Transport Services Flashcards
What Transport Services exist on an Exchange 2016 service?
- Front End Transport Service
- Transport Service
• Mailbox Transport Service
which consists of:
○ Mailbox Transport Submission Service
○ Mailbox Transport Delivery Service
What is the Front End Transport service?
- Stateless proxy for inbound, external, SMTP traffic
* Optionally, a stateless proxy for outbound traffic
What is the Transport Service?
- handles all SMTP mail flow for the organization
- it performs categorization (determining how to route an e-mail message)
- performs content inspection
- can queue messages
What are the Mailbox Transport services?
• Consists of:
○ Mailbox Transport Submission Service
○ Mailbox Transport Delivery Service
• The services that communicate directly with mailbox databases for both inbound and outbound email
Which transport services can queue messages?
Only the Transport Service.
The Front End Transport and Mailbox Transport services cannot.
Which transport services can perform content inspection?
Only the Transport Service
The Front End Transport and Mailbox Transport services cannot.
What can the Front End Transport service communicate with?
Only the Transport Service, on either the same mailbox server or on another mailbox server.
What can the Transport service communicate with?
- Front End Transport Service (on same or other Mailbox servers)
- Transport Service (on other Mailbox servers)
- Mailbox Transport services (on same or other Mailbox servers)
- Edge Transport servers or external SMTP servers for outbound mail
What can the Mailbox Transport services communicate with?
- Mailbox databases
* the Transport Service (on same or other Mailbox servers)
What is the Transport Pipeline?
A collection of services, connections, components, and queues that work together to route all messages to the categorizer in the Transport service on an Exchange 2016 Mailbox server inside the organization.
What is a Routing Destination?
The final destination that Exchange has determined for a message, based on the recipient information.
What are possible Routing Destinations?
- a Mailbox database (for internal recipients)
- a Connector (for external recipients)
• a Distribution group expansion server
– if one has been specified for that group. (otherwise, any mailbox server can perform the distribution group expansion)
– When a DG is expanded, it may result in duplication of message for sending to different routing destinations
What might cause an e-mail message to be duplicated and sent to different routing destinations?
This can happen, for example, if a distribution group is made up of both internal mailboxes and external contacts.
The message would be duplicated, with one having a routing destination of a mailbox database for the internal recipients, and one with a routing destination of a Send Connector for the external recipients.
What is a Delivery Group?
Each Routing Destination has one or more mailbox servers that are responsible for delivering messages to their destination.
Those servers are referred to as a Delivery Group.
What are possible Delivery Groups?
• a DAG
– The DAG members are the Delivery Group for any mailboxes that are hosted within that DAG
• Mailbox server(s) within an AD site
– If a mailbox is not hosted in a DAG, then any of the mailbox servers in the same site as the mailbox itself are used as the Delivery Group.
• Source Mailbox servers for a connector
– if a message needs to be sent over a Send Connector, then the source servers for that connector are the Delivery Group.
• an Active Directory site.
– if a hub site has been configured, and it is the least-cost route to the destination, then the mailbox servers in that site are a Delivery Group.
How are messages with multiple recipients routed?
For distribution groups:
– The nearest Transport service is used, and the group is expanded to determine routing destinations.
For multiple recipients:
– The first 20 recipients are used to determine the best Transport service to use. Then, that transport service decides what to do from there.
How does Exchange route email messages?
The Categorizer calculates the least-cost route to the delivery group for that recipient.
What is least-cost routing?
The Active Directory replication topology configures AD Site links with a “cost value.”
The lower the “cost,” the better the speed or bandwidth of the WAN connection between the two sites.
Exchange looks at these same IP Site Cost Links when calculating the least-cost route between two servers in the same organization.
What does “queuing to the point of failure” refer to?
If a site has an outage, then Exchange will try to send messages destined for that site to the closest site and queued until the destination site is online again.
This will only happen if the target mailbox is NOT hosted in a DAG, or if all DAG sites are down.
What will happen if a destination mailbox is hosted on a server that is having an outage?
If the mailbox is hosted in a DAG and the DAG member hosting the mailbox is down, then the sending server will try to deliver the message to:
(in order of preference)
- the closets DAG member,
- any DAG member,
- queuing to the point of failure if no DAG members are available
If the mailbox is not hosted by a DAG, then the sending server will simply queue to the point of failure.
What is an “Exchange Cost”?
A parameter that can be set on an AD Site Link, which will be used by Exchange to calculate route cost instead of the IP Site Link cost.
Exchange Cost value will override the IP Site Link cost value.
This is useful in situations where the AD topology doesn’t match how you want your Exchange routing to behave.
How do you configure an Exchange Cost?
Exchange Cost values are assigned to AD Sites Links.
Set-ADSiteLink
“SiteA-SiteB”
-ExchangeCost 5
How will a “Hub Site” influence Exchange routing between sites?
When an AD Site has been configured as a Hub Site,
and that site exists on the Least Cost Routing path between two other sites,
then a message will be sent to an exchange server in the Hub site, instead of directly to the destination server.
What is the PowerShell command to create a new Hub Site?
To set an AD Site as a Hub Site:
Set-ADSite
“Name of Site”
-HubSiteEnabled:$true
What is Shadow Redundancy?
A feature that makes a copy of each e-mail message onto another Mailbox server, to protect the message from failure of a server that is processing the message.
The goal is to improve availability and resilience of e-mail in transit.
How does Shadow Redundancy work?
When Front End Transport sends a message to a Transport service, the message will also be copied to another Transport service before the first Transport service acknowledges receipt.
When the Shadow copy has been successfully created, the first mailbox server sends back the acknowledgement to the sending device or server.
What is a Shadow Queue?
Part of Shadow Redundancy.
The Shadow Queue is where Shadow copies of e-mail messages are stored on the Shadow server for that message.
How is a server selected to copy a message onto for Shadow Redundancy?
If the receiving server is NOT a DAG member:
- a mailbox server in the local directory site is used as the Shadow server.
If the receiving server is a DAG member:
- another DAG member is used as the Shadow server.
If the DAG spans multiple Active Directory sites:
- Shadow Redundancy will try to pick a DAG member in a different site, so you have site resilience for the message.
How do you configure Shadow Redundancy?
- It is enabled by default.
- Disabling it is not recommended.
- It is enabled and configured organization-wide. Cannot be configured per-server.
- You can set it to either accept or reject messages if shadow redundancy fails.
What happens if a Shadow copy fails to be created?
You have the option to set whether Messages are accepted or rejected if a shadow copy fails to be created.
If you have multiple servers in each AD site, or if servers are members of a DAG, you can set messages to be rejected if the Shadow copy fails to be created, if you want to prioritize resiliency over continued delivery.
In single-server sites/environments, Shadow copies will always fail to be created, so they must be accepted in this setting.
What is the PowerShell command to Reject a message when a shadow copy fails to be created?
Set-TransportConfig
-RejectMessageOnShadowFailure:$true
Part of Transport, this feature will protect message resiliency while the message is in transit.
What is:
Shadow Redundancy
Part of Transport, this feature will protect message resiliency after the message has been delivered.
What is:
Safety Net
What is Safety Net?
Safety Net stores copies of successfully delivered messages, so they can be resubmitted into the transport pipeline if necessary.
When might Safety Net be used to resubmit a message?
- After a lossy failover of a database in a DAG.
* During activation of a lagged database copy.
How long will Safety Net hold copies of delivered messages for?
The default hold time is 2 days, but it is configurable at the organization level.
How should Safety Net be configured if you are using Lagged Database Copies?
The hold time of Safety Net should be set to match or exceed the longest lag time on your lagged database copies.
Powershell command to configure Safety Net Hold Time?
Set-TransportConfig
- SafetyNetHoldTime
10. 00:00:00
(format is dd.hh:mm:ss)
How is Safety Net itself made redundant?
As long as Shadow Redundancy hasn’t been disabled, then Safety Net is made redundant by Shadow Safety Net.
The the Primary Safety Net is unavailable for resubmission requests, Shadow Safety Net can service the requests.
At what point are messages moved into the Safety Net?
After a message has been successfully delivered to the recipient, the primary copy of the message is moved to Safety Net, and the shadow copy of the message is moved to the Shadow Safety Net (as long as Shadow Redundancy hasn’t been disabled).
What database is Safety Net part of?
Safety Net is part of the Transport Database.
What is a Receive Connector?
Receive Connectors are used to receive e-mail from clients, servers, or other transport services.
What are the default Receive Connectors that come configured when you install Exchange?
- There are 5 default receive connectors created by Exchange.
- Each will have the server name in the Connector’s name:
• Front End Transport Service Receive Connectors:
○ Default Frontend
○ Client FrontEnd
○ Outbound Proxy Frontend
• Transport Service Receive Connectors:
○ Default
○ Client Proxy
What is this Receive Connector, and what port does it use?
Default Frontend
- Bound to the Front End Transport Service
- Listens on Port: TCP 25
- Entry point for e-mail from non-Exchange systems into the organization.
What is this Receive Connector, and what port does it use?
Client FrontEnd
- Bound to the Front End Transport Service
- Listens on Port: 587
- Requires client authentication
- Used for secure, authenticated SMTP (protected by TLS)
- Used, for example, by POP and IMAP clients, and by internal devices or apps that are configured to use authentication.
What is this Receive Connector, and what port does it use?
Outbound Proxy Frontend
- Bound to the Front End Transport Service
- Listens on Port: 717
- Only used if a send connector is configured to proxy outbound e-mail via Front End Transport service.
- By default, send connectors are NOT configured in this way, so this Receive Connector is almost never used in the real world.
What is this Receive Connector, and what port does it use?
Default
- Bound to the Transport Service
- Listens on Port: 2525
- Accepts connections from other transport services on the same or different servers.
- That would include connections:
- that have been proxied from the Default Frontend connector,
- from Edge Transport servers,
- from Mailbox Transport services,
- or from other transport services on other servers.
What is this Receive Connector, and what port does it use?
Client Proxy
- Bound to the Transport Service
- Listens on Port: 465
- Accepts client connections proxied by Frontend services
For what reasons should you modify any of the Default Receive Connectors?
Generally, you should not modify them, except to do the following:
- Change message size limits
- Change rate limits
- Configure a TLS certificate to use on a particular connector
What is a Relay Connector?
A particular type of Receive Connector.
If a device or application needs to send e-mail messages, (such as for notifications, alerts, scan-to-email, etc.), Exchange can perform that service.
A connector built for this purpose is commonly called a Relay Connector.
What is the difference between Internal Relay and External Relay, and how are they configured?
Internal Relay works without any special configuration required.
The default Frontend Receive Connector is already configured to accept SMTP connections from anywhere, and the Exchange server will allow an anonymous sender to send e-mail to any internal recipients (since that is also how internet e-mail is treated).
But, if a sender tries to send e-mail to an external recipient, they must either use authentication (which will go through the Client FrontEnd Receive Connector), or an External Relay Connector must be configured to allow anonymous senders from specific IP addresses.
If more than one Front End Receive Connector is configured to listen on Port 25 for internal senders, how does an Exchange server know which connector to use?
The most specific match wins.
The Default Frontend Receive Connector is configured for all IP addresses. So, it is the least-specific configuration.
If a non-default Receive Connector has been configured with a specific set or range of IP addresses, then Exchange will use the connector that most specifically matches the source’s IP address.
What is an “Accepted Domain”?
Accepted DOmains are SMTP namespaces that an Exchange organization is able to send and recieve e-mail for.
A domain must be added as an accepted domain before it can be used in e-mail addresses assigned to recipients.
What types of Accepted Domains are there?
There are three types of Accepted Domains:
- Authoritative
- Internal Relay
- External Relay
What is this type of Accepted Domain:
Authoritative
- A domain that the Exchange organization is solely responsible for.
- Exchange will only deliver e-mails to local recipients (that are within the local Exchange organization)
- All e-mail for unknown recipients is rejected.
What is this type of Accepted Domain:
Internal Relay
- A domain that the Exchange organization shares responsibility for.
- Recipients can be within the local Exchange organization, or can be located in an external mail system (whether another Exchange organization, or a third-party mail system).
- Exchange will first look for a local recipient, and if none are found, it will relay the mail off to another mail system.
- (A Send Connector handles the routing of the message to the other mail system.)
What is this type of Accepted Domain:
External Relay
- A domain for which Exchange has no local recipients, but can relay the messages to another mail system.
- Only recipients that are located in an external mail system.
- (A Send Connector handles the routing of the message to the other mail system.)
How can you assign e-mail addresses using sub-domains to recipients?
You can either configure the sub-domain as an Accepted Domain,
(for example, sales.company.com)
or you can add an Accepted Domain using a Wildcard.
(for example, *.company.com)
which will allow ANY subdomain to be used in e-mail addresses assigned to recipients.
Using e-mail address policies, how can you assign e-mail addresses using sub-domains to recipients?
If your Accepted Domain was configured with a wildcard, you cannot use its sub-domains in E-mail Address Policies.
Instead, you would need to add the specific sub-domains as their own Accepted Domain.
What is a “primary e-mail address”?
A recipient can have more than one e-mail address, but the primary e-mail address is the address that the e-mails they send will appear to be FROM.
What is this:
EAP
E-mail Address Policy/Policies
How can you use e-mail address policies to remove unwanted e-mail addresses?
You cannot.
EAPs can be used to add e-mail addresses, but they never remove e-mail addresses.
In an EAP address format, what does this variable mean?
%g
Given name
In an EAP address format, what does this variable mean?
%i
Middle initial
In an EAP address format, what does this variable mean?
%s
Surname
In an EAP address format, what does this variable mean?
%d
Display name
In an EAP address format, what does this variable mean?
%m
Exchange alias
In an EAP address format, what does this variable mean?
%xs
Use first X letters of surname
In an EAP address format, what does this variable mean?
%xg
Use first X letters of given name
If you don’t specify an e-mail address format in an EAP, what will the resulting e-mail address be?
It will default to use the user’s Exchange Alias. For example,
alias@domain.com