5412 discovery issue IMC7.0

[Mauro Furini]

Hi all,

    in a customer client I have some issue when try to discovery the two core, two 5412, the IMC (7.0 E0202 Basic) they are discovered only as a pingable device.

The other switch (2650, 2910 and 5406) are correctly discovered.

Anyone as an ideas?

At this customer we can't use telnet or ssh to manage the switch, but also the old PCM+ has worked in this way without any problem.

thanks

[LindsayHill]

There's three possible reasons for this:

• SNMP configuration mismatch between IMC & the switch. Double-check your community strings/SNMPv3 users & encryption settings
• ACL - very common problem if you're installing an NMS on a new IP address
• firewall between IMC & the switch.

The only other possible scenario is routing, but we know that's working, since ICMP is working.

You could run Wireshark on the IMC server to watch the SNMP packets leaving the IMC server, and see that lack of response from the switch.

[Mauro Furini]

Thanks for the reply, but the snmp configurations are the same of others switchs correctly discovered, the IMC server is on the same subnet of switch, this exclude acl and firewall issue...

Also, the snmp configuration server side is the same of the actual PCM+ server that is working fine...

:(

[LindsayHill]

How does being on the same subnet exclude firewall/ACL issues?

Run Wireshark on the IMC server. Watch the SNMP packets leaving. If you're using SNMPv2, look at the community string. See if you see any response. If you don't, then the problem is not the IMC setup.

[Mauro Furini]

The HP support told me also to run the batch operations  "check access settings" to verify this issue, maybe a write community error on the snmpv3 configuration.

But I have some doubt 'cause this template work fine with other switch...

Next week (After customer holidays) I will try.

Thanks

[LindsayHill]

The "Check Access Settings" option will do things like send SNMP requests, and try and login via SSH/Telnet, just to check if it all works.

It can tell you if things are all configured correctly or not, but you already know that it's broken somewhere. So you can run it, but it will fail.

You don't need working SNMP read-write to discover devices, you only need read-only.

Use wireshark. Verify that SNMP packets are leaving the IMC server, and no response is coming back. Then debug it the same as you would any other network connectivity issue.

[Mauro Furini · Accepted solution]

Problem solved, we have completely missed this line:

ip authorized-managers 192.168.1.1 255.255.255.255 access manager

on the configuration of the two core, it refer only to the ip of the PCM+, so the IMC was unable to manage this switch.

Thanks!