CiscoWorks RME Device Inventory

[mjgleeson]

I'm getting the following error when trying to delete some "Non Reponding" devices. There are about 40 devices that give this error every time I try to delete them..

Any ideas?

Are there any database management/tidyup tools I can run, or should be running periodically?

Error: The requested URL cannot be reached because Premature end of script headers: d:/program files/cscopx/cgi-bin/import/pimport/statusdevupdate.pl

Description: 500 Server Error is the HTTP status code returned by the Apache web server if it encounters an unexpected condition that prevents it from fulfilling the request.

--------------------------------------------------------------------------------

The application has run into a unforeseen error. This may be due to an incorrect installation of the product. Please refer to your product's installation guide.

--------------------------

Versions:

CiscoWorks Common Services with SP3 2.2

Resource Manager Essentials 3.5

Resource Manager Essentials Incremental Device Support 3.5

[David Stanford]

What IDU version for RME are you running? The last time I saw something like this in RME 3.5 it was due to individual device corruption within the DB.

Has anyone deleted the config files on the hd related to these devices in the inventory?

Are you seeing any 500 server errors anywhere else in the application?

[mjgleeson]

Incremental Device Update 13.0.

We are getting other errors scattered about regarding prolems reading the db, hence the query about any db maintenance we can do.

[David Stanford]

I don't think there's really any db maintenance you can do, but you can try they following to see if there is individual device corruption:

1) first login to the app as admin and then go to dbreader:

http://ciscoworks:1741/dbreader/dbreader.html

2) Enter the credentials for username, password and db

3) Select the button Get Database Tables

4) Look at the table DEV_FQN for your device and record the Device_Id

5) Then go to the table MANAGED_DEV and make sure this Device_Id is there (if not look in UNMANAGED_DEV)

6) Now go back to the main dbreader page and enter the following statement:

delete from DEV_FQN where Device_Id=? (you would sub ? with the actual id)

Once this is done, try going back to the RME GUI and adding the device manually, then try viewing its status and see if you still get the errors

[mjgleeson]

Sounds useful. When I check the status of the devices, it says "device no longer exists in database".

I have the user/pass, obviously, but what will the db name be?

[David Stanford]

the db name is rme (case sensitive)

[mjgleeson]

Thanks for the advice...

Is there any easy way I can get rid of the database and start from scratch?

Is it just a case of renaming/deleting files, or does it have to be done through the dbreader url?

[David Stanford]

you can completely remove the RME database and restore an orig:

1) Stop the daemons

2) Solaris:

/opt/CSCOpx/bin/perl /opt/CSCOpx/bin/dbRestoreOrig.pl dsn=rme dmprefix=Essentials

Windows:

c:\progra~1\CSCOpx\bin

perl dbRestoreOrig.pl dsn=rme dmprefix=Essentials

3) Restart the daemon manager

This will completely wipe out the RME db, including devices, configs, images

[mjgleeson]

Thanks again davistan!

Got my db wiped and started again.. All healthy now, but I want to do a backup of the db before I add 1000 ish new devices... It's a windows setup btw.

Is there a tool for backing up/restoring the inventory db in the similar way to wiping it completely?

[David Stanford]

Hi,

LMS has its own backup tool built in:

Server configuration -- Administration -- Database management

and you can select backup now or schedule a backup