CompTIA® Server+ Exam Notes : Troubleshooting Software Problems

7. Troubleshooting

7.3 Troubleshooting software problems

The following are good practices that are recommended to be followed for securing a server

1. Password protecting the Servers

2. Lock the Servers room, or limit access to the room to authorized server administrators only.

3. Lock the cabinet that the server is mounted or sitting in if you cannot lock the actual room.

4. Locate all the server back-ups, the server, and the emergency recovery disks at the same place for convenience.

You want to prevent your server from being broken through the network. This can be prevented by using:

1. Appropriate firewall to limit access to the server

2. Choosing a difficult to guess password, so that it is difficult to guess.

3. Checking the logs periodically to monitor if anybody tried to login to your server without any success, so that you can take appropriate measures.

OS boot failure: The message "Missing operating system" occurs when the system found a boot device, but could not boot because operating system files are missing or corrupt.

The message "No boot device available" appears when the bootable drive itself is not found. In such case, both floppy and hard disk drives are faulty or not present. It is important to be familiar with various boot time messages and probable causes.

You need to follow the manufacturer's recommendation regarding updates. Microsoft recommends that you enable automatic updates on your client as well as server operating systems so that the latest patches and fixes may be applied.

A network mapper is a tool that identifies what the devices connected to the network and the operating systems being used, if any. Firewall, proxy server, and web security gateway are used for network/host security. System mapper is given to divert the attention from the basic question.

Software backout: Software backout procedure is a term used for restoring the system or software in the event of any recent failure due to upgrade.

Memory leaks are bugs where the application fails to release memory when no longer needed. Over time, memory leaks affect the performance of both the particular application as well as the operating system. A large leak might result in unacceptable response times due to excessive paging. Eventually the application as well as other parts of the operating system will experience failures.

Performance Monitor and Resource Monitor can monitor and graph resource use over time. You may use these tools and a few other tools Application Verifier to find the problem.

Common Problems:

A runaway process is a process that consumes one or more processors, causing your system to become sluggish and sometimes causing other applications to freeze or crash. Often you can find such problems quickly by using Task Manager to view information about running processes. The common way to solve a runaway-process problem is to just kill the process, but this is only a temporary fix; the runaway process will likely recur. To troubleshoot processes that consume large amounts of CPU clock cycles, two tools by name Process Explorer and Adplus are useful.

In Linux, you may use the command "kill <process ID>" to kill the offending process. Alternatively, you may use the System Monitor GUI to kill the process. It looks a lot like the Windows Task manager.

When a system admin finds that a required service is not starting, there are two things to do:

1. Manually Start the Service: If you notice that a required service is not running, then the first thing that you should do is to attempt to manually start the service. To do so, just right click on the service and choose the Start command from the resulting shortcut menu. It is possible that the service starts without any problems. You can check the Dependencies tab and verify if the Service depends on another Service or Services, and troubleshoot for those Service(s).

2. Check the Event Log: If the manual start fails, the next best thing to do is to check the Event Viewer. If there are hundreds of events, then you use Event Viewer's find feature to locate the event of interest. When you find a log entry related to your problem, just double-click on the entry to view it. If the Event Log doesn't provide specific information, you can use the information provided and search Microsoft Technet for solution.

Over a period of time, it is possible that a Server computer (for that matter, even a PC) becomes slow in performance. This might happen due to various reasons, including the following:

1. The disk is almost full with junk files, temp files, and other un-necessary programs. This may be fixed by running disk clean-up software.

2. Defragment: It is possible that the hard drive has been fragmented considerably over a period of time. By defragmenting the drives, the performance may improve.

3. OS updates: Some times, by installing the latest Operating System updates, OS performance is improved.

4. Corrupt files: It is possible that some process has got corrupt due to some or other reason. You may check if any process is taking a lot of system resources by using Task Manager. You may need to end the process till it is resolved.

5. Virus and other malware: Last, but not the least, it is possible that some virus or malware has infected your server computer. Scan the computer from time to time to ensure that the system is free of any malware.

Cause of common problems

Corrupt or missing file(s): When the operating system fails to start and the hardware is not a fault, it is because of one or more essential files that the operating system needs to start have come corrupt or missing. Common error messages include an invalid boot disk and inaccessible boot drive or missing BOOTMGR. In Windows Server 2012 R2, the causes are most likely due to corrupted system file, BCD, boot sector, or master boot record (MBR).

There might be various reasons for a user not being able to access a resource on the network. This includes the following:

1. The admin has not given sufficient rights to access the resource

2. The user computer is not configured properly. For example, improper DNS server address or the Default gateway address or a simple IP address conflict.

3. The DNS server itself may not be functioning at that point of time.

4. If using an AD, check if the AD is working, and that the user is given sufficient permissions.

A patch is an important component of software update and it usually fixes some known bugs. If you need to apply it on a server, where the downtime is critical, it is recommended practice to install the patch on a lab server first. Install it on the production server once you find it satisfactory.

Previous   Contents   Next