cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
7382
Views
0
Helpful
8
Replies

Health status change to "Data Layer Partially Diverged" in Single APIC cluster after Upgrade to 1.1(1j)

failee2007
Level 1
Level 1

Any idea? The vmmmgr process seem to have problem, and creating EPC in apic does not create port-group in vm DVS. We are unable to downgrade back to 1.0(3n) because controller health is not in "fully fit" state. 

 

 

 

 

8 Replies 8

Zach Seils
Level 7
Level 7

Can you please see if any core dump files have been created?  You can check with the command 'show cores' on the APIC.

Zach

yes, i have attach the show cores file here. 

 

 

This is just showing core files on the switches (which we'll look at later).

On the APIC, can you check the /data/techsupport directory for an core files?

Thanks,

Zach

Hi Zach, 

I just rebuilt the fabric instead of troubleshoot it today, because we have a poc for customer very soon....and once i downgrade to 1.0.(4h). The vmmmgr service resumed immediately. But when i try to upgrade 1.1.1j again, the vmmmgr service down again...will rebuild the fabic again tomorrow and stay at 1.0.4h first. 

By the way, the cluster health is return to "Data Layer Partially Diverged", and i can not downgrade the controller firmware becoz it is not fully fit, anyway to bypass it so next time i dont have to rebuild the whole fabric again?

There are 2 core file generated after upgrade to 1.1.1j. Please let me know how to upload the core file to you. Thank you very much.

dbgexp_coreexp-default_apic1_sysid-1_2015-06-24T01-34UTC_svc_ifc_vmmmgr.bin_log.2107.tar.gz
dbgexp_coreexp-default_apic1_sysid-1_2015-06-24T01-35UTC_svc_ifc_vmmmgr.bin_log.2153.tar.gz

 

 

I had this exact same problem with one of my APICs after doing an OS downgrade on my fabric.  I tried rebooting the node several times, but still I kept getting the partially diverged error.  In the end I removed/decomm’d apic1 did an eraseconfig setup, then rebooted it.  Then I went thru the initial setup config, commissioned it on another apic, then rebooted the leaf.  All seems to be fully fit now.  I guess that forced a total DB replication to the new node, and is what fixed my issue. 

Hi Zach, 

i have uploaded the core file. Thank.

You can upload the files to:

ftp.cisco.com/incoming/TAC

Please let me know once they are there.

Zach

Hi failee2007,

Feel free to contact me directly @ seils@cisco.com for help with this issue.

Zach

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Save 25% on Day-2 Operations Add-On License