PowerFlex: How to Set up replication on peer systems
Summary: After deploying Storage Data Replicator (SDR) packages on both peer systems, create peer connections, volume pairs and Consistency Groups. Then add the volume pairs to the Consistency Groups. ...
Instructions
Prerequisites
Ensure that SDR packages are deployed on peer systems. A minimum of two SDRs are required on each system.
Ensure that you have network connectivity between the peer systems.
About this task
At various stages of the procedure, the following query commands can be used to verify replication configuration status:
scli --query_all (to query asset capacity, system information)
scli --query_all_sdr
scli --query_all_replication_peer_system
scli --query_all_volumes
scli --query_all_replication_pairs
scli --query_all_replication_consistency_groups
Most of the actions shown in this workflow can also be done using the PowerFlex UI.
The CLI commands described in this workflow are typical examples, and for clarity, do not include all possible options.
The full syntax and options for each command are described in the PowerFlex CLI Reference Guide.
Steps
- Open a CLI session with each peer system.
scli --login --username <NAME> [--password <PASSWORD>] [--ldap_authentication | --native_authentication] [(--approve_certificate_once | --approve_certificate]
--accept_banner_by_scripts_only
- Extract a root certificate from each system so that it can be imported into the other system. Perform this step on both systems.
scli --extract_root_ca --certificate_file <PATH_AND_CERTIFICATE_FILE_NAME>
The certificate file name is a user-defined name.
- Copy each certificate to the other system either by physically loading it on the server, or over the network.
For example:
scp <FILE_PATH_AND_FILE_NAME> <USER>@<PEER_SYSTEM_IP_ADDRESS/FILE_PATH>
- On each system, add the certificate that you copied to it in the previous step.
scli --add_trusted_ca --certificate_file <PATH_AND_CERTIFICATE_FILE_NAME> --comment <COMMENT>
- Use the following command to check that the certificate was added successfully:
scli --list_trusted_ca
From this point and onwards, the procedures described below can also be done using the PowerFlex UI.
- Obtain the primary MDM IP addresses and system ID for both systems.
- Use the
scli --query_allcommand to find the system ID:
Look for the ID in the System Info section of the output.scli --query_all - Use the following command to find the primary MDM IP addresses:
scli --query_cluster
- Use the
- Ensure that at least two SDRs have been added to each peer system, and that they have "Connected" status.
- Query each system and check the SDRs:
scli --query_all_sdr
- Query each system and check the SDRs:
If you are performing the initial setup of the replication system, you may have to set the replication journal capacity. For more information about calculating the replication journal capacity, refer to "Prepare the CSV file" in the Deploy PowerFlex Guide.
scli --set_replication_journal_capacity --protection_domain_name <NAME> --storage_pool_name <NAME> --replication_journal_capacity_max_ratio <VALUE_IN_PERCENTAGE>
- Use the following command to define the maximum capacity for the remote system to use for replication consistency groups (RCG) sending to this Protection Domain.
scli --set_remote_protection_domain_capacity_threshold (--protection_domain_id <ID> | --protection_domain_name <NAME>) (--remote_protection_domain_capacity_threshold <THRESHOLD>)
- Use the following command to define the threshold on the local Protection Domain of moving the replication consistency group (RCG) to slim mode.
scli --set_protection_domain_slim_mode_threshold (--protection_domain_id <ID> | --protection_domain_name <NAME>) (--protection_domain_slim_mode_threshold <PERCENT>)
If SDRs are not present, add them to the system using this command:
scli --add_sdr --sdr_ip <IP_ADDRESS> [--sdr_ip_role <ROLE>] [--sdr_name <NAME>] (--protection_domain_id <ID> | --protection_domain_name <NAME>) [--sdr_port <PORT>] [--profile <PROFILE>] [--force_clean [--i_am_sure]]
Where:
<ROLE> is the role used by the SDR IP address: application_only, storage_only, external_only, application_and_storage, application_and_external, storage_and_external, or all
<PROFILE> is the SDR performance profile: high or compact
- Create the connection between peer systems. On each system, use the command:
Where:scli --add_replication_peer_system --peer_system_ip <MDM_IP_ADDRESSES> --peer_system_id <SYSTEM_ID> --peer_system_name <SYSTEM_NAME>
<MDM_IP_ADDRESSES>lists the MDM IP addresses from the peer system.The IP address of each MDM member must be listed (both primary and secondary) to support MDM switchover scenarios. If external networks are used, the external network IP where the MDM process is running should be used.<SYSTEM_ID> is the System ID of the peer system.<SYSTEM_NAME>is the name that is assigned to the peer system.- Create volume pairs are used as source and destination volumes for replication. Each volume must have a matching pair of exactly the same size on the peer system. For example:
scli --add_volume --protection_domain_name <PROTECTION_DOMAIN_NAME> --storage_pool_name <STORAGE_POOL_NAME> --size_gb <SIZE> --volume_name <VOLUME_NAME>
- Either now, or later, map the intended source side of a volume pair to Software-Defined Controllers (SDCs). For example:
scli --map_volume_to_sdc --volume_name <NAME> --sdc_name <NAME>
On the destination side of a volume pair, do not map the volume. It is read-only once replication configuration is complete. Do not create a pair containing a destination volume that is mapped to SDCs with read_write permission.
It is possible to map the volume on the destination side with access_mode as Read Only or without Read/Write access. For example:
scli --map_volume_to_sdc --volume_name <NAME> --sdc_name <NAME> --access_mode <read_only>|<no_access>
- Create one or more Consistency Groups on the system which contains the source volumes. The data is copied from this side to the peer system (destination).
Each Consistency Group aggregates communication between volume pairs, and parameters assigned to a Consistency Group apply to all the volume pairs in the Consistency Group.
The system on which the Consistency Group is created is the source side of the volume pairs. It is possible to create Consistency Groups on both systems, thereby creating a scenario where data is being copied in both directions.
scli --add_replication_consistency_group --destination_system_name <DESTINATION_SYSTEM_NAME> --protection_domain_name <PROTECTION_DOMAIN_NAME> --remote_protection_domain_name <REMOTE_PROTECTION_DOMAIN_NAME> --rpo <SECONDS> --replication_consistency_group_name <REPLICATION_CONSISTENCY_GROUP_NAME>
Where:
<DESTINATION_SYSTEM_NAME>is the name of the destination replication system to which data is copied.<PROTECTION_DOMAIN_NAME>is the name of the Protection Domain in the system that you are configuring (source side).<REMOTE_PROTECTION_DOMAIN_NAME>is the name of the Protection Domain in the destination replication system.<SECONDS>is the Recovery Point Objective (RPO) in seconds. RPO is the point in time to which data can be restored following an event. If you performed a backup at 6 PM, and a server failed at 8 PM, your RPO would be two hours, and any data created during that two-hour time span would be lost.<REPLICATION_CONSISTENCY_GROUP_NAME>is the name of this Consistency Group.
Add replication pairs to the Consistency Group. Repeat this command for each pair.scli --add_replication_pair --destination_volume_name <DESTINATION_VOLUME_NAME> --source_volume_name <SOURCE_VOLUME_NAME> --replication_consistency_group_name <REPLICATION_CONSISTENCY_GROUP_NAME>Where:
<DESTINATION_VOLUME_NAME>is the name of the volume to which data is copied.<SOURCE_VOLUME_NAME>is the name of the source volume.<REPLICATION_CONSISTENCY_GROUP_NAME>is the name of the Consistency Group to which you are adding the replication pair.
If the source volume is large, initial copying of the data to the destination volume may take some time.
Results
When all required Consistency Groups have been created, and replication pairs have been added to them, the replication configuration is complete.
Additional Information
Related documents (Log in as a registered Dell Support user may be required to view these documents):