In a hard disk, the read and write operations are managed by Disk I/O. The speed of the server in reading and writing information to disk straight away affect one’s server performance and the performance of cPanel and WHM. A load of one’s server increases if the system struggles with high disk I/O wait time.
Problems Attacking the Disk:
There are nearly
seven problems that attack the high disk I/O. They are:
High Server Load:
Where the average system load exceeds 1.
Where one gets notifications about an offline service or it is conveyed that the system cannot restart a service.
Slow Hosted Websites:
Whereby the hosted websites may take more than a minute to load.
Slow Delivery of Email:
Due to which the Exim service performs comparatively slowly or does not respond.
Slow Connection for Email:
Hampering the POP or IMAP services, where they perform comparatively slowly or do not respond.
Slow Webmail Interfaces:
Where it performs slowly or do not respond as it should (for example Roundcube, Horde, or SquirrelMail).
Slow WHM or cPanel Interfaces:
Which results in the slow performance of the WHM or cPanel interfaces when one adds email accounts, databases, or other items.
If there are problems, surely there are solutions even. We can combat the problems with certain valuable recommendations. It will be easier if we discuss the problems and the recommendations side by side.
How to Determine the Disk I/O Wait on One’s Server
One needs to use the top command to find the average wait time on one’s server. The %wa statistic at the top of the output will indicate one’s server’s average disk wait.
The CPU cores must wait to process data on hard disk if the I/O wait percentage is greater than one divided by the number of your CPU cores. As for example, if the system possesses four CPU cores and the server %wa statistic is 8.0, then the actual %wa is 2.0. Because of the actual %wa larger than 1.0, the CPU cores must wait before they can process data on hard disk.
Then one can use the sar command to determine the history of the server’s disk I/O wait.
The sar command provides the history of the server’s load averages. Using this command, one can determine how many times the server experienced high disk I/O.
Problems and Recommendations:
- hard disk specifications with low RPM speed or slow interface technology, one can upgrade the hard disk on their server or split the application load between separate hard disks.
- If there is a problem of no availability of bandwidth on the hard disk, one can upgrade the hard disk on their server or split the application load between separate hard disks.
- If one faces the problem where write caching is disabled, it is recommended to enable write caching on the disk. This will solve the problem.
- The problem of degraded RAID array can be solved by checking the RAID array for a hardware malfunction. One should positively test and verify the hardware.
- When stuck in a problem where software RAID array on the server reports are busy and CPU uses slow parity calculation, one can arrive at a solution by checking the RAID array for a hardware malfunction. Here also, the hardware should be tested and verified.
- Lastly, if the software processes are occurring slowly, then one should upgrade the hard disk on their server or split the application load between separate hard disks.
Disk I/O is an integral part of any organization infrastructure. Any outage could result in loss of crucial time and data. This interface helps administrators manage and monitor the Disk I/O.