ADSM-L

NetWare DS backup problems

1996-10-17 13:57:27
Subject: NetWare DS backup problems
From: Tom Anstey <tom.anstey AT COMPUTING-SERVICES.OXFORD.AC DOT UK>
Date: Thu, 17 Oct 1996 17:57:27 GMT0BST
We have been having problems with the backup of the NDS to ADSM.

The server is running NW4.10, with the latest patches applied, and the
client 2.1.0.5.

Before we applied PTF 5 it looks as though the backup of Directory
Services may have been failing altogether.

ADSM seems to be trying to backup non-existant objects in the NDS
which then produce failures.

If anyone has had similar experiences and managed to find a solution,
then we would be very grateful for your assistance.

The message below is part of an email dialogue that has been taking
place "in-house" regarding the situation:


Tom
>
> > 1) Has NSH ever completed it's backup successfully since PTF 5 was
> > installed?
> >    If possible, can you get the evidence for that from the
> >    dsmsched.log?
> >
>
> I think PTF 5 was installed 13th Sept judging from the dates of
> DSMC.NLM
>
> There have been partially successful backups after that date shown
> by the DSMSCHED.LOG which did not exhibit the problem with the
> 'Server detected and error' message although there are odd failures
> whihc don't seem to correspond with existing Netware NDS tree
> objects. Its almost as if, in programming terms, the string that is
> build to indicate what NDS object distinguished name to backup is
> incorrect and therefore indicates an object that doesn't exist.
>
> i.e.
> 03-10-1996 01:59:23 Directory-->         2,048,000
> NSF_OUCS\DIRECTORY:/.O=Oxford/.OU=WYCL.O=Oxford  Sent
>
> does exist as the organisational unit (NDS terms) .OU=WYCL.O=Oxford
> but the following don't exist
>
> 03-10-1996 01:59:27 ANS4228E Sending of object
> 'NSF_OUCS\DIRECTORY:/.O=Oxford/CH.O=Oxford' failed
> 03-10-1996 01:59:27 ANS4089E File not found during Backup, Archive
> or Migrate processing 03-10-1996 02:07:26 Directory-->
> 2,048,000 NSF_OUCS\DIRECTORY:/.O=Oxford/JR.O=Oxford  ** Unsuccessful
> ** 03-10-1996 02:07:26 ANS4228E Sending of object
> 'NSF_OUCS\DIRECTORY:/.O=Oxford/JR.O=Oxford' failed 03-10-1996
> 02:07:26 ANS4089E File not found during Backup, Archive or Migrate
> processing 03-10-1996 02:07:26 Directory-->         2,048,000
> NSF_OUCS\DIRECTORY:/.O=Oxford/JR.O=Oxford  ** Unsuccessful **
> 03-10-1996 02:07:26 ANS4228E Sending of object
> 'NSF_OUCS\DIRECTORY:/.O=Oxford/JR.O=Oxford' failed 03-10-1996
> 02:07:26 ANS4089E File not found during Backup, Archive or Migrate
> processing 03-10-1996 02:07:26 Directory-->         2,048,000
> NSF_OUCS\DIRECTORY:/.O=Oxford/RI.O=Oxford  ** Unsuccessful **
> 03-10-1996 02:07:26 ANS4228E Sending of object
> 'NSF_OUCS\DIRECTORY:/.O=Oxford/RI.O=Oxford' failed 03-10-1996
> 02:07:26 ANS4089E File not found during Backup, Archive or Migrate
> processing
>
>
> >
> > 2) Can you please have a look at the protections for the entry
> > shown in the
> >    dsmerr.log file at the time of failure, i.e.
> >
> >    NSF_OUCS\DIRECTORY: /.O=Oxford/CH.O=Oxford
>
> This object doesn't exist in the tree anyway, but all areas of the
> NDS have the same restrictions so there is no reason for one to be
> backed up but not the other.
>
> >
> >    What they are trying to ask is whether the ADSM backup process
> >    has the appropriate priv to access this entry.
> >
>
> I think the answer has to be yes, if it existed. The way that the
> log displays the the object it is backup up is not the typical way
> of representing an NDS object so it may be that I am wrong in the
> way I am interpreting the existance of an object.
>
> I could generate a text file that has the complete NDS layout if
> that is of use with respect to the existence/non-existence of
> certain objects?
>
> Hope that helps,
>
> Ian
> ----------------------
> Ian Griggs
> Micros Consultant
> Oxford University Computing Services
>
>

==============================
Tom Anstey

Oxford University Computing Services
13 Banbury Road
Oxford
OX2 6NN
<Prev in Thread] Current Thread [Next in Thread>
  • NetWare DS backup problems, Tom Anstey <=