Nowadays everyone is talking about “The Cloud”, but it is actually the dedicated servers, either physical or virtual ones (“Cloud” is actually a form of a virtual dedicated server) which are mostly use computing instances for public or private internal workloads. No matter where your dedicated server hosting instances are located you’ll find 6 things one needs to do on regular basis on their dedicated server infrastructure.
1. Routine OS & Software Upgrades
OS updates, especially those for Linux distributions are quite frequent. Make sure that you have schedule updates. Consider using any software application like Oracle’s Ksplice, which offers Linux security fixes, kernel and user space patching without rebooting the system. Automated patch management tools and monitoring are important as the send alerts when a system is out of date. If you can automate, do it. Don’t rely on manual updates as there is a risk to miss important security updates. Hackers are usually scan systems for vulnerabilities within minutes after any security issue is disclosed. If you cannot automate OS updates, then make sure you have a schedule to update the OS and software updates. It is good idea to have a weekly updates plan.
2. Server Management Software & Application Updates
Most servers used to deliver service for websites, email services, FTP and many other IT services for public use have automation software, known as control panels. Those are very important to be regularly updated. Sometimes those updates require software updates PHP, Perl, Python and other scripts. Make sure that your web server software is also up to date.
Do you know that software applications used for delivery of public services account for more than 95% of all security breaches? So each and every software application installed on your server must be up-to-date. Popular apps like WordPress are very vulnerable. Do not leave them out of date.
3. Storage Partitions & Disk Usage
Never allow any production system to be turned into a storage node. Regularly delete old logs, emails and anything whips is not in use. It is important to keep your servers and systems clean. The less the data, the faster would be the recovery. If your the disk usage exceeds 90% of the storage capacity, you have 2 options - to reduce usage or to increase storage space by adding either virtual or physical storage. When your partition is close to 100%, you risk server failure and database corruption and data loss.
4. Remote Management Tools & Resource Utilization
All appliances and infrastructure colocated in a data center, including any leased dedicated servers, have to be accesses remotely. Make sure you have proper remote management software in place, including an out-of-the-band plan. You obviously need to have a procedure for remote management. It includes the use of tools such as remote console, remote reboot and rescue mode.
Organize the resource monitoring and make sure you always have information about resource utilization (Storage, CPU, RAM, Network). Check Observium, it is a good low-cost IT monitoring application. Sysstat/sysstat is also an option. for Linux based computing instances.
5. Backup & Verify Disaster Recovery Plans
All production system must have up-to-date archive copies and operational backup copies and/or snapshots. Check that your backup work. Run some recovery tests before deleting any kind of data. it is important to have both local archive copy of the data and a remote one. Location and speed is always important.
Any disaster recovery plan should be simple and easy to follow. It is important to have standard procedure part of your business documentation. It should say who does backups, how does one do it, who is responsible to give authorize and approve all operations. Test the disaster recovery plan.
6. Inventory, Supply Chain & Replacement Parts
When you own hardware, it is important to have a good inventory database. The necessary replacement hardware components, should be in place, or based on what your Service Level Agreement says and the way the supply chain is organized, there should be a plan for hardware repair and replacement of failed components.
Make sure that you are buying expertise and service, not just hardware. Your vendor should facilitate servicing of the hardware, servers and appliances they sell.