08-16-2024 12:42 PM
Hello community,
currently we are running 5520 AireOS and testing C9800 SSO cluster without DNA center.
In the future we are planning to discover, provision and manage 2x C9800 SSO clusters in DNA Center. What are your experiences and thoughts so far? Would you recommend completely configuring the clusters via DNA with the available templates or is it better to do it via C9800 CLI/GUI? I keep hearing controversial opinions about DNA and C9800.
Thank you in advance!
Regards
08-16-2024 11:46 PM
- Normally , if the controllers are intended to be managed by DNA Center , then HA SSO must be deployed with DNA Center too :
https://www.cisco.com/c/en/us/td/docs/cloud-systems-management/network-automation-and-management/dna-center/2-3-3/user_guide/b_cisco_dna_center_ug_2_3_3/m_provision-wireless-devices.html#id_91520
M.
08-17-2024 12:45 AM
@klnnnnng wrote:
2x C9800 SSO clusters in DNA Center.
1. What is the exact model of the WLC?
2. Present total amount of APs? Total amount of APs in the next 3 years?
3. Present peak daily wireless client?
08-17-2024 03:44 AM
Hi Leo,
9800-40, 2000+ AP's and 10k client peak.
Regards
08-17-2024 05:46 AM
@klnnnnng wrote:
9800-40, 2000+ AP's and 10k client peak.
9800-40 cannot handle >1000 x APs.
Read the 03 May 2024 revision of the Cisco Catalyst 9800 Series Configuration Best Practices. Half of the document talks about the role of the WNC daemon (aka "load balancer"). (There are 4 WNCD queue in the 9800-40 and 8 WNCD in the 9800-80.) This document talks about the "80% rule":
C9800 design is no different and, generally, Cisco recommends limiting the load to around 80% of the AP and client scale.
The 80% scale is just a recommendation to start planning the design and deployment of a catalyst wireless network as this is tested and validated number.
For C9800-80, for example, this means 4800 APs and/or around 50k clients. Does this mean that you cannot have six thousand APs on a single C9800-80? No, not really; Cisco has a lot of successful deployments at maximum scale. The 80% scale is just a recommendation to start planning the design and deployment of a catalyst wireless network.
Poor quality of the firmware has caused the WNCD to spring memory leaks when certain conditions are exceeded. Among them are:
Below is the control-plane memory utilization of our 9800-80 (firmware 17.12.3) with an uptime of 12 weeks:
Before the spike, our controller had 2900 APs. As soon as the AP count exceeded 3000 APs, the control-plane memory utilization increase due to memory leaks.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide