Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[Samba] error during DRS repl ADD: No rDN found in replPropertyMetaData

201 views
Skip to first unread message

ash-...@comtek.co.uk

unread,
May 19, 2016, 9:10:03 AM5/19/16
to
The system described by
https://lists.samba.org/archive/samba/2016-May/199829.html (Invalid data
for index DN=@INDEX:OBJECTCLASS:DNSNODE) now appears to perform DNS
updates correctly, all systems are 4.2.10-Debian, and we've been able to
add a user and a new DC. (Thanks for the help!)

Synchronisation between v-ward (the new local DC), and empire isn't
entirely working, though.

> root@v-ward:/home/abc# /usr/bin/samba-tool drs replicate
v-ward.chester-dc.example.com empire.chester-dc.example.com
DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com
> ERROR(<class 'samba.drs_utils.drsException'>): DsReplicaSync failed -
drsException: DsReplicaSync failed (58, 'WERR_BAD_NET_RESP')
> File "/usr/lib/python2.7/dist-packages/samba/netcmd/drs.py", line
345, in run
> drs_utils.sendDsReplicaSync(self.drsuapi, self.drsuapi_handle,
source_dsa_guid, NC, req_options)
> File "/usr/lib/python2.7/dist-packages/samba/drs_utils.py", line
83, in sendDsReplicaSync
> raise drsException("DsReplicaSync failed %s" % estr)

Looking in the log file, I see:

> [2016/05/19 13:41:52.219968, 0]
../source4/dsdb/repl/replicated_objects.c:783(dsdb_replicated_objects_commit)
> Failed to apply records: replmd_replicated_apply_add: error during
DRS repl ADD: No rDN found in replPropertyMetaData for
DC=DEEL032,DC=chester-dc.example.com,CN=MicrosoftDNS,DC=DomainDnsZones,DC=chester-dc,example!
> : Constraint violation
> [2016/05/19 13:41:52.223745, 0]
../source4/dsdb/repl/drepl_out_helpers.c:770(dreplsrv_op_pull_source_apply_changes_trigger)
> Failed to commit objects:
WERR_GENERAL_FAILURE/NT_STATUS_INVALID_NETWORK_RESPONSE

I've found a similar case (
https://lists.samba.org/archive/samba/2014-September/185225.html ), but
it doesn't seem like there was a good resolution.

We are considering simply deleting the
DC=DEEL032,DC=chester-dc.example.com,CN=MicrosoftDNS,DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com
object. Would this be unwise?

I don't know if this is relevant ( Andrew Bartlett suggested "It is also
harmless, so I wouldn't worry too much until we can fix up
dbcheck" in a similar case -
https://lists.samba.org/archive/samba/2014-October/186439.html ), but we
also the the following when running dbcheck --cross-ncs

> 0 root@empire:/home/abc[0] samba-tool dbcheck --cross-ncs --fix
> Checking 18686 objects
> ERROR: wrong dn[DC=DEELR013,CN=Deleted
Objects,DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com] dc='DEELR013'
name='DEELR013\nDEL:1fa8058d-c987-4518-958d-10352c93c28a'
new_dn[DC=DEELR013\0ADEL:1fa8058d-c987-4518-958d-10352c93c28a,CN=Deleted
Objects,DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com]
> Rename DC=DEELR013,CN=Deleted
Objects,DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com to
DC=DEELR013\0ADEL:1fa8058d-c987-4518-958d-10352c93c28a,CN=Deleted
Objects,DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com? [y/N/all/none] y
> Failed to rename object DC=DEELR013,CN=Deleted
Objects,DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com into
DC=DEELR013\0ADEL:1fa8058d-c987-4518-958d-10352c93c28a,CN=Deleted
Objects,DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com : (64,
'objectclass: structural objectClass dnsNode is not a valid child class
for CN=Deleted Objects,DC=DomainDnsZones,DC=chester-dc,DC=example,DC=com')
> Checked 18686 objects (1 errors)

Can anybody recommend our next course of action?

Thanks

--
To unsubscribe from this list go to the following URL and read the
instructions: https://lists.samba.org/mailman/options/samba

mathias dufresne

unread,
May 23, 2016, 8:30:03 AM5/23/16
to
An awful response: change tombstoneLifetime : )

When deleting objects they go to recycle bin then to deleted objects then
are deleted. This if you have some recycle bin working which is not the
case by default I reckon.

tombstoneLifetime is the number of days AD has to keep deleted objects
before the real deletion. If you use the recycle bin this very same
tombstoneLifetime is also used to define how long objects will stay into
the recycle bin. There is a parameter named "msDS-deletedObjectLifetime"
which is meant to define a different delay to keep data into recycle bin
which is not defined by default I think (samba does not activate by default
recycle bin so no need of that parameter).

So tombstoneLifetime = 1 and your deleted objects should be fully removed
in few days (one if my understanding is correct, but I bet on 2).

ash-...@comtek.co.uk

unread,
May 23, 2016, 8:50:03 AM5/23/16
to
On 23/05/16 13:26, mathias dufresne wrote:
> An awful response: change tombstoneLifetime : )
>
> When deleting objects they go to recycle bin then to deleted objects then
> are deleted. This if you have some recycle bin working which is not the
> case by default I reckon.
>
> tombstoneLifetime is the number of days AD has to keep deleted objects
> before the real deletion. If you use the recycle bin this very same
> tombstoneLifetime is also used to define how long objects will stay into
> the recycle bin. There is a parameter named "msDS-deletedObjectLifetime"
> which is meant to define a different delay to keep data into recycle bin
> which is not defined by default I think (samba does not activate by default
> recycle bin so no need of that parameter).
>
> So tombstoneLifetime = 1 and your deleted objects should be fully removed
> in few days (one if my understanding is correct, but I bet on 2).
>
That seems like a reasonable solution to delete the old deleted objects
(which I don't think we need anyway), Andrew Bartlett was suggesting in
https://lists.samba.org/archive/samba/2014-October/186439.html that the
error harmless, though.

Is the Deleted Objects error even relevant to the DsReplicaSync failed
(58, 'WERR_BAD_NET_RESP') ?

I'm happy to let the deleted objects expire naturally if we can just
make replication work

Andrew Bartlett

unread,
May 27, 2016, 6:10:04 PM5/27/16
to
Please file a bug on this one. I have a set of patches for this, and it is always good to have a real-world user bug to attach to it.

As mentioned previously, it is harmless, but now that we can get it right, we should. The patches may not be backported, but it will help tracking if we do.

Thanks!

Andrew Bartlett

--
Andrew Bartlett http://samba.org/~abartlet/
Authentication Developer, Samba Team http://samba.org
Samba Developer, Catalyst IT http://catalyst.net.nz/services/samba
0 new messages