![]() ![]() Monitoring Tools Have Their PlaceĪutomating health checks should not replace your monitoring tools, they should compliment them. Even without more complex network platforms being developed, existing platforms contain a large amount of information just waiting for a system to analyze it and provide insights that would be difficult for someone to do manually on a repeated basis. To be completely honest I don’t have any experience with platforms providing AIOps, but I’m excited to see how AIOps will evolve and help us in our roles as network engineers. It is used in a few different areas of IT such as security monitoring and application performance monitoring. The ability of ML and AI to parse and correlate data from multiple different sources in realtime is why AIOps is starting to take off. AIOps is the idea of leveraging machine learning (ML) and artificial intelligence (AI) to analyze large data sets to identify trends that may become, or already are, impacting. Any article on automation and health checks would probably be doing itself a diservice without mentioning AIOps. Where does AIOps come in to play?īefore diving into my approach to automated health checks I wanted to briefly bring up AIOps (artificial intelligence for IT operations). I hope to cover each of these goals throughout this post to provide some insight into the benefits of automated health checks. ![]() Because scripts can be executed and completed in less than a few minutes, performing due diligence on these simpler incidents can save your organization in the long run. While automations may seem overkill for a low priority and low impacting incident, you may catch an unrelated issue and be able to resolve it before there is noticeable impact to users and services. ![]() Record a historical snapshot of current stateīy creating automations to help you during critical incidents you also provide significant value to even low priority incidents.Automate the easy and mundane health checks, so you can focus on the complex aspects.An automation should compliment, not replace, our monitoring tools.I believe that it can be a low risk and high reward when it comes to health checks and I often have the following goals when creating automated health checks: When faced with limited time to perform a task you start to look for a more efficient way automation.Īutomation can be a powerful tool and as our understanding and acceptance of the risk versus reward factor increases we will find more ways to incorporate it in our daily lives in network operations. It’s not that I don’t mind this approach, it certainly makes me a better engineer having prove without a doubt that the network is healthy, but we are always faced with prioritizing which checks we perform first and can sometimes overlook a symptom early on in an investigation as the attendees on a 30 person conference call are asking for minute-by-minute updates. Then take into account all of the devices in a network path between a source and destination, of which many of these devices now include rapidly evolving network technologies, and it becomes a painful battle to beat the clock as your business is being impacted by a critical incident.Īs network engineers we’re used to everyone else saying “it’s the network causing the issue” and having to painstakenly investigate every component and provide evidence as to whether or not the network is contributing to the issue the users or applications are experiencing. Health checks on a site router can now consume valuable time in driving the incident to resolution. In the past a few checks of system health (CPU, memory, hardware status, interface status and errors, routing changes) has now evolved to also include app-routing and application aware functions (app routes, SLA’s, BFD, NetFlow, QoS, among many others). Take for example SD-WAN what used to be a relatively simple health check on a router now requires deeper knowledge of the specific traffic traversing the device. ![]() The increasing complexity of network infrastructure devices has drastically changed what is required to perform a thorough health check to ensure it is operating correctly and efficiently. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |