Forum Discussion
HA & Disaster Recovery with Auto Failover and Auto Failback
Does F5 offer a Disaster Recovery configuration that includes an HA pair that is synced with a DR node failover device with auto failback?
Scenario: Both the Active and Standby go hard down and the DR node (Forced offline) automatically releases from Forced Offline to Online Active with no manual intervention.
When either of the original HA pair members then come back online, they take over as the Active standalone or active and standby sync, while the DR node goes back into Forced Offline automatically.
- rob_carr
Cirrocumulus
Not exactly.
However, you can simulate the effect you are looking for by creating a sync-failover device group with three members, where the traffic group is set to use 'Failover using Preferred Device Order and then Load Aware' and has 'Always Failback to First Device if it is Available' enabled.
You then specify your devices in the order site 1 device A, site 1 device B, site 2 device. The site 2 device will only be used when neither site 1 device A or B are available. If the site 2 device is active, and either (or both) site 1 device becomes active, the traffic group will fail back to one of the site 1 devices.
- youssef1
Cumulonimbus
Hi,
Natively you can't do it but using an irule it's possible:
can you try the following irule (I don't test it, keep me in touch)
when CLIENT_ACCEPTED { set mypick [LB::select] If we have both node UP we force traffic trough node 1 if { {[active_members poolName] == 2} } { First node if { [LB::status pool prod_443 member 1.1.1.1 443] eq "up" } { pool prod_443 member 1.1.1.1 second node } elseif { LB::status pool prod_443 member 2.2.2.2 443] eq "up" } { pool prod_443 member 2.2.2.2 } else { eval $mypick } } }
Regards
Recent Discussions
Related Content
* Getting Started on DevCentral
* Community Guidelines
* Community Terms of Use / EULA
* Community Ranking Explained
* Community Resources
* Contact the DevCentral Team
* Update MFA on account.f5.com