Planned outage on site 2 (H2 MM secondary) and return to three-site configuration

Edit online

In this scenario, there is a planned outage expected on site 2. Disaster recovery capability is still needed so the copy will be set up between site 1 and site 3. When site 2 is available again, you can restart the session and use the original session configuration. Procedures are shown for both the multi-target Metro Mirror - Global Mirror session type as well as the Metro Global Mirror cascaded session type.

About this task

Follow these steps to carry out this scenario.

Procedure for Multi-Target Metro Mirror - Global Mirror session type

Edit online

Procedure

Start H1->H2 H1->H3
Run this command to begin Metro Mirror - Global Mirror on your system when the production I/O is running on H1. After the session is in a Prepared state, both sites 2 and 3 will have a consistent copy of the data.

Note: Production I/O does not stop running to site 1 to perform the following steps.
Suspend H1H2

This command will suspend the replication between Site 1 and Site 2. The data at Site 2 will be consistent but the data will not contain any new writes issued to the Site 1 volumes. The session will end up in a Suspended Partial state. Replication will continue between Site 1 and Site 3.
Note: In order to ensure the data at Site 2 is consistent, a freeze of the I/O is done prior to suspending the replication.
Failover H2

This command is optional and can be performed if Site 2 is still accessible. It will failover the relationships between Site 1 and Site 2 so that the volumes at Site 2 will allow writes. Production I/O can continue on H1. Testing can be done on the volumes at Site 2 without affecting production. When ready to return to a full three site configuration continue to the next step.
Start H1->H2 H1->H3
After the session reaches a Prepared state, site 2 and 3 will have a consistent copy of the data again.

Procedure for Cascaded Metro Global Mirror session type

Edit online

Procedure

Start H1->H2->H3
Run this command to begin Metro Global Mirror on your system when the production I/O is running on H1. After the session is in a Prepared state, both sites 2 and 3 will have a consistent copy of the data.

Note: Production I/O does not stop running to site 1 to perform the following steps.
Start H1->H3
This command starts a Global Mirror session between site 1 and site 3, during the outage of H2. Production I/O can continue on H1. After the session has reached a Prepared state, a consistent copy of the data will reside on Site 3.
1. If running [() DS8000]® devices on H1 with the correct level of microcode, and enough time has passed since the initial start of the session, the creation of the H1 to H3 pairs will not require a full copy.
2. Data on Site 2 is no longer consistent at this point.
3. After Site 2 is brought back up and is ready to hold a consistent copy of the data, go to the next step to return to the full three-site configuration. Production I/O never has to stop writing to H1 in this scenario.
Start H1->H2->H3
After the session reaches a Prepared state, site 2 and 3 will have a consistent copy of the data again.