We have a requirement to load balance web traffic for services such as Outlook Web Access across two datacentre locations. I understand that the GSS devices would be what we require (?)
In addition we require internal load balancing of some Exchange CAS servers in an RPC Client Access Server array which stretches between two datacentres as well. I have been informed that the CISCO device to be used for this would be the ACE in conjunction with the GSS.
Could someone please be so kind to explain how this would work. So far I think it would work as follows:
For external access:
1.Client makes a DNS request for webmail.company.com
2.Query makes its way to the GSS device via normal DNS
3.The GSS device polls the ACE boxes and determines best IP (?) How does the GSS poll the ACE, what protocol is used (??)
4.The GSS device sends back a response to originating DC which in turn repsonds to the client with an IP
For internal access
1. Client makes a DNS request for outlook.company.com (RPC Client Access Array)
2. Internal AD DNS has a NS record for outlook.company.com which directs the internal DNS server to the GSS in the DMZ
3.The GSS device polls the ACE boxes and determines best IP
4.The GSS device sends back a response to the AD DNS server which then returns the response to the client
So in the event that both ACE servers are down in the local site or both CAS servers are down the GSS will return the IP address of a CAS server in the 2nd DC.
Is this correct or am I way off here. An explanation would be greatly appreciated.
GSS would be set up with keepalives which would poll poll ace vips in both datacenters. The keepalive type can be icmp, tcp, http or kal-ap.
The gss then can be set up to either use ordered list balancing which would return answers for the primary datacenter and return answers from the backup datacenter if a vip on the primary goes down. Or you can use round robin to balance between the 2.
GSS when using KAL-AP keeplaive from ace can balance answers based on load reported back to GSS from ace. But in ace this load value is purely a function of how many servers are in service in the server farm not actual load. You can however use something like maxconns on ace to take a server out of rotation and manipulate this value based on connections on servers.
Somewhat more complicated is configuring the GSS to use proximity which can reach out to routers acting as drp agents and measure rtt times and return the answer that is most proximate to the requesting d-proxy.
You can also control the answer eturned based on source address lists say if you want users that are internal to get a different answer than users that are external, you can make source address lists defining the internal networks and use them to determine which dns rule and subsequently which answer to return.
The unmanaged mode is also known as Network only switching, which is introduced in Brazos release. It adds the flexibility for customer to use only network automation for service appliance.
If a device is configured a...
Usually, we can access ESXi Shell by pressing Alt+F1 from ESXi DCUI (Direct Console User Interface).
But on HyperFlex system, it just shows black window.
This is expected behavior because HyperFlex redirects ESXi Shell output to SoL...
Configuring an Export Policy Using the GUI
This procedure explains how to configure an Export policy using the APIC GUI. Follow these steps to trigger a backup of your data:
On the menu bar, choose Admi...