Juan Pablo Castro’s Post

For years, my career in the managed hosting industry revolved around one critical mission: ensuring uptime. It was all about meeting high standards for RTO and RPO—how quickly services could be restored and how much data loss was acceptable. The focus was on high availability, clustering, and geo-dispersed data centers, building applications designed to endure. That’s where those coveted 99.99% SLAs stem from. A huge part of achieving this was relentless testing—failovers, backup and restore processes, planned and unplanned outages, simulations—you name it. In this industry, testing isn’t optional; it’s a way of life. Every test is about identifying risks and mitigating them before they become real problems. So, when it comes to security, why should it be any different? Sure, a robust platform is essential, but the key to true security is testing your defenses, over and over again. We do Red- and purple teaming at Trend Micro and it will really help you to identify some of that risk. Get in touch!

Here is Forrester’s perspective of the CloudStrike global outage and what to do next!

In today's rapidly evolving digital landscape, network storage solutions have become essential for businesses and individuals alike. These systems offer secure, scalable, and efficient ways to store and manage vast amounts of data. Explore the critical role of network storage solutions in enhancing data accessibility, security, and overall IT infrastructure. #NetworkStorage #DigitalAge #NASStorage #DataSecurity #DigitalAccessibility #NetworkAttachedStorage #SecureStorage #scaleoutNAS #scaleoutstorage https://lnkd.in/dBYuZPfZ

Discover how Guidehouse Technology Solutions can modernize your organization’s technology and digital framework to mitigate security risks and reduce costs.

#SecurityAssessment #Top10 questions #TrustEverybodyButCutTheCards #Technology stacks change too quickly to expect anyone to keep up with all of the security concerns, but good #securityprinciples stand the test of time and all we have to do is apply them. With each new technology new security concerns arise constantly. Nevertheless any new infrastructure paradigm can be secured thanks to: • #network #segmentation • #workload separation • #usermanagement • #accesscontrol The key questions are: 1- #Where does the infrastructure live? 2- #Who has #access to the infrastructure and at what level? 3- Is the infrastructure #managed, by who, and at what cadence? 4- What workloads are running on the infrastructure? 5- Do you have #sensitive workloads running on the same infrastructure as non-sensitive workloads? 6- At what #permission level do the workloads run? 7- What can the workloads access within and outside the infrastructure? 8- What #users can access the workloads? 9- What level of #access do they have to the workloads? 10- How do we ensure #availability of the infrastructure? If the system owner can’t answer these questions, then there is a different set of problems that need to be addressed first. If the system owner can’t be identified, good luck. Source: https://lnkd.in/dNznX3V2 #TrustEverybodyButCutTheCards

Visibility into the health and performance of cloud-based systems is crucial for optimizing resource allocation and ensuring seamless scalability to meet growing demands. #CloudPerformance #Scalability #TechConsulting https://lnkd.in/eNJzP7RU

Răzvan Moise touches in this article the topic of Management of Critical Infrastructure. This comes in a difficult period marked by the recent global tech outage. You can read it here:

CrowdStrike Hardware failure can be mitigated by provisioning resources across data centres in one region or multi-region. And making them on standby or running partially in case of any event(Active-Passive), or they can be running at full capacity across regions or centres(Multi-site / Active-Active). But I think all these are too hyped. They lead to over-provisioning etc. I often ask myself, how often does AWS or any cloud provider announce that their entire Data centre is down, or their entire region is down? Rarely happens, isn’t it? Now let’s talk about distributed applications running on the resources you provisioned.  Applications can be written in on one of the languages, Servers, DBs, OSs etc. There are several strategies to utilize when introduce updates in order to avoid a disaster. When a new update is released , no matter who released it, even if it’s Microsoft, you don’t have to trust it 100% and quickly apply the updates to all applications. You can employ following strategies. 1. In-place deployments - applying updates in place on running machines. Although this approach does not imply infrastructure re-provisioning or change, it is dangerous. Because you perform updates to the running application which is a disaster. This affect the availability of the application. 2. ROLLING updates - introduce the updates to several selected applications and give them some time to monitor whether there are no unintended side effects. Maybe several minutes, maybe several hours etc according to the risk based on RPO (Recovery Point Objective) and RTO (Recovery Time Objective). Then you can introduce updates to additional apps if it goes well or rollback. 3. ROLLING with additional batches - introduce new machines/instances, then run the application with new updates on the new machine/instance. If all goes well, apply the updates to some of the machines. Then roll it on. 4. Blue/Green - create a complete new environment run updates on the new environment , when testing is healthy, switch to new environment 5. Canary deployment, incrementally deploy updates and monitor the performance.

The infamous Microsoft-CrowdStrike outage last July, revealed the fragility of our modern digital infrastructure. 🏮 A faulty update disrupted critical systems worldwide, leading to over $5 billion in losses for Fortune 500 companies. 😱 🔍 What Can Be Done? observability platforms might hold the key to preventing such disasters. 🎙️In an interview with AIM, Sawaram Suthar, Founding Director of Middleware , emphasizes: "Real-time feedback mechanisms in observability solutions notify teams immediately, reducing the mean time to detect and respond. This proactive approach allows teams to resolve performance issues before they escalate into full-blown outages." 🔽 Key Benefits of Observability 🔽 🔷 Early Issue Detection: Real-time insights help engineering teams tackle problems quickly. 🔷Unified Data View: Centralized logs and metrics streamline root cause analysis. 🔷Predictive Analytics: AI-driven models foresee potential issues, enabling preemptive actions. Understanding the potential of observability tools can significantly mitigate risks, ensuring that businesses remain resilient in the face of technical challenges. ➡ Read the full story here: https://lnkd.in/gcy2yVgQ 🙌 A massive shoutout to Pritam Bordoloi for featuring us. 🙌 #Observability #ITOutages #Middleware #DigitalInfrastructure #AIOps #Monitoring #Microsoft #Crowdstrike

Thanks, AIM, for featuring my thoughts on how Observability Can Help Prevent Another CrowdStrike Outage. I know it's a bit late, but it is good to share as Microsoft still didn't come out from the outage 🤣 #observability #crowdstrike #outage #microsoft