For NetWare network users, it is not safe to store data with only one disk, because any physical device may fail or expire. So, when there is only one hard drive in your server, do you want to add a hard drive to back up the network data? Disk mirroring means that two hard drives are connected to the same hard disk control card, using a hard disk control card to manage two hard drives. When the user writes data to the server, the disk mirroring technique is written to two hard drives, so that once a hard disk is corrupted, it can obtain data from another hard disk to maintain the normal operation of the network. First, install and set two hard drives 1. The newly added hard drive for new hard drives is called slave, that is, mirror disk; the original hard disk in the server is called a master. (MASTER). From the plate best with the main disc for the same capacity of the same manufacturer. From the capacity of the disk should be greater than the capacity of the primary disc, because the large-capacity hard disk can be mirrored to small-capacity hard drives, otherwise. When using a large-capacity hard disk mirror small capacity hard disk, adjust the size of NetWare Partition Size so that the size of the two hard disk network partitions is the same, otherwise the error information of "size mismatch" will appear. In order to reduce some cumbersome operations and avoid unnecessary errors, the author recommends the same hard disk with the same capacity. 2. Setting the installation of the hard disk server second hard disk and installs the method and precautions for installing the second hard drive in a normal PC basically. First, according to the jumper on the hard disk manual, set the newly added hard disk to slave the disk, after the completion, the slave is installed on the hard disk bracket of the chassis, and the power connector is connected (the host power supply must be disconnected when operation), and then according to the requirements Connect from the disk to the same cable on the connection the main disc. There are generally two or more hard disk connections (IDE is 40PIN connector, while SCSI is 50PIN connector), it is recommended to connect the primary disc at a joint closest to the hard disk control card, from the disc connection The connector behind the main disk connector. If you install the hard drive of the SCSI interface, there is a terminal resistance (generally milky white) at each 50PIN connector. When a joint is accessing the hard disk, the side of the terminal must be removed. Regardless of the number of hard drives, the SCSI cable tail must be connected to the terminal resistance, and the hard disk of the IDE interface does not have these troubles. When the hard disk setting is complete and installed, turn on the host power, enter the CMOS state, set the primary relationship between the two hard drives in the Standard CMOS Setup option: Set the C disk to the master, the D disk is set to slave, and save the set result . 3. Hard disk partitions and formatted with the fdisk command to partition the slave, pay attention to the same partition from the disk and the main disc, namely: the size of the DOS partition and non-DOS partition should be the same as the primary disc. It is recommended to view the partition information of the primary disc in the case of the slave partition. After the end of the disk partition, the DOS partition from the disk is formatted. The specific operation command is: Format D: / S / V. Note: Be sure to format the slave to a bootable system, so when the primary disk is damaged, you can start the server from the disk. The main disc can copy all the files and directories on the C: on the disk from the partition and formatting of the disk to the slave D: on, so that the information in the two hard disk DOS partitions is identical. Second, the setting of the software 1. Built free space to NetWare Partition When we use DOS's FDISK command to partition the hard disk, the maximum of the hard disk is divided into non-DOS partitions. The size of non-DOS partition is the size of Free Space after entering the network operating system.
Free space is not managed by the network operating system, so it must be converted to NetWare Partition so that the network operating system calls. To do this, INSTALL files can be loaded on the server: Load Install, "Installation Options" main window, select "Partition Tables" option under "Disk Options" in the main window, "Available Disk Drives" list: * Device # 0 (5d001) Seagate 7345-SCSI Device # 1 (5D010) Seagate 7345-SCSI Select Device # 0, showing the WE SPACE space of the main disc has become NetWare Partition. Select Device # 1, then select the "CREATE NETWARE Partition" option in the window, "Partition Information" appears, press the ESC button and select "YES" to determine, at this time, the free space of the second hard disk becomes NetWare Partition, Non-DOS partition of the second hard disk has become NetWare Partition that can be used by the network operating system. 2. Perform a disk image next to the previous step. In Install applications, select the "mirror" option under "Disk Options" will display the unmethatic information as shown below: Not mirrored: non mirrored partition [24bd034a] # 1 not mirrored: non mirrored partition [24bd034b] # 3 After pressing Enter, the "mirrored netware partition" will be displayed: in Sync- NetWare Partition # 3 on Device # 1 Represents the first hard disk # 1 has completed the "in sync" operation, press "INS" button to appear "Available Partitions" window, its display information is: NetWare Partition # 1 On Device # 0 After pressing the Enter key, the "Mirrored NetWare Partitions" window appears, and its display information is: in Sync- NetWare Partition # 1 on Device # 0 in Sync- NetWare Part # 3 On Device # 1 After pressing the "INS" button, the "partition mirroring status" information window appears: Partition Logical Partition [24bd034a] # 1 indicates that the disk mirror is complete. Third, the effect of verifying the disk image is logged in from the workstation with Supervisor, and copies a file (Test.txt) to the server. Let the workstation logout, turn off the server, turn on the chassis and change the primary relationship between the two hard drives, set the original primary disk to slave, set the original slave plate as the primary disk, or simply remove it off The original primary disc will be set from the plate as the primary disk and reset the CMOS.
Start the server, log in from the workstation, see if there is a file (Test.txt) that is copied in front, if there is, explain the mirror success. Another way to verify that the disk mirroring effect is to use the Monitor detector. When you type the load monitor on the server, select the "Disk Information" option of the "Available Options" window, see the "Mirror Status" status If it is "mirrored", it means that the disk image is successful. Fourth, the application of the mirror disk is in the run of the server. When the primary disk does not recover after physical failure or the data corruption in the main disk, the mirror disk (slave) can be changed to the primary disc, and the independent disk is used. The specific method is to remove the original main disc, and set the slave jumper after the slave hard disk C:, start the server after the setting of the CMOS related parameters, run the install program, cancel the original "mirrored", will It becomes "not mirrored" status, and this server will work in a single hard disk. At the same time, in order to prevent the failure, it is recommended to repair damaged hard drives as soon as possible. (Wang Qun, Jiangsu)
