Home > Storage > PowerMax and VMAX > Storage Admin > Implementing Dell SRDF SRA with VMware SRM > Configuring VMAX/PowerMax replicated devices
Once the SRDF SRA and SRM Array Managers are configured, dynamic SRDF-replicated devices can be discovered and managed. This section describes the discovery and configuration process that allows these devices to be eventually included in SRM Protection Groups and Recovery Plans.
VMware SRM automatically invokes the SRDF SRA to perform a discovery of the replicated LUNs between the relevant storage arrays as soon as a new array pair has been enabled. A successful completion of the process results in the detailed discovery of the replicated environment including the serial number of VMAX/PowerMax storage arrays at the protected site and the recovery site, and the number of replicated devices on the protected site VMAX/PowerMax storage array. For each device pair the target and source VMAX/PowerMax device number is displayed as well as the direction and status of the replication. If the device pair hosts a VMFS volume, the volume name will appear.
Furthermore, when the devices are a part of a device group or composite group on BOTH the local and remote Solutions Enabler server, the name of the group will be listed. If only the R1 devices are grouped or only the R2 devices are grouped the devices will not be discovered as valid candidates and errors such as below will be displayed in the SRDF SRA log:
[WARNING]: Skipping this source device as the peer device is contained within a remote group [<remote group name>]
Recall that groups are a mandatory prerequisite if planning on using the test failover functionality. SRM will fail the test if they are not present.
Note: The respective local (RDF1) and remote (RDF2) device/composite groups for a given set of devices do not need to be named identically to be accepted as valid group pairs by the SRDF SRA, unless using a SRDF/Star configuration.
To discover pairs in SRM use the button DISCOVER DEVICES as in Figure 39.
If a device does not meet the SRDF SRA requirements it will not be shown in the discovered devices list. Devices must meet the following requirements to be discovered as candidate devices by the SRA:
The SRDF SRA initiates the device discovery by asking for all SRDF device pairs and Solutions Enabler then returns them. Most incorrect configuration exclusions will be logged by the SRDF SRA.
Note: During a discovery operation, if the SRDF SRA encounters SRDF Star environments with sites in a “disconnected” state, the SRA will attempt to connect and protect these sites as well as enable Star. This may include configurations that the customer does not plan on using with SRM. Sites in an “isolated”, “halted”, “tripped”, “pathfail”, or “connected” states will be ignored. Discovery operations that execute these Star protection work-flows will experience longer than average discovery durations.To prevent the SRDF SRA from connecting any disconnected Star configurations, please enable the advanced parameter IgnoreDisconnectedStar found in Device discovery advanced options.
The SRDF SRA requires that all devices are in device groups or composite groups with consistency enabled prior to creating protection groups. When devices are not grouped, each replicated device is presented in its own datastore group since SRM and the SRA do not know of device dependencies and application relationships beyond that of a virtual machine spanning multiple devices. This can lead to incorrect configurations when creating protection groups and recovery plans as required devices may not be properly grouped.
Note: With SRDF/A, all devices in the SRDF group must be failed over in unison. Failover of a subset of devices in a SRDF group is not allowed and is blocked by the array. Therefore, if there are SRDF/A devices that are not to be failed over in unison, it is required that they be put into separate SRDF groups in accordance with the desired level of failover granularity.
The SRDF SRA presents the devices to SRM in a standard format such as the example below:
<SourceDevice id="014C" state="read-write">
<Name>014C</Name>
<TargetDevice key="0181" />
<Identity>
<Wwn>60:00:09:70:00:01:92:60:36:03:53:30:30:31:34:43</Ww
n>
</Identity>
</SourceDevice>
SRM takes the effective (external) WWNs/Mobility IDs of the devices and attempts to match them with devices that contain a VMFS volume or are in use as an RDM. SRM examines virtual machine/device correlations and consistency groupings to create the datastore groups presented in SRM for addition to a protection group.
The following three advanced options of the SRDF SRA control the behavior of device discovery: