresses: SMTP:BBB@example.com
proxyAddresses: smtp:CCC@example.com
So any mail that user sends will use <BBB@example.com> as sender address; incoming mail addressed to either <BBB@example.com> or <CCC@example.com> will end up at that users mailbox; but where does <AAA@example.com> come into play?
So the questions I have are
- Is there any specific purpose where the
mailattribute is used? - Does it ever make sense to have different
mail:andproxyAddresses:SMTPentries as above? - If the answer to the preceding question is "no": What am I doing wrong that causes
mail:andproxyAddresses:SMTPto be no longer kept in sync?
Best Answer
As you said, the proxy address attribute can contain multiple values whereas the mail address contains only a single value. This can be useful when configuring multiple addresses for a single account. See the below config:
In this instance, the first attribute "SMTP:aaa@example.com", being uppercase, defines the user's primary email address. And will also be the default address for Outlook and Outlook online.
There are 3 attributes that need to be configured to ensure Accounts are synced properly between your on-premise domain controller and AzureAD/Exchange Online. These are mail, mailNick and proxyAddress.
An example of a working configuration would be as follows: