P4500 - Cannot login to one of the nodes

[NicholasA]

Hi all,

We have a P4500 running 5 nodes, and out of the blue, the CMC cannot connect to one of the nodes. As far as I can see, the node is still functional.

A couple of tests I did:

I tried generating the Support Bundle, with the following error for the node in question
"An error occurred trying to obtain log information from device: 10.254.251.117. duplicate entry: 10.254.251.118.tar.gz"

I did connecte to all the nodes via the CLIQ, so credentials to the system also work

I then used the local cliq binary, to generate the NSM Logs from the node, with the following error:
"Couldn't create nsm support info Failed to bundle OS configs., lhn/public/system/info/vloggen/"

In the CMC, all nodes except for the problematic are connected. The attached screenshots are the messages when trying to connect to the one in question

This is an inherited system, with no support and I would appreciate any pointers where to start looking.

[VikasMandal · Accepted solution]

Dear @NicholasA

Thank you for posting your question.

In order to resolve this issue, you may restart the eman/hydra service on all nodes in the management group. These are non-disruptive commands and can be used while system is running.

For each node in the management group, use putty to open an ssh session on port 16022.
This will give you a login prompt. Use the management group credentials. 

At the CLIQ> prompt, type these commands (including the quotes): 

utility run="servicectl eman stop"
utility run="servicectl eman start"
utility run="service hydra restart"
(These commands may ask you to press "y" to provide your approval.)

Close each putty session when done. Close and restart the CMC.
Try to login again and run another management group support bundle.

If the issue persists, then a root level support access is required (Only through HPE support). In case you need to open a case with HPE support, I would recommend you to renew the contract for better assistance.

Hope this helps.

Regards,
Vikas

[NicholasA]

Dear Vikas,

Thank you so much for your response. I followed your guide, and the CMC is now connecting to the misbehaving node again.

To be precise, I only restarted the services on the system that was manifesting the problem, and it is resolved. Support Bundle was also generated succesfully for all the operational Nodes.

Thanks for the help!

Nicholas