Prevent to display specific "fabric connectivity issues"

[alessandro.s]

Hi all,

the Assurance/Issues dashbord in our DNA Center  shows a lot of "fabric connectivity dhcp overlay"and "fabric connectivity dns overlay" issues, looking at the details of the issues i can see messages like the following:

'Co-located Fabric Border and Control Plane' "Border Node" Lost Connectivity to the DHCP Server Y.Y.Y.Y in the Virtual Network VN_Guest

the problem here is that Y.Y.Y.Y is not in Virtual Network VN_Guest so the behavior of non reaching this DHCP server from this VN is normal so what i wanted to know is : there's a way to say to DNAC that Y.Y.Y.Y belongs to another Virtual Network so i won't see any event about this?

Regards,

Alessandro

[marce1000]

 - FYI : https://bst.cloudapps.cisco.com/bugsearch/bug/CSCvr01356 . check if the Workaround could be applicable to your case , 

 M.

[Sylvain_Che]

Hi @alessandro.s ,

You have several options here:

- Allow the Guest_VN to reach your DHCP server Y.Y.Y.Y. Check CiscoLive 2020 - BRKSDN-2500.pdf on slide 81 and make sure to open firewall.

- Decrease the severity of the issue in DNAC Issue settings.

FYI, in DNAC 2.3.3.0+ and IOS-XE 17.6.2+, this KPI is removed from Assurance dashboard.

To strictly answer your question, there is no way in DNAC to say that your DHCP server belongs to another VRF.

Best regards,

Sylvain.

[alessandro.s]

Hi @marce1000 and @Sylvain_Che ,

thank you both for your replies, as i can understand the solution provided in CiscoLive 2020 - BRKSDN-2500.pdf is a description of the workaround described in https://bst.cloudapps.cisco.com/bugsearch/bug/CSCvr01356 , in every case in the next days i'll go to upgrade DNAC from 2.2.3.5 to 2.3.3.5 so, based on the fact that my BNs runs 17.6.3 version i'll not get any of these alarms

Thank you!

[alessandro.s]

Hi @Sylvain_Che ,

last weekend i upgraded to 2.3.3.6 version but i still display those events in the assurane dashboard so i don't now if there's something i'm missing or i misread the document you posted, by now i think i'll go through the workaround.

Regards,

[Sylvain_Che]

Is NETCONF enabled on the Border nodes?

[alessandro.s]

Hi Sylvain,

yes, NETCONF is enabled on border nodes, also i was checking the config for the workaround and i saw that it was still applied (at least for one of the involved VNs) so i decided to open a TAC to go deeper

[Sylvain_Che]

Ok. Please share with us the final outcome on TAC solved the issue.

Sylvain.

[alessandro.s]

Hi Sylvain,

i just realized i did not update you about, so TAC team told me that this behavior is expected and the feature is working as per design, the only thing i can do is ti disable the alert or to change the priority level and i did the latter lowering the priority level from P2 to P3 to avoid to see the alerts in the dashboard