Reacting to website failure: A foolproof process (infographic)
Posted February 12th, 2018 in Infographics.
Tagged: detection, escalation, failure, infographic, monitoring, process, verification.
Faultless websites, online stores or web apps simply don’t exist. Despite an exemplary infrastructure, duplicated systems and excellent mechanisms, something can always go wrong.
We offer you a proven process that you can apply in a crisis situation – adapting it accordingly to your conditions, of course.
You can publish our infographic on your blog or website. Use the following code to do this.
<a href="https://super-monitoring.com/images/reacting-to-website-failure.png" target="_blank"><img src="https://super-monitoring.com/images/reacting-to-website-failure.png" alt="Reacting to website failure: A foolproof process (infographic)" style="max-width:1200px;width:100%;" /></a><br /> <br /> Source: <a href="https://www.supermonitoring.com/blog/reacting-website-failure-foolproof-process-infographic/" target="_blank"><em>Reacting to website failure: A foolproof process (infographic)</em></a>
Text version of the process
| When you aren’t using monitoring | When you are using monitoring |
|---|---|
1. Detecting a failure. | |
| You notice a problem with the website on your own or – even worse – you learn about it from the users. In any case, some time passes from the occurrence of the failure. | You receive an alert via email and/or SMS as soon as the failure occurs. |
2. Failure verification. | |
| Make sure that the failure has actually occurred – and it’s not a local problem on your end. To confirm that the website is unavailable you can use, e.g. www.website-down.com. | Monitoring will alert you only after it confirms the failure from several independent locations. |
3. Checking the type of failure. | |
| If you aren’t an IT specialist, you can make a preliminary diagnosis based on the symptoms you are seeing – e.g. messages displayed by the browser: a) “This page could not be found” / “The IP address could not be found” – the domain has expired or an error has occurred in the configuration of the DNS (i.e. the server assigning domain addresses to IP addresses); b) “Connection timed out” / “The server took too long to respond.” – connection to the server could not be established; c) “Connection is not secure” / “Connection is not private” – the SSL certificate has expired or an error has occurred in its configuration on the server; d) ” 500″ / “Internal server error” – something is wrong with the server’s configuration; e) “503” / “Service unavailable” – most often the server resources are depleted (server is overloaded); f) “404” / “Not found” – the sub-page or file with the given address doesn’t exist; g) “403” / “Access denied” – something is wrong with the given location’s access rights; h) empty page templates are displayed, with no content – most likely a database failure; i) one of the website’s features doesn’t work – e.g. navigation, button, form. | The email alert sent by the monitoring system contains information about the type of failure. Moreover, in the user panel, you will find additional data – such as the HTTP header, HTML content, screenshot, cascade diagram or HAR file. |
4. Escalation. | |
| Depending on the supposed type of a failure, report the problem to the appropriate specialist: – cases a) to e) – server administrators – cases f) to i) – programmers responsible for the website. | You can configure the monitoring application so that it also sends alerts directly to administrators and programmers. |
5. Information for the users. | |
| Do not try to hide the failure from your users as it can only harm you. Inform them immediately about: – the occurrence of a failure – failure being fixed – prolonged unavailability. After the failure has been fixed, explain why it appeared and what measures you are going to take to ensure that the situation does not occur again. | You can prepare a page for your users that will inform them about the status of your website, displaying the current status of the failure taken straight from the monitoring system. You can also use this page to post messages about the progress of works aimed at removing the failure. |
6. Conclusions. | |
| You can learn something from any failure. Both in the case of systems that failed and procedures for dealing with critical situations. Let every such event help you reduce the likelihood of future failures. | |
