This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

NCM Manual Download of Configs - timing issue

Hi Folks,

we have a NCM Only installation where we usually do automated config backups during the night.

Now some of our admins complained that they are not able to either download a running or a startup config of a Cisco device that is TACACS enabled.

We narrowed it down to the following:

- When you download either config (Running or Startup) you need to wait approx 30 seconds before you can download another config (manually) - we were able to verify this with a few nodes.

- This happens with either prototcol (SSH or Telnet)

- This only happens on some nodes, not on all nodes

We assume the following:

- TACACS is configured that only 1 connection per user at a time is possible

- when we disconnect from a node, the information needs to be sent/processed to the tacacs server depending on the tacacs config, and this is the duration when we can not log in again to the device

We are not maintaining the TACACS System and only have a brief understanding of that, but we need to get as much information as possible to ask the TACACS admins to help us out.

Can someone give us some ideas on how to troubleshoot

Thanks,

Holger

  • I'm not personally too familiar with the TACACS side of this but on the NCM side you can break your backup scheduled job into two parts, one pass for running configs and another starting a few minutes later for startups. Hopefully that gets around this issue for you if there isn't an easy way to fix it in the TACACS setup.

    -Marc Netterfield

        Loop1 Systems: SolarWinds Training and Professional Services

  • I use this exact solution--with TACACS.  One job for backing up startup configs, a separate job run later for catching running-configs.  It works well.

  • Which version of NCM are you using?

    And what TACACS solutions are you running--ACS? Or something entirely different?

    I've had many problems with ACS logging/reporting capabilities and reliability, and am anxiously awaiting for our ISE to replace it.

    But the earlier advice is appropriate:  One job for backing up startup configs, a separate job run later for catching running-configs.  It works well.

  • Hi Guys,

    thanks for your comments.

    the automatic download jobs work perfectly fine. It is just when the fellow device admins are impatient and download startup and running config manually on the webpage. We have about 120 admins working with NCM And many of them are not patient. The Orion Team is constantly cancelling Download Jobs that are "stuck"'on itializing because others were not waiting 30 seconds before hitting the Download button again.

    I am almost certain that this is a TACACS timing issue however at the moment it is my/Solarwinds fault And I want to get some proof that it is not.

    thanks

  • 120 Admins using NCM?  Impressive?

    How many nodes are on your network?

    How many elements do you manage?

    I have 4 pollers, NPM, NCM, NTA, just under 800 managed nodes, elements are about 35,000.

  • 35.000 Devices/Nodes over 3 Polling Engines NCM Only (unfortunately)

    import of nodes via SDK from a CMDB

    my largest installations when it comes to number of nodes.

    But not the most complex installation

    cheers

    holger

  • I didn't answer your question correctly... the Customer has over 70.000 Network devices, but only half of them need to be managed by NCM. So the Network is a lot bigger than the NCM Installation represents.