Summary
This chapter described the basic concepts behind RAID and the Solaris Volume Manager (SVM). This chapter described the various levels of RAID along with the differences between them, as well as the elements of SVM and how they can be used to provide a reliable data storage solution. We also covered the creation and monitoring of the state database replicas and how to mirror and unmirror file systems. Finally, you learned about Veritas Volume Manager, a third-party product used predominantly in larger systems with disk arrays.
Key Terms
Exercises
10.1. Monitoring Disk Usage
In this exercise, you'll see how to use the iostat utility to monitor disk usage. You will need a Solaris 10 workstation with local disk storage and a file system with at least 50 Megabytes of free space. You will also need CDE window sessions. For this exercise, you do not have to make use of metadevices because the utility will display information on standard disks as well as metadevices. The commands are identical whether or not you are running Solaris Volume Manager. Make sure you have write permission to the file system.
Estimated Time: 5 minutes
1. | In the first window, start the iostat utility so that extended information about each disk or metadevice can be displayed. Also, you will enter a parameter to produce output every 3 seconds. Enter the following command at the command prompt:
| 2. | The output will be displayed and will be updated every 3 seconds. Watch the %b column, which tells you how busy the disk, or metadevice, is at the moment.
| 3. | In the second window, change to the directory where you have at least 50 Megabytes of free disk space and create an empty file of this size, as shown in the following code. My example directory is /data. The file to be created is called testfile.
cd /data
mkfile 50M testfile | 4. | The file will take several seconds to be created, but watch the output being displayed in the first window and notice the increase in the %b column of output. You should see the affected file system suddenly become a lot busier. Continue to monitor the output when the command has completed and notice that the disk returns to its normal usage level.
| 5. | Press Ctrl+C to stop the iostat output in the first window and delete the file created when you have finished, as shown here :
|
Exam Questions
| 1. | Which of the following is a device that represents several disks or disk slices?
| | 2. | Which of the following provides redundancy of data in the event of a disk or hardware failure?
| | 3. | Which of the following types of addressing interlaces component blocks across all of the slices?
| | 4. | Which of the following volumes organizes the data sequentially across slices?
| | 5. | Which of the following devices was created using Veritas Volume Manager?
| | 6. | In Veritas Volume Manager, which of the following is a group of physical slices that appear to the system as a single, logical device?
| | 7. | Which of the following is a collection of slices reserved to be automatically substituted in case of slice failure in either a submirror or RAID 5 metadevice?
| | 8. | Which of the following is a set of contiguous disk blocks; these are the basic units in which the Veritas Volume Manager allocates disk space?
| | 9. | Which of the following replicates data by using parity information, so that in the case of missing data, the missing data can be regenerated using available data and the parity information?
| | 10. | Which of the following has an eight-partition limit per disk?
| |
| |
| |
| Standard Solaris SPARC disk
|
| |
| | 11. | Which of the following commands would create 3-state database replicas on slice c0t0d0s3?
| |
| metainit -a -f -c3 c0t0d0s3
|
| metadb -a -f -c3 c0t0d0s3
|
| |
| | 12. | Which of the following commands would create a one-way mirror (d1), using metadevice d14 as the submirror?
|
Answers to Exam Questions
| 1. | B. A volume (often called a metadevice) is a group of physical slices that appear to the system as a single, logical device. A volume is used to increase storage capacity and increase data availability. For more information, see the "Solaris SVM" section.
| | 2. | A. A mirror is composed of one or more simple metadevices called submirrors. A mirror replicates all writes to a single logical device (the mirror) and then to multiple devices (the submirrors) while distributing read operations. This provides redundancy of data in the event of a disk or hardware failure. For more information, see the "Solaris SVM" section.
| | 3. | D. A stripe is similar to concatenation, except that the addressing of the component blocks is interlaced on all of the slices rather than sequentially. For more information, see the "SVM Volumes" section.
| | 4. | C. Concatenations work in much the same way as the Unix cat command is used to concatenate two or more files to create one larger file. If partitions are concatenated, the addressing of the component blocks is done on the components sequentially. The file system can use the entire concatenation. For more information, see the "SVM Volumes" section.
| | 5. | A. The names of the block devices for virtual volumes created using Veritas Volume Manager are found in the /dev/vx/dsk/<disk_group>/<volume_name> directory. For more information, see the "Veritas Volume Manager" section.
| | 6. | B. A volume is a virtual disk device that appears to applications, databases, and file systems like a physical logical device, but does not have the physical limitations of a physical disk partition. For more information, see the "Veritas Volume Manager" section.
| | 7. | A. A hot spare pool is a collection of slices (hot spares) reserved to be automatically substituted in case of slice failure in either a submirror or RAID 5 metadevice. For more information, see the "Solaris SVM" section.
| | 8. | C. A set of contiguous disk blocks, subdisks are the basic units in which the Volume Manager allocates disk space. For more information, see the "Veritas Volume Manager" section.
| | 9. | D. RAID 5 replicates data by using parity information. In the case of missing data, the data can be regenerated using available data and the parity information. For more information, see the "RAID" section.
| | 10. | D. In a standard Solaris SPARC disk-partitioned environment, there is an eight-partition limit per disk. For more information, see the "Solaris SVM" section.
| | 11. | C. The command metadb -a -f -c3 c0t0d0s3 would create the required state database replicas; see the "Creating the State Database" section.
| | 12. | B. The command metainit d1 -m d14 would create a one-way mirror; see the "Creating a Mirror" section.
|
 |
