cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
325
Views
0
Helpful
2
Replies

Need advice on deploying redundant CMs and Core WAEs

d-fillmore
Level 2
Level 2

Hi, soon I'm going to be deploying a WAAS solution that uses 2 CMs and 2 Core WAEs - both running in inline mode.

In the documentation, I can't find anything about how the redundancu works for either the CMs or the Core WAEs.

Do you manually configure one of the CMs as a master and the other as a backup?

I intend on connecting the two core WAEs in a serial cluster and I've read that they will employ spillover load-balancing. What determines which on is the primary unit? do they each need to be configured individually?

I've not got a box in front of me yet so a lot of this may become clear once I've got the GUI to play with but if anyone can answer these questions or point me in the direction of some relevant documentation I'd greatly appreciate it.

Thanks in advance

Dom

1 Accepted Solution

Accepted Solutions

dstolt
Cisco Employee
Cisco Employee

Dom,

Bring the Primary CM online as device mode central-manager and central-manager role primary (default). Standby is brought online as a device mode central manager, but with the configuration as standby.

standbyCM #device mode central-manager

save config/reload

standbyCM (config)# central-manager role standby

standbyCM)# central-manager address primaryCM.customer.com (or IP address)

standbyCM #cms enable

Once it's registered with the Primary CM, it's IP address is replicated to all the devices registered and online under the current CM via the CMS database. If the primary fails, you have to manually promote the standby CM to primary and then all the devices registered will then start to come online in the new primary CM as they start to check into the old primary and it doesn't respond. You DON'T have to change any IP addresses.

For the core WAEs, the closest WAE to the hosts will be the primary unit as that is the one that will first insert the TCP options in the tcp handshake on the traffic. You should configure them individually as they will both act independantly, however you can push configs to both at once using a Device Group in the CM. You may also want to consider moving to out of band approach in the core in the future as best practice.

Check out the quick config guide and other doco on Cisco.com for a pretty good step by step configuration guide to set up the boxes.

Hope that helps,

Dan

View solution in original post

2 Replies 2

dstolt
Cisco Employee
Cisco Employee

Dom,

Bring the Primary CM online as device mode central-manager and central-manager role primary (default). Standby is brought online as a device mode central manager, but with the configuration as standby.

standbyCM #device mode central-manager

save config/reload

standbyCM (config)# central-manager role standby

standbyCM)# central-manager address primaryCM.customer.com (or IP address)

standbyCM #cms enable

Once it's registered with the Primary CM, it's IP address is replicated to all the devices registered and online under the current CM via the CMS database. If the primary fails, you have to manually promote the standby CM to primary and then all the devices registered will then start to come online in the new primary CM as they start to check into the old primary and it doesn't respond. You DON'T have to change any IP addresses.

For the core WAEs, the closest WAE to the hosts will be the primary unit as that is the one that will first insert the TCP options in the tcp handshake on the traffic. You should configure them individually as they will both act independantly, however you can push configs to both at once using a Device Group in the CM. You may also want to consider moving to out of band approach in the core in the future as best practice.

Check out the quick config guide and other doco on Cisco.com for a pretty good step by step configuration guide to set up the boxes.

Hope that helps,

Dan

Thanks Dan