[cifs-protocol] [REG:115041612643080 ] MS-ADTS 6.2 - Knowledge consistency checker
garming at catalyst.net.nz
Wed Apr 22 20:45:29 MDT 2015
Everything that you've mentioned, I've managed to grasp. Likewise, I've
seen most of the available documentation on RODC and KCC including the
ones you've posted.
> If the local site color is white or number of colored vertices is less than two, KCC does not need to build spanning tree for that NC.
This check for a white vertex evidently happens in CreateConnections of
ADTS 18.104.22.168.4.5. This means, as it describes, that the topology does
not need to be computed. However, if the RODC does not compute the
topology, it doesn't explain how the inbound connection objects it needs
are created when its in its own site. At least from the KCC, all the
regular DC see the RODC as invisible (Why doesn't the KCC on writable
domain controllers try to build connections from an RODC? on
Is the connection creation outside of the KCC? The only thing I know for
certain that a KCC on an RODC does is update the FRS connection to point
at the same place as the DRS connection. But the description itself
obviously has checks to imply otherwise. Notably when it is in its own
site, according to ADTS 22.214.171.124.1 RODC apparently acts as an
intersite-topology generator for itself. Following the guidelines for
"is present", every indication seems to show that all the NC replicas on
the RODC satisfy these conditions. However, it seems irrelevant since
the KCC never gets far enough to use this value - again with the RODC
being invisible to everyone but itself.
> Either the local site does not host this NC, or the spanning tree would not have any edge. In either case, we do not need to compute the topology. KCC moves to the next NC in the list.
> Recall that vertex initialization sets the color to ‘white’. The colored vertices are sites which contain a writeable or partial replica of the NC specified by CrossRef. Sites which contain at least one writeable replica are considered 'red' vertices. Sites which contain at least one partial replica but no writeable replicas are considered 'black' vertices. All other sites are considered 'white' vertices as marked during initialization, and are not colored.
> Please take a look at the following references, including the non-MS-ADTS ones, they may be useful.
> MS-ADTS 6.2.2 Overview
> . . .
> To simplify the task descriptions, the following concepts are used:
> • An NC replica that "is present" on a DC. Given NC replica r of NC n and a DC with nTDSDSA object o, r "is present" on the DC if both of the following conditions is true:
> o o!hasMasterNCs contains n or o!msDS-hasFullReplicaNCs contains n or o!hasPartialReplicaNCs contains n.
> o One of the following two conditions is true:
> o!msDS-HasInstantiatedNCs contains no value v where the dsname portion of v = n. (In this case n is in the process of being instantiated.)
> o!msDS-HasInstantiatedNCs contains a value v, where the dsname part of v = n, and the binary part of v (DWORD in big-endian byte order) is an integer such that the IT_NC_GOING bit is clear. (In this case n is instantiated, and is not in the process of being uninstantiated.)
> • An NC replica that "should be present" on a DC. Given NC replica r of NC n and a DC with nTDSDSA object o, r "should be present" on the DC if r is one of the following:
> o A writable replica of the config NC, the schema NC, or the DC's default NC on a writable DC.
> o A full read-only replica of the config NC, the schema NC, or the DC's default NC on an RODC.
> o A writable replica of an application NC for which there exists a crossRef object cr such that cr!nCName = n and cr!msDS-NC-Replica-Locations contains a reference to o.
> o A full read-only replica of an application NC for which there exists a crossRef object cr such that cr!nCName = n and cr.ms-DS-NC-RO-Replica-Locations contains a reference to o.
> o If the DC is a GC server (that is, if bit NTDSDSA_OPT_IS_GC is set in o!options), a partial replica of a domain NC n such that n ≠ the DC's default NC, and there exists a crossRef object cr such that cr!nCName = n.
> • An nTDSConnection object "implies" a tuple in the repsFrom abstract attribute of an NC replica (and a corresponding edge in an NC replica graph). An nTDSConnection object cn implies a tuple in r!repsFrom for NC replica r of NC n on the DC with nTDSDSA object t, if each of the following is true:
> o cn is a child of t.
> o cn!fromServer references an nTDSDSA object s.
> o An NC replica of n "is present" on s.
> o r "should be present" on t.
> o The NC replica on s is a full replica or r is a partial replica.
> o n is not a domain NC, or r is a partial replica, or cn!transportType has no value, or cn!transportType has an RDN of CN=IP.
> Review Bridgehead Server Load-Balancing Improvements with Windows Server 2008 RODCs
> Active Directory Replication Concepts
> RODC Frequently Asked Questions
> -----Original Message-----
> From: Edgar Olougouna
> Sent: Thursday, April 16, 2015 10:23 AM
> To: Garming Sam
> Cc: cifs-protocol at samba.org; MSSolve Case Email
> Subject: [REG:115041612643080 ] MS-ADTS 6.2 - Knowledge consistency checker
> [case number in subject]
> [bcc dochelp; cc casemail]
> Hello Garming,
> Glad to hear from you. The case number 115041612643080 has been created for this inquiry. I will investigate and assist you on this.
> -----Original Message-----
> From: Garming Sam [mailto:garming at catalyst.net.nz]
> Sent: Wednesday, April 15, 2015 11:25 PM
> To: Interoperability Documentation Help
> Cc: cifs-protocol at samba.org
> Subject: MS-ADTS 6.2 - Knowledge consistency checker
> In attempting to implement the inter-site topology generation algorithm, it appears that cases around RODC are lacking clarity in the documentation.
> For instance described in 126.96.36.199.3 - Site Graph Concepts, essentially Red describes sites with writable replicas, Black describes sites with only partial read-only and White with neither writable nor partial read-only replicas. But there is no mention of full read-only replicas.
> The only assumption that could be made here is that they are white, however, there are two problems:
> 1. In 188.8.131.52.4.3 - Site Graph Construction under ColorVertices, Red describes sites with one or more DCs with full replicas - not writable.
> This would make full read-only replica Red.
> 2. Being marked as White, during intersite topology generation, sites with only an RODC do not appear in the spanning-tree computation.
> Therefore, they will never create any connections.
> Being marked as Red however, there does not appear to be any restrictions in the graph construction on who may replicate from it (like a Black node for instance). Is there something here I am misunderstanding?
> What color is a site containing only an RODC intended to be? And are there any additional undocumented special cases for RODC?
> Garming Sam
More information about the cifs-protocol