Objects could not be deleted in objdb / database errors

2000-06-16 09:35:08
Subject: Objects could not be deleted in objdb / database errors
From: "Michael Seibold" <Michael.Seibold AT Gek DOT de>
To: nv-l AT lists.tivoli DOT com
Date: Fri, 16 Jun 2000 15:35:08 +0200
This is a MIME message. If you are reading this text, you may want to 
consider changing to a mail reader or gateway that understands how to 
properly handle MIME multipart messages.
Netview V5.1.2 / AIX 4.3.1

Coming back from holidays I found some new Objects in Netview which were not 
discovered correctly because no snmp communities were defined in Netview, so I 
decided to delete them and let them be discovered again.

Problem: after deleting an object (mainframe) from the map (only the 
default-map exists, no client maps) the object remains in the object database.

All daemons are running obviously normal.

Verified with ovtopodump and ovmapdump: no trace of the deleted Object 
(ip-addresses, object-name).
ovobjprint shows the object as still there.
ovmapcount shows Exists count current = 0   correct = 0, Managed count current 
= 0, correct = 0.

I tried almost all options of ovtopofix, ovmapcount, ovwdbdmap, no result.
Tried all options with smitty nv6000 to resolve database inconsistencies and 
doing database compression, no result.

I can add the object again with Edit -> Add -> Object, getting the warning that 
it already exists and only a new symbol will be created. Deleting it again has 
the same result as before, it remains in ovwdb.

ovtopofix -r is of no help, as the objects are only in the object database and 
not in topology db.

The log files found in /usr/OV/log show no errors.

There is a lot of free disk space.

Only some hint's in nettl.log (see attached logfile), beginning after ovstop 
nvsecd and with the starting of /etc/netnmrc showing that there must be a 
problem with ovwdb. One of the messages is "Cache update failure in 
ovwDbSetFieldValue". We have about 8.000 objects in the db, according to 
mapcount. ovwdb cache size was set to 10.000 using smit.

Any ideas bevore I delete the databases and start all over??


Michael Seibold

Attachment: nettl.log3
Description: Text document

<Prev in Thread] Current Thread [Next in Thread>