'rediscover' servers

[Joseph Madden]

Folks,

We deployed HPSIM a few months ago. We discovered that most of the servers didn't have trust setup correctly and had older versions of the agents going all the way back to early 5.0 agents. So we upgraded agents and setup trust.

The question is most of these device show as 'Unmanaged' and I guess I thought the discovery process (which runs every 2 hours) would 'rediscover' these devices and gather the snmp data/im agent data they couldn't gather before.

I am I missing something?

Thanks
Joe

[jim goodman]

Have you re-run the Identify Systems under Options-->Discovery?

- Jim

[Joseph Madden]

Yep, Good idea and I did, and still no change - seems like it is not doing the snmpget of system information - any other ideas?

Thanks

Joe

[jim goodman]

I am sure you have already verified the SNMP Security and Trap Settings as well HpSIMs SNMP settings. And also rebooted the servers after the updates.

Here is something to try... set the SNMP Settings on one of the servers to accept SNMP Packets from any Host, then restart the agents, After the Agents restart look at the properties and see if the new settings are still there and re-run the Device Identification task.

Another idea, Have you deleted any of these from HpSIM then manually re-discover, then run Device Identification?

- Jim

[jim goodman]

I forgot to add....

Also on the Device System page (click on the name) Next to Protocols: make sure you see SNMP listed.

[Joseph Madden]

OK Jim, that last bit I believe is the ticket.

I checked Links(tab)=> System Protocal Settings and the snmp read/write strings were set wrong on the management server.

So I changed it on a couple of the servers using 'values below' instead of global and then ran Options=>Discovery=>Identify systems and it works.

So when I go to Options=>Protocol Settings=>Global Protocol Settings...it tomcat errors out on me - I have bounced the server and it is still doing it.

So what I am in the process of doing is Options=>Protocol Settings=>System Protocol Settings on All Sytstems and then rereun Identify Systems to see if that holds. Any idea on why the blow-up on the global page.

Thanks
Joe

[Joseph Madden]

So that didn't have the intended effect I was hoping - I actually lost some servers - in the process of updating the local settings for all systems - I lost the 'hardcoded' snmp read/write strings, and then when the discovery process ran it changed a few of them back to unmanaged.

Joe

[jim goodman · Accepted solution]

Well at least you are seeing some change. Is the Tomcat error happening all of the time? What is the exact Error?

Also I understand you have some of the servers now show to be identified and some are still unmanaged?

Do this, go into the unmanaged server Insight Managment Agents and go to the SNMP link and look the Security settings, you should see your community string and you r-w as well ip addresses for the localhost and the CMS. If you don't it is possible that the settings didn't take.

Manually apply them to that clicking apply in the popup to change and apply again on the main window then restart the agent from there. After a few minutes re-run the Identify Systems task under Options-->Discovery.

- Jim

[Joseph Madden]

Jim,

I am to the point that I get the 'problem' - it is I believe a function of the CMS not 'using' the right strings to query/communicate with the managed nodes - and I should be able to use the Global Settings editor which errors out on me.

I am going to create a new thread on this point and attach the error out.

Thanks for your help with this.

Joe

[jim goodman]

Awesome, all the best.

- Jim