Coyote Point Systems Equalizer Spezifikationen PDF herunterladen (Seite 456)

Failover

Refer to "Peer Interface Subnet States and Substates" on page 438 for descriptions of the Peer

states and substate conditions.

Configuring Active/Active Failover Between

Two EQ/OS 10 Systems

Active/Active (A/A) failover allows clusters to be active on both Peers that are configured into failover. For the

same failure situations that cause a Peer to take over all the cluster and floating IP addresses in an Active/Passive

failover configuration, Active/Active failover operates the same way - that is that the healthy Peer will take over all

of the cluster and failover IPs.

An Active/Active failover configuration consists of two peers. Equalizer’s clusters are instantiated on both peers

and organized into “Failover Groups”. If the one peer’s connectivity for the failover group’s resources is judged to

be “healthier” than the peer on which the group is running, then the group “fails over” to the other peer.

It should be noted that if and when the "sick" Peer is healed, there is no automatic migration of the clusters back to

it. You can, however, invoke a "rebalance" command to make this happen.

Failover Groups

Active/Active failover introduces the concept of "Failover Groups". A Failover Group consists of all the smallest

set of resources that may be moved between Peers and can consist of one or more clusters, servers, and failover

IPs.

Failover Groups are dynamically determined by the configuration and cannot be specified by the user.

In the simplest case, there is a maximum of 1 Failover group per subnet. However, based on the cluster/match

rule/server pool/server configuration, a Failover Group may contain more than one subnet. Basically, the algorithm

is that:

l a cluster subnet and, for

non-Spoof

match rules, all associated server subnets must all be in the same

Failover Group. This means that all clusters and failover IPs on any of these subnets are in the same

Failover Group.

l if

spoof

is set for all the cluster match rules, then the server subnets are not factored into the Failover group.

This means that as clusters and servers are added or deleted, the Failover group configuration may change.

Configuring Active/Active Failover (CLI)

1. Configure failover using the current procedures described in "Configuring Active/Passive Failover (CLI)" on

page 443.

2. Activate the Active/Active failover mode by setting the active-active flag on the local Peers. This flag must

be set on both Peers for A/A to be enabled. If the flag is set on only one, or no Peers, failover operates as in

the current Active/Passive mode.

You will need to access both Equalizers so it may be easier to open two TTY sessions. Each one should access

the local peers. Enter the following for each local peer:

456

1 2 ... 451 452 453 454 455 456 457 458 459 460 461 ... 593 594

Keine Kommentare

Coyote Point Systems Equalizer Spezifikationen Seite 456