Many issues using OneCollect/ConfigAdvisor with FlexPod

[TMACMD_old]

running both the new onecollect (for switches only) and config advisor.(flexpod, iscsi)

both are not capturing the data from the 9K switches correctly.

for instance, "show running-config" seems to be chopped well before the beginning. It is missing the first 130 lines!

It may be related to not correctly getting the data above, but the output of config advisor is also saying that TELNET is ENABLED and SSH is DISABLED and yet the opposite is true.

Also showing VLAN issues which I know are not true

Also showing issues when trying to use CDP. The Netapp side picks up the CDP info, but the switch is advertising the data from the peer-link. Therefor the data does not match up.

Logic could/should be in place to catch this as I am seeing it more frequently as a security mechanism on the switch.

also, the software mentions bootarg.init.switchless_cluster.enable is set to tru and should be set to "1" -> no place on the support site can I find this mentioned.

Please prove that that is correct. every document I find on the support site indicates that bootarg.init.switchless_cluster.enable=true is the correct way.

[NithinR]

Hi,

   Sorry for the inconvenience caused! We will check the output truncating issue with Cisco 9k series.

For the ConfigAdvisor related issue, could you please send across the collected data files to ng-configadvisor-questions@netapp.com. bootarg.init.switchless_cluster.enable=true is the right setting.

Thanks,

   Nithin

[Mickhanna]

I have the same issue with ConfigAdvisor showing an issue with the boot variable.  Nothing is flagged in autosupport that I can find.  I have only been able to find this thread about the issue.

Error Boot variables bootarg.init.switchless_cluster.enable are not set properly in node wcsonetapp-01. Incorrect values to these bootargs could be detrimental to system health and/or performance. The correct value(s) should be bootarg.init.switchless_cluster.enable="1"The corrective action is to boot each node in the cluster into the loader prompt and use the setenv command on each variable that requires a change

[ajeffrey · Accepted solution]

You might try using the new Converged Systems Advisor for validating FlexPod. It covers Config Advisor rules as well as totally new rules for FlexPod (ONTAP, UCS, Nexus, MDS, VMware).

Try it out at https://csa.netapp.com

see https://mysupport.netapp.com/products/p/ncsa.html for details