4.3 Given a scenario, troubleshoot hard drives and RAID arrays with appropriate tools
Important RAID (Redundant Array of Inexpensive Disks) details are given below:
RAID 0: This has striping but no redundancy of data. It offers the best performance but no fault-tolerance.
RAID 1: This is also known as disk mirroring. It consists of two or more drives that duplicate the storage of data. There is no striping. Read performance is better since either disk can be read at the same time.
RAID 2: It uses disk striping across disks with some disks storing error checking and correcting (ECC) information. RAID 2 is hardly used because it is expensive and does not provide fault tolerance like other forms of RAID levels.
RAID 3: It uses disk striping and dedicates one drive to storing parity information. The embedded error checking (ECC) information is used to detect errors.
RAID 4: Block-level striping with dedicated parity. RAID-4 offers no advantage over RAID 3
RAID 5: Uses block-level striping with distributed parity. Thus, all read and write operations can be overlapped. RAID-5 stores parity information but not redundant data (but parity information can be used to reconstruct data). RAID 5 requires at least three disks for the array. It's suitable for multi-user systems in which performance is not critical.
If the hard-disk is making sound, the most likely problem is that the hard disk read/write head is scratching the disk surface. It often results in the total failure of the disk. If you find that you can still read/write to the disk, backup the hard disk and replace immediately.
A hard disk should never be low level formatted at the customer premises. It is highly recommended that it is done at the manufacturer's or at any authorized center.
It is very cumbersome to change the partition sizes, once the hard disk is partitioned and used. It may require backing up all the data and restoring after repartitioning.
It is recommended that the backup tape is stored at a location away from the building where the backup was taken. For most companies, backups contain important data and loosing backups may affect the continuity of one’s business. If a backup is stored in the same building, it may get damaged in fire or any other natural calamities along with the computers. As a result, both the server, as well as back fail at the same time. Therefore, it is recommended to store the backup at a different location.
Some of the important System Monitor counters are:
Memory>Available Mbytes: measures the amount of physical memory that is available. Typically > 4MB. If less than 4 MB, consider adding more memory.
Memory>Pages/Sec: Shows the number of times that the disk has been accessed, because requested information was not available in memory. If the value of the counter is not below 20, you should add more memory. A value of 4 or 5 is typical.
Paging File>%Usage: Indicates the % of allocated page file utilization. Should be less than 99%.
Processor>%Processor Time: measure the time that the processor is busy. Should be typically less than 80%
Processor>Interrupts/Sec: Indicates the average number of hardware interrupts that the processor receives each second. If more than 3,500, you can suspect a program or faulty hardware.
PhysicalDisk>%Disk Time: Measures the amount of time that the physical disk is busy servicing read or write requests. If more than 90%, you can improve the performance by adding another disk channel.
PhysicalDisk>%Current Disk Queue Length: indicates the number of pending disk requests that need to be processed. The value should be less than 2. The disk problems might arise from less memory, resulting in usage of excessive paging. Ensure that the memory is sufficient before attending to the disk problem.
LogicalDisk > %Free Space counter: Indicates the amount of logical disk’s free disk space. Typical value is 10% or above.
Copyright © Anand Software and Training Private Limited.