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 Historical content remains viewable.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

Schinst Error -613

0 views
Skip to first unread message

wayne...@greenhillsschool.org

unread,
May 6, 2003, 11:22:11 AM5/6/03
to
I have a newly set up Netware 6 server with SP3 applied. I'm trying to
follow the instructions for setting up NFS access to the server, but I'm
having trouble with the SCHINST portion of the setup. It fails when it
tries to create the object .CN-NFAUUser.O=Greenhills with the error code
-613. Can anybody tell me what this error code means? What do I need to do
to get this working? Thank you for your help. Here is what the SCHINST.LOG
file says:

Info:10:5-6-2003 11:01:48 am : Modifying NDS schema from file :
sys:/system/schema/nisupgd.sch.

Info:147:5-6-2003 11:01:48 am :

Login was successful.

Error:144:5-6-2003 11:01:48 am : Could not add the object :
.CN=NFAUUser.O=Greenhills
You need to check the NIS Administrator Context Error Code : -613.

Wayne Eaker
Systems Administrator
Greenhills School

Brad Doster

unread,
May 6, 2003, 11:52:28 PM5/6/03
to
In article <DqQta.1691$ez4...@prv-forum2.provo.novell.com>, wrote:
> I'm trying to
> follow the instructions for setting up NFS access to the server, but I'm
> having trouble with the SCHINST portion of the setup.
>
What instructions are you following? As I recall, you should not need to
run SCHINST manually, though it has been awhile since I've done the
install...

> It fails when it
> tries to create the object .CN-NFAUUser.O=Greenhills with the error code
> -613.
>

Below is what the error code listing has to say in NetWare Remote Manager.
In that it's an NDS error, if you need further clarification on it, it's
best to ask in the NDS forums.

-613 FFFFFD9B SYNTAX VIOLATION

Source

NDS® or Novell® eDirectoryTM

Explanation

A request was made to modify an NDS or eDirectory attribute using data
that does not conform to the syntax specified within the NDS or eDirectory
schema attribute definition.

Warning: Applying all solutions mentioned in this topic could make the
problem worse if the actual cause of the problem is not known. Before
following a course of action, make sure that you understand the cause of
the error and the consequences for the actions suggested.

Possible Cause

This error occurs most often when an attribute value being added to an
object is incorrect or when the value is pointing to an object that does
not exist in the tree.

Action

Verify that the object name being added as a value is correct, and verify
that the object exists in the tree.

Possible Cause

An attempt was made to modify an NDS or eDirectory attribute whose syntax
is a distinguished name. The following are possible causes of the error:

The source server was unable to locate the distinguished name for the
NDS or eDirectory object provided.

The distinguished name for the NDS or eDirectory object provided refers
to a deleted object. Additional information on this instance of the error
can be obtained using DSTRACE on the source server with the Collisions
flag set.

The distinguished name for the NDS or eDirectory object provided is an
illegal NDS or eDirectory name.

Action

Make sure the object name does not contain illegal characters.

Possible Cause

A fault occurred in the application that is being used to manipulate the
attribute values.

Action

Contact the developer of the application.

Possible Cause

The error might also occur when a bindery-type object (usually created by
a third-party program) becomes corrupt.

Action

If the -613 errors are on these types of objects, delete all objects
created by the program, and then reinstall the program.

Possible Cause

This error is also seen during the install or upgrade to NetWare® 4.1 (or
later versions).

Action

If you are installing or upgrading to NetWare 4.1 (or later versions), run
Partition Continuity in NDS manager to view any synchronization errors on
the partition that you are installing the server into. Using NDS manager,
run Repair Local Database on the server holding the master replica of the
partition.

If a new NetWare 4.1 (or later version) server hangs during installation
while copying over the existing NDS or eDirectory objects, run DSTRACE on
the server holding the master replica of the partition that you are
installing the server into. Do this by setting the SET DSTRACE=+SYNC
command.

If an error exists after running DSTRACE, run DSREPAIR on the master.

If the problem still exists, run the SET DSTRACE=+SYNC command to find the
object returning the error and check the description field (and any other
field which accepts a string) for any unusual ASCII characters. Correct
the corruption or delete the object, and the replica should complete
copying to the new server.

bd
NSC Volunteer SysOp
www.InsightNetSolutions.net

wayne...@greenhillsschool.org

unread,
May 7, 2003, 10:03:31 AM5/7/03
to
Actually, I'm not sure where I found it now. Specifically it was on what to
do if you get the Root user not defined error, and told you how to set
things up again. But, actually, I have three servers that I installed by
the "normal" procedures of just following the install, and NFS has never
worked on any of them. I haven't needed it up to this point, so I've let it
slide. But, now I really could use it, and nothing I do seems to have any
effect. (NFA for Windows and Mac both work fine.)

Basically, no NDS objects were ever created relating to NFS, even though I
installed the NFA for Unix on all three NW6 servers. I can go into
ConsoleOne and create a user and set his UID to 0, but that doesn't seem to
make any difference. NDSILIB still gets screwed up when running NFSStart
because it says the root user is not defined. I've tried changing the
search path through NFSAdmin, and it made no difference. This happens on
all three servers.

I feel like something is missing from my directory, but I have no idea what
it is or how to create it. Any leads you could provide would be great.

Wayne Eaker
System Administrator
Greenhills School

Brad Doster

unread,
May 7, 2003, 10:43:14 PM5/7/03
to
Well, I think I would try uninstalling it, restarting the server,
reinstalling, restarting the server, then reapplying the service pack
you're using and any other patches that might be relevant, then see
where that goes.


Hmmm.... randonm thought... do the servers in question hold RW replicas
of the O=Greenhills container? And silly question... you do have S
rights to the root of the tree, correct?

wayne...@greenhillsschool.org

unread,
May 8, 2003, 8:51:00 AM5/8/03
to
Okay, I tried re-installing, as you suggested, and I'm at the same place.
Everything installed with no errors, but nothing was created in the
directory, and I still get the Root not mapped to User message when
starting NFS.

I'm pretty new to Netware, so I'm not exactly sure how to verify what
you're asking about RW replicas. I checked the tree, and the admin account
does have S rights to the root of the tree.

Thanks again for your help.

Wayne Eaker
Systems Administrator
Greenhills School

> Well, I think I would try uninstalling it, restarting the server,

Brad Doster

unread,
May 8, 2003, 9:47:39 AM5/8/03
to
In article <Uosua.3959$ez4....@prv-forum2.provo.novell.com>, wrote:
> I'm pretty new to Netware, so I'm not exactly sure how to verify what
> you're asking about RW replicas.
>
Well, you can see you replica layout with ConsoleOne, NDS Manager
(sys:public\win32\ndsmgr32.exe) or DSREPAIR.NLM, among other tools.
DSREPAIR might be the most direct in this case, though you'll need to
do this on each server in question...

LOAD DSREPAIR
Advanced Options
Replica and partition operations
Is .Greenhills listed? If yes, is its Replica type "Master" or
"Read/Write"? If yes, then we're OK on that server.

wayne...@greenhillsschool.org

unread,
May 8, 2003, 10:47:49 AM5/8/03
to
Thanks for the pointers. I checked it out on all three of my servers and
.Greenhills is not listed, but ".[Root]." is. Two of them are R/W and the
other is the Master. Is it a problem that the organization is not listed?

Brad Doster

unread,
May 8, 2003, 10:59:54 PM5/8/03
to
That's probably OK -- Greenhills may not have been partitioned off, so
it is part of the [Root] partition. BUT, that doesn't help us with
your problem, so it's not all good news.

Did you look into/try any of the items listed with the error message
info I posted?

Wayne Eaker

unread,
May 9, 2003, 1:12:24 PM5/9/03
to
I checked out that information, and also I saw a post by Matt Ross in the
native-file-access group that was helpful. He was having the same Root not
mapped problem, and you had sent him a TID about extending the schema
manually.
(http://support.novell.com/cgi-bin/search/searchtid.cgi?/10054743.htm) I
tried that, and it seemed to work fine. DSMISC.LOG showed everything either
working, or "already exists and is identical." However, when I went back to
run SCHINST -N again (as per the TID), it fails running the same schema
extension extension that I just successfully ran. I ran
SYS:\SYSTEM\SCHEMA\NISUPGD.SCH with no problems, but when it ran
sys:/system/schema/nisupgd.sch, it stopped on the following error:

Lower bound does not match existing schema attribute definition.
Failed to add schema attribute userPassword
An error occurred modifying the NDS schema for file
sys:/system/schema/nisupgd.sch.
Error Description This attribute already exists. (DSI-5.00-185)

So, I think I have things narrowed down, but I don't know what I can do
about this. Following the instructions in the TID did not lead to success,
as NFAUUser was not created by the SCHINST -N command. Does this information
I've found today give you any ideas of what I might try next?

I really appreciate all your help.

Wayne Eaker
Systems Administrator
Greenhills School

"Brad Doster" <b...@NSCSysOps.net> wrote in message
news:VA.00001d2...@nscsysops.net...

Brad Doster

unread,
May 9, 2003, 11:23:20 PM5/9/03
to
Did you do this part of the TID...

If these schema extensions fail, see the eDirectory/Directory Services
Health Check document at
http://support.novell.com/cgi-bin/search/searchtid.cgi?/10060600.htm
After verifying eDirectory/NDS health repeat the above schema extension
steps.

..and then continue from there? I'm starting to suspect NDS problems
as the source of the issue here.

Wayne Eaker

unread,
May 12, 2003, 11:44:19 AM5/12/03
to
Yes, I followed that TID, and everything checked out okay. I got no
indication of any problem. I also followed the other TID there that suggests
checking the schema for an attribute called UNIX:UID and deleting it if it
exists. It didn't, the only similar ones that exist in my schema are UID and
uidNumber.

So, I'm at a total loss now...everything reports that it's okay, but the
schema extension and creation of the NFAUUser refuses to occur.

In the meantime, I have gotten it to almost work using CIFS and mounting the
volume as smbfs volume in Linux. It works, but there is a problem reading
some files. (and a subsequent altering of their modified date to far in the
future) But, that's a problem for a different forum.

Thanks for all your help. If you know of anything else that I can try, I'd
appreciate it.
Wayne.

"Brad Doster" <b...@NSCSysOps.net> wrote in message

news:VA.00001d4...@nscsysops.net...

Brad Doster

unread,
May 13, 2003, 1:02:21 AM5/13/03
to
Sorry, but you've got me stumped on this one as well, at least for the
moment <g>.

Brad Doster

unread,
May 13, 2003, 1:17:58 AM5/13/03
to
Sorry, but you've got me stumped on this one as well, at least for the
moment <g>.

bd
NSC Volunteer SysOp
www.InsightNetSolutions.net

Wayne Eaker

unread,
May 22, 2003, 11:18:36 AM5/22/03
to
Just for people searching this group who might have the same problem as me
in the future, I want to post my "solution". Since NFS refused to install
properly on my servers, I decided to try CIFS. I tried for a solid week to
get the smbfs mount option on Linux to properly connect to the CIFS NFA on
Netware 6, but I finally had to give that up. There is some kind of problem
with the timestamps. At random intervals, the timestamps get returned as
crazy thing from the year 2031 and whatnot. This is probably due to some
problem in the smbfs support on Linux. (This server is Redhat 7.2.) Also,
there is a delay in timestamp updates that was odd.* Unfortunately, since
the program I was working on requires proper timestamps, this wasn't
acceptable.

But, then I found something on Redhat called ncpfs. I wasn't aware of it
before, but this will allow you to mount Netware volumes over IPX or IP (on
Redhat). This saved the day, and it took about 5 minutes to set up...two
rpms to download and install, one man page to read. If you're on Redhat and
being completely frustrated by Unix NFA (the "Root not mapped" error seems
common on these boards), give this a try. The two rpms you need are ncpfs
and ipxutils(even if you are going to use IP, it's a dependency of ncpfs).

Wayne Eaker
System Administrator
Greenhills School

*I'll describe it the timestamp delay problem here, because it seems like it
might be a Novell issue on this one. Here's the scenario:
1) The Netware volume is exported via CIFS and mounted on the Linux box
using smbfs. (Yeah, I know, unsupported, but as an SLA licensee, everything
I do is unsupported.)
2) My computer was 1 minute and 40 seconds ahead of the server time. The
Linux box and the Netware server are synchronized. (to within about 10
seconds)
3) I save a file to the server, and then immediately after that, check the
time as seen from the Linux box using the stat() function in Perl. That time
is a new timestamp using the server's time. (1:40 behind the time on my
desktop.)
4) After some time (I wasn't able to nail it down, but approximately 20-30
seconds), the timestamp will change on that file with no interaction from
me. If I stat() the file after waiting, the timestamp is the same timestamp,
but it's using my desktop computer's time. At some point, 1:40 was added to
the timestamp.
5) Also, if I use stat() twice on a file quickly, the second timestamp will
be the one from the desktop, no matter if I waited 20-30 seconds or not.
Stat()ing the file seems to force the update, but not until after it returns
the first value.

It's a strange thing, and I ran around in circles trying to nail it down for
quite some time, but then I found ncpfs, and now there's no reason to
bother. But, I'll put up here for the record.

"Brad Doster" <b...@NSCSysOps.net> wrote in message

news:VA.00001d5...@nscsysops.net...

Brad Doster

unread,
May 22, 2003, 10:02:13 PM5/22/03
to
Thanks for the update!

Brian Schonecker

unread,
Oct 9, 2003, 10:47:59 AM10/9/03
to
I don't know if I'm opening a can of worms but I've been fighting this very
problem for about a week now on NW6.5.

I'm virtually crippled not being able to use NFS. NFS
_none_ of the TIDs suggested worked. I'm tempted to call Novell.

B

<wayne...@greenhillsschool.org> wrote in message
news:DqQta.1691$ez4...@prv-forum2.provo.novell.com...

Brad Doster

unread,
Oct 9, 2003, 11:07:25 PM10/9/03
to
Looks like this thread and the one you replied in the NFAP forum have
fallen off of the server. I've already replied in the NFAP forum, so
lets keep the conversation there and let this one die out.
0 new messages