A well-maintained server can be the perfect IT labour, working day-in, day-out without kicking up a fuss. Simply serving web pages without complaint. However, ensuring that your server operates that smoothly involves a degree of regular monitoring. And doing a couple of things that prevent various server failures. You don’t want to be in a position where you haven’t made a backup for years. Or where major problems, such as hard drive failure, has gone unnoticed for months.
Though managed servers mean you don’t need to worry about these server maintenance issues, they’re not always practical. For any number of reasons you may need to manage your own server. And if that’s the case, we suggest you pay attention to the following checklist. Tick these off the list and you’ll have a reliable server.
1. Manage your storage
Heaps of server failure is storage related. Here are some of the most important storage aspects you should keep an eye on.
Check disk space.
Servers can fall over simply because your machine has run out of disk space while no one noticed. It’s not just how much space remains for that extra log or cache file. You should also clear out your disk in order to avoid security problems. Because old applications can invite hacks that disrupt your server.
Also keep in mind that a small server footprint ensures that your backups are small, and that recovery is quicker. When doing server maintenance be particularly mindful of partitions that exceed 90% in use. Because this is the point you should start freeing up space. At 100%, your server can simply stop responding. The result? A corrupt database and a big restoration effort.
Check your RAID.
RAID is intended to give you a reprieve if a drive fails. But if you never replace that drive, RAID will not continue to offer protection. Since multiple drive failures can lead to data loss. However, if you quickly replace a drive, your RAID facility can repair the volume without data loss.
For this reason, you need to monitor your RAID status when doing server maintenance. So you can quickly act in case of disk failure. Hence, fixing the problem before a second disk failure causes you to loose data on your volume.
Check your backups.
Just because you set up your backups years ago doesn’t mean they still work flawlessly now. You need to actively restore your backups to a machine to see if you can recover from dramatic failure.
In addition, when doing server maintenance, check your backup logs regularly for any errors. They may indicate a backup has failed due to a broken device. Or a connection doesn’t work as you expect. This helps just in case you need that last-minute backup to recover something.
2. Updates, updates, updates!
We’ve all read how important it is to keep your software up to date, and this counts for your server too. Here’s what you need to keep in mind:
Update your server control panel. As part of your regular server maintenance tasks you need to check that your hosting control panel is running the latest version. Keep it up to date and you avoid all kinds of vulnerabilities. Skip it, and it could provide an entry point to hacks.
Note that you may need to perform some manual updates too, for example if you update WHM and cPanel you need to update the PHP version on your machine to make sure some of the known issues are fixed as only updating the control panel won’t update PHP or the web server you use.
OS and application updates. You’re probably using Linux for your server. Linux is frequently updated and it can be tricky to keep ahead but these updates are important and you should try your best to keep your server up to date as part of regular server maintenance. Automatic tools help so that patches are not delayed, avoiding the risk of a zero-day vulnerability getting exploited.
Rapid response is key and is also crucial when keeping your applications up to date. Web apps are the source of a lot of security breaches and just like your OS should regularly be updated to make sure you’re not victim of a hack.
3. Remote hardware server maintenance
Disk space aside, you should regularly check in on your server to make sure it is running optimally. You can monitor server hardware remotely, so don’t wait around – log in every day to ensure there are no glitches you don’t know about:
Check for server hardware problems. Are you running on a backup power supply all of a sudden? Better get your main PSU replaced before the second one fails too. You’ll only get the opportunity to do so if you check in for server hardware problems as part of routine server maintenance.
Another server hardware aspect you should keep an eye on is network, RAM and CPU utilisation. If your server does not have the resources to service requests those requests will be delayed or simply not served. Nobody likes slow pages or error pages, so consider installing something like systat that can help you monitor your machine.
Verify your remote management tools. Yes, you can perform server administration and server maintenance remotely, but don’t get caught by a tool that does not work. Many sysadmins rely on managing servers remotely and an up and running remote monitoring tool is core to getting their work done.
Remember to also check functions that you don’t use every day. Just because you only use a certain function once a year does not mean you may not urgently need it at some point. Check that you can access rescue mode and remote boot if and when you need to for server management.
4. Actively manage credentials and security
Credentials are like keys to the premises and should be treated as such, yet often when doing server management sysadmins can be quite relaxed about credentials. Don’t be, instead:
Review accounts. The assistant you employed five years ago should no longer have access to your machine, but you may forget to disable their credentials if you don’t regularly review who has access. Clients, developers and the like should all get their access revoked on time.
Also note that storing data from clients who are no longer working with you is not a good idea as it brings both a legal and a security risk. As part of server maintenance but depending on your service contracts you should consider clearing out user data which is no longer needed.
Passwords and security. Users are lazy when it comes to passwords and determined actors can steal and abuse credentials. That’s why changing passwords regularly is important: consider doing so at least every 6 to 12 months.
Indeed, as part of server maintenance you should take a regular review of your server security, tools such as Nessus can help. These audits can cover your server configuration, check whether you have performed OS updates and generally ensure that your server security efforts have not missed a beat.
Effective server maintenance: Every step you need to take2019-01-162019-01-16https://www.plesk.com/wp-content/themes/plesk/images/plesk_logo_144x144_black.pngPleskhttps://www.plesk.com/wp-content/uploads/2019/01/header-blog_1920x400-server-maintenance-basics.png200px200pxPlesk is available in Cloud Server Plesk.Contact us to find out our latest offers!