![](../ng932.gif)
Configuring disaster recovery fail over to another high availability group
You can create a configuration where a high availability (HA) queue manager fails over to another HA group at the recovery site.
If you are running an HA queue manager at a site that experiences a severe site-wide outage, you can fail over to a disaster recovery instance of that queue manager at a remote site. The queue manager then runs as part of an HA group at the remote site.
This configuration is represented in the following two diagrams.
Figure 1 shows the queue manager terentia1
running on its preferred appliance,
Pollux
, at the Rome site. Figure 2 shows the desired situation after the Rome site
fails, where the queue manager terentia1
is running on its preferred appliance,
Hugin
, at the Stockholm site. The arrangement is reciprocal. In normal operation,
the Rome site has a DR instance of the Stockholm HA queue manager Freya2
.
For instructions on how to implement a fail over in this situation, see Operating in a DR/HA environment.
![shows queue manager running on original HA group](graphics/HADRSETUPnew.gif)
![shows failed over queue manager](graphics/HADRSETUP2new.gif)