Table of Contents
2 Boot without /etc/rc
3 Change password
4 Clean configuration and initialize all disks OR Initialize owned disks OR Assign ownership and initialize disks for root volume OR No disks assigned (use `disk assign' from Maintenance Mode)
5 Maintenance Mode
6 Update flash from backup config
7 Install new software first
8 Reboot node
After initiating the Boot Menu a list of choices is presented that allows the user to modify the boot process.
This causes a normal full boot sequence to be done, after which the system behaves just as if the system where booted without interruption.
option 2: Boot without /etc/rc
This does a normal boot, but bypasses execution of the /etc/rc file. Following this, the system is running normally, but without the configuration normally provided to it in the /etc/rc file. The commands in the /etc/rc file can be typed manually to bring the system to a fully operational state. Generally, this command is used when there is something in the /etc/rc file which is causing the node to misbehave. Often, only an ifconfig command and an nfs on or a cifs restart command are done manually, allowing NFS or CIFS to become operational; then the /etc/rc file is edited to remove the offending lines, and then the system is rebooted.
option 3: Change password
This allows the node password to be changed. It is usually used when the administrator has forgotten the current password, and so cannot use the online passwd command.
option 4: Clean configuration and initialize all disks
This commands zeroes all the node's disks and re-enters the setup menu. It is typically used only once, at system installation time. This option asks for confirmation; once confirmed, there is no way to retrieve data previously on the disks. Zeroing the disks may take time (sometimes hours), depending on how many disks there are, whether they need to be zeroed or not, and their capacity.
On systems with software-based disk ownership, option 4 initializes disks that are assigned to the system. If no disks have been assigned on systems other than V-Series systems, the software attempts to assign a minimum set of disks for the aggregate containing the root volume. After disks are assigned, they are zeroed and the user enters the setup menu. For V-Series systems the user must use option 5 to assign at least one disk (LUN) from the storage subsystem, then use option 4 to create the root volume. After disks are assigned, they are zeroed and the setup menu is entered.
option 5: Maintenance Mode
This enters a mode in which a small subset of commands is available, and is usually employed to diagnose hardware (often disk-related) problems. In maintenance mode, WAFL aggregates and traditional volumes are recognized but are not used, the /etc/rc file is not interpreted, and few system services are started. NFS and CIFS cannot be used. Disk reconstructions do not occur. No filesystem upgrade occurs, even if the system is newer than the OS release previously installed.
option 6: Update flash from backup config
Restores system configuration data that has been lost due to a change in the boot device state. This option is usually employed following a CompactFlash boot device replacement. The system boots normally to load the WAFL root volume and retrieve the backup configuration data. After the backup configuration data is restored onto the boot device the boot process is interrupted and the system automatically reboots.
option 7: Install new software first
Allows the download and installation of a new OS release on the boot device before completely booting the system.
In order to download the install package a temporary network interface must be configured. If needed, the user is prompted to enter the network device name (for example, e0a) and the system automatically reboots. Following this the IP address, netmask, default gateway address and URL of the install package must be provided.
option 8: Reboot node
Allows the system to be rebooted from the Boot Menu. This option can be used with option 7 or at any time the Boot Menu needs to be aborted.
When using option 6 in a HA configuration the system can stop in the Waiting for Giveback state before loading the WAFL root volume and restoring the backup configuration data. When this happens a cf giveback must be done on the partner system. Note that this can be disruptive as the system will automatically reboot to restore the backup configuration following the giveback.
Table of Contents