Cisco Works Server sending excessive ARPs

[mbrown]

I have CW2K running on Solaris 7. Several months ago, the IP addresses of a number of network devices were changed. The devices were deleted and re-added to the CW databases. Today, it was discovered that the CW server is ARPing out to the old IP addresses. As they no longer exist, the CW server does not receive any replies. It is continuously ARPing and causing an excessive number of broadcasts. Does anyone have any experience with this issue?

[rmushtaq]

There could be several factors for this:

- Could be due to that fact that the device isn't completely deleted from the database.

- Could also be that CM still has a record of the device is some one's CDP table.

- You also may want to check to see if the packets are being produced by another source.

[mbrown]

I have verified that the devices are not in other CDP tables, and that the packets are being produced by the CiscoWorks Server.

How do I verify that the devices are completely deleted from the database and, if not, delete them completely?

[rmushtaq]

You'll have to access the database manually and need to know the db schema as well. Without knowing the schema you may end up hosing the db completely.

One workaround that you may consider is to export your RME inventory to a .csv file using Export to File funtion. Then re-initialize the RME datbase to clean it up completely. After that, import the RME inventory back using Import from File feature. Caution: You may loose your old device configs/archives etc completely if you follow this. Do it at your own risk.

[mbrown]

Thanks for the info. I'll give it a try.