HP StorageWorks 2/16V Brocade Fabric Manager Administrator's Guide (53-1000019 - Page 227

Collecting FDMI Device Information 14 Collecting FDMI Device Information FDMI device information is updated once a day as a server-side service, with a few exceptions: • Downloading firmware to HBAs starts FDMI data collection so that the data matches the download. • FDMI data for a new fabric is collected when it is discovered. Because the devices and ports must be in the database prior to storing FDMI data, data collection does not start until the device discovery is complete. You can, at any time, choose Discovery > Refresh to update the FDMI device information. If FDMI data cannot be collected, an alert appears with a brief description of the problem. The alert remains unresolved until the server can collect FDMI for that fabric. Note There must be at least one FDMI-capable switch in the fabric for FDMI information to be collected. Rebooting Switches This section provides information on creating reboot groups and performing sequenced reboots. A reboot group contains one or more switches from a single fabric. You cannot group switches from different fabrics, and each switch can belong only to one reboot group. Create a reboot group containing multiple switches from the same fabric if you want the selected switches to reboot simultaneously. You might want to create reboot groups containing switches that run the same firmware, serve the same function, reside in the same physical location, or share any other attribute. After creating reboot groups, you can then configure a sequential reboot of the reboot groups. A sequenced reboot allows different areas of your SAN to reboot and stabilize before other switches in the fabric begin to reboot. This reduces the load of interswitch traffic on the SAN. For example, you can use the following reboot strategies: • Simultaneously reboot switches that run the same firmware. • Simultaneously reboot switches of the same model. • Reboot the core switches of a fabric, then the edge switches. • Reboot the backbone of a large SAN, then reboot other sections. • Reboot distant physical locations sequentially. Refer to the following sections for information about creating reboot groups containing single or multiple switches, adding switches to existing reboot groups, and configuring a sequenced reboot: • "Creating a Reboot Group with Multiple Switches" on page 14-20 • "Creating Single-Switch Reboot Groups" on page 14-21 • "Assigning Switches to a Reboot Group" on page 14-22 • "Performing a Sequenced Reboot" on page 14-23 Fabric Manager Administrator's Guide Publication Number: 53-1000196-01-HP 14-19