AD Sites

[MStrider]

Hi

We have an infrastructure consisting of 4 domain controllers, 2 on each
site, which are connected by a 256k leased line.

I have recently had some replication problems, and would like some views on
the best replication methods.

Below is the current configuration:

Site A has DC1 and DC2
Site B has DC3 and DC4

Site A

DC1 - Replicates from DC3 twice per hour using Transport: IP
DC1 - Replicates from DC2 twice per hour using Transport: RPC
DC2 - Replicates from DC1 twice per hour using Transport: RPC

Site B

DC3 - Replicates from DC1 twice per hour using Transport: IP
DC3 - Replicates from DC4 twice per hour using Transport: RPC
DC4 - Replicates from DC3 twice per hour using Transport: RPC

We have approx 40 workstations and 60 users.

Any suggestions would be gratefully received.

Also, and pointers on which transprt to use for which scenario would be
cool.

Thanks

Martin

 

[Simon Geary]

Replication uses one of two transport mechanisms, RPC over IP or SMTP. So
when you list your transports as either IP or RPC those two things actually
mean the same thing. All you replication will be done using RPC over IP.
SMTP transport can only be used between DC's in different domains. That's
just a terminology thing, you don't have a problem here. Use RPC over IP for
everything which is what you are doing anyway.

Your intra-site replication times are a little mean. Twice per hour for
servers in the same site is a little low. Unless you have a very slow LAN I
would look at bringing this down to something like every 5 minutes.

What replication problems are you actually having?

 

[MStrider]

event log error as below:

Event Type: Warning
Event Source: NtFrs
Event Category: None
Event ID: 13508
Date: 02/06/2004
Time: 12:33:11
User: N/A
Computer: DC4
Description:
The File Replication Service is having trouble enabling replication from DC3
to DC4 for c:\winnt\sysvol\domain using the DNS name DC3.here.com. FRS will
keep retrying.
Following are some of the reasons you would see this warning.

[1] FRS can not correctly resolve the DNS name DC3.here.com from this
computer.
[2] FRS is not running on DC3.here.com.
[3] The topology information in the Active Directory for this replica has
not yet replicated to all the Domain Controllers.

This event log message will appear once per connection, After the problem
is fixed you will see another event log message indicating that the
connection has been established.
Data:
0000: 00 00 00 00 ....

 

[Simon Geary]

Is the 13508 error followed in the logs by a 13509? If it does, this means
the problem sorted itself out and you have nothing to worry about. If it
doesn't, or if there are a lot of 13508\13509 combinations, then do some
basic troubleshooting.

Check DNS settings and DNS name resolution from each server (most likely
cause of problem)
Check FRS service is running
Run dcdiag /v and netdiag /v on DC3 and DC4 to check for any obvious errors.
Run the following command on both DC's: ntfrsutl version <FQDN of remote
domain controller>

Are any other DC's getting the same errors or is it specific to DC3 and DC4?

event log error as below:

Event Type: Warning
Event Source: NtFrs
Event Category: None
Event ID: 13508
Date: 02/06/2004
Time: 12:33:11
User: N/A
Computer: DC4
Description:
The File Replication Service is having trouble enabling replication from DC3
to DC4 for c:\winnt\sysvol\domain using the DNS name DC3.here.com. FRS will
keep retrying.
Following are some of the reasons you would see this warning.

[1] FRS can not correctly resolve the DNS name DC3.here.com from this
computer.
[2] FRS is not running on DC3.here.com.
[3] The topology information in the Active Directory for this replica has
not yet replicated to all the Domain Controllers.

This event log message will appear once per connection, After the problem
is fixed you will see another event log message indicating that the
connection has been established.
Data:
0000: 00 00 00 00 ....

Replication uses one of two transport mechanisms, RPC over IP or SMTP. So
when you list your transports as either IP or RPC those two things actually
mean the same thing. All you replication will be done using RPC over IP.
SMTP transport can only be used between DC's in different domains. That's
just a terminology thing, you don't have a problem here. Use RPC over IP for
everything which is what you are doing anyway.

Your intra-site replication times are a little mean. Twice per hour for
servers in the same site is a little low. Unless you have a very slow

LAN

I
would look at bringing this down to something like every 5 minutes.

What replication problems are you actually having?

views
on

 

[MStrider]

just dc3 and 4

I will try your suggestions and let you know what happens

no 13509 showing in the log

cheers

Is the 13508 error followed in the logs by a 13509? If it does, this means
the problem sorted itself out and you have nothing to worry about. If it
doesn't, or if there are a lot of 13508\13509 combinations, then do some
basic troubleshooting.

Check DNS settings and DNS name resolution from each server (most likely
cause of problem)
Check FRS service is running
Run dcdiag /v and netdiag /v on DC3 and DC4 to check for any obvious errors.
Run the following command on both DC's: ntfrsutl version <FQDN of remote
domain controller>

Are any other DC's getting the same errors or is it specific to DC3 and DC4?

event log error as below:

Event Type: Warning
Event Source: NtFrs
Event Category: None
Event ID: 13508
Date: 02/06/2004
Time: 12:33:11
User: N/A
Computer: DC4
Description:
The File Replication Service is having trouble enabling replication from DC3
to DC4 for c:\winnt\sysvol\domain using the DNS name DC3.here.com. FRS will
keep retrying.
Following are some of the reasons you would see this warning.

[1] FRS can not correctly resolve the DNS name DC3.here.com from this
computer.
[2] FRS is not running on DC3.here.com.
[3] The topology information in the Active Directory for this replica has
not yet replicated to all the Domain Controllers.

This event log message will appear once per connection, After the problem
is fixed you will see another event log message indicating that the
connection has been established.
Data:
0000: 00 00 00 00 ....

Replication uses one of two transport mechanisms, RPC over IP or SMTP. So
when you list your transports as either IP or RPC those two things actually
mean the same thing. All you replication will be done using RPC over IP.
SMTP transport can only be used between DC's in different domains. That's
just a terminology thing, you don't have a problem here. Use RPC over

IP
for

everything which is what you are doing anyway.

Your intra-site replication times are a little mean. Twice per hour for
servers in the same site is a little low. Unless you have a very slow

LAN

I
would look at bringing this down to something like every 5 minutes.

What replication problems are you actually having?

Hi

We have an infrastructure consisting of 4 domain controllers, 2 on each
site, which are connected by a 256k leased line.

I have recently had some replication problems, and would like some views
on
the best replication methods.

Below is the current configuration:

Site A has DC1 and DC2
Site B has DC3 and DC4

Site A

DC1 - Replicates from DC3 twice per hour using Transport: IP
DC1 - Replicates from DC2 twice per hour using Transport: RPC
DC2 - Replicates from DC1 twice per hour using Transport: RPC

Site B

DC3 - Replicates from DC1 twice per hour using Transport: IP
DC3 - Replicates from DC4 twice per hour using Transport: RPC
DC4 - Replicates from DC3 twice per hour using Transport: RPC

We have approx 40 workstations and 60 users.

Any suggestions would be gratefully received.

Also, and pointers on which transprt to use for which scenario would be
cool.

Thanks

Martin