My website is down.
It’s something developers, or anyone responsible for the performance of a website, hate to hear.
Once you have confirmed the website actually is down, you’ll need to embark on a mission to get to the bottom of why the server has failed, and this can take some time.
Fortunately, there are some ways you can reduce server downtime…
Understand the cause
I know what you’re thinking…
“So, the first of your six ways to reduce server time—your opening gambit, Nimbus is ‘understand what’s causing it. Wow, great insight. I’m glad I clicked on this.”
Bear with me.
Of course, you need to understand the cause before you can fix the issue, but by identifying and documenting the causes of server downtime you can start to anticipate those problems and more easily avoid them in the future.
Recognise that a particular component is failing every six months? You need to be prepared to replace that component before that sixth month comes around. Realise that your client keeps spilling their coffee all over the PSU? Buy them a sippy cup.
Understanding the causes of issues is crucial for avoiding them in the first place.
Have a robust go-live process
What’s your process for implementing software updates, upgrading hardware or adding additional devices to your network? Do you even have a process?
Well, you need one. Here are some tips to get you started:
- Never roll anything out at the end of the day—unless you want to be staying at the office until the early hours bug fixing.
- Never on a Friday—if something goes wrong shortly after implementation, there’s a good chance it won’t be resolved until the following Monday. Go to the pub instead.
- Don’t implement during peak hours—if you’re aware of certain times of day when server traffic is at its highest, this is not a good time to roll out changes.
- Create a separate test environment—you should always test any software updates in a dedicated environment. That way, if it causes any problems, you can work out why and fix it before implementing it in the live environment.
This is probably the most common cause of server/website downtime.
There can hardly be anything more frustrating than putting in the time and money to develop a fantastic website, only to go with poor quality hosting.
Most of the time, this will be because you’re on a cheap ‘shared’ hosting platform. You know the deal. Your beautiful website is crammed into a server with potentially hundreds of others. It’s like putting on your best dress for a party hosted in your mate’s shed.
Switch up to cloud hosting (a.k.a cloud VPS hosting) or dedicated hosting. Trust us, it’ll be worth the investment.
You’re stuck in a ‘traffic jam’
Getting loads of traffic to your website is great, but not if it starts causing website slowdown or a full on crash.
Again, this is going to be much more common if you’re with a shared hosting provider. If you start to exceed your allocated server processes, number of connections, database queries and so on, your provider might just temporarily take your site offline. Why? To protect everyone else who is sharing the server with you. The only thing they don’t put a limit on is how ****** off you can get with them.
Again, the solution here is to go with a virtual private server (VPS) or dedicated hosting provider *cough*Nimbus*cough*. Note, though, that even if you do have your own VPS the server can go down if you’re maxing out your current resources. If so, you’ll need to consider an upgrade in order to handle traffic spikes.
Set up monitoring
Unfortunately, even with the very best providers, server downtime is an inevitability. However, with the help of server uptime and performance monitoring, you can quickly identify when issues arise, tackle them quickly and, as a result, reduce network downtime.
Sick and tired of your hosting provider letting you down? Get in touch with the team here at Nimbus, or simply check out our solution for free.