Catalin Petru Ciubotariu

Do you agree with Christophe about working on the helpdesk? 🎙️: Careers in the Cloud - EP32: Christophe Troessaert: Go From IT Novice to Freelance Cloud Expert Watch on YouTube: https://lnkd.in/ePPa9piM #TechPodcast #CareerAdvice #CareersInTheCloud

10

1 Comment

"DNS Belgium wants to eventually remove its infrastructure from AWS and migrate to a European cloud provider. The administrator of the .be, .vlaanderen and .brussels internet extensions wants to further optimise its services in this way. At the same time, the organisation wants to set an example for others to no longer be dependent on non-European service providers in geopolitically turbulent times." https://lnkd.in/eypxvHJu

883

57 Comments

🔍Data Professionals - Azure or AWS? Here's what I'm seeing… ☁️⚙️ When speaking to candidates within my network, we often discuss the cloud technology that they are more familiar with. In Belgium, companies tend to use either Azure or AWS so for more junior profiles, it can be a tough choice deciding which cloud technology to focus on. From a hiring perspective, here’s what I’m noticing in the market: 🔷 Azure is gaining traction in enterprise environments — especially where there's a strong Microsoft stack already in place. Roles focused on Synapse, ADF, Power BI, and Purview are in demand, especially in finance, government, and healthcare. 🟧 AWS still dominates in tech-first and product-driven companies. Redshift, Glue, EMR, S3 - these tools come up again and again in data engineering and architecture interviews. Flexibility and scale are the main draws here. 👥 From my experience hiring managers are open to multicloud experience, but deep expertise in one (not just surface-level knowledge of both) still wins interviews. So I’m curious: 👉 Which cloud are you working with most in your data projects? 👉 Have you seen hiring trends shift in your region or industry? 👉 Are you seeing more orgs embrace multicloud or stick with one provider? Would love to hear your take — whether you're hiring, job hunting, or deep in the weeds building data platforms.👇

12

💡 What is an Application Load Balancer (ALB) in Azure? In Azure, it’s called the Azure Application Gateway — a Layer 7 (Application Layer) load balancer. It helps to distribute web traffic (HTTP/HTTPS) across multiple servers or web apps — so that no single server is overloaded. ⸻ 🧠 Simple Analogy: Think of it like a traffic police officer standing at a busy junction: • Cars = user requests (like people visiting your website) • Roads = your web servers or app instances • The officer (Application Gateway) directs cars to the least busy or healthiest road to keep traffic flowing smoothly. ⸻ ⚙️ How It Works — Step by Step: Let’s say you have a website like www.myshop.com hosted on multiple servers. 1. User Request Comes In A customer types www.myshop.com in their browser. 2. DNS → Application Gateway The DNS directs the user’s request to the public IP of your Application Gateway (the load balancer). 3. Application Gateway Checks Rules It looks at: • URL (e.g., /products, /login) • HTTP headers • Hostname Based on these, it decides which backend server or pool should handle it. 4. Health Probe Check Before sending traffic, it keeps checking all backend servers to ensure they’re healthy and running (like ping tests). If one is down, traffic is not sent to it. 5. Request Forwarded to Backend Pool It forwards the request to one of the healthy web servers or app services (based on load-balancing rules). 6. Response Sent Back The selected server processes the request and sends the response back to the user through the Application Gateway. 💻 Example Use Case: You run an online shopping site with: • Frontend: Application Gateway • Backend Pool: 3 web servers running the site When 10,000 customers visit: • The gateway splits the traffic evenly (e.g., 3,333 requests per server) • If one server fails → traffic automatically redirected to healthy ones • WAF protects your site from malicious requests #Azure Networking #Cloud #Networking

8

A ransomware attack just cancelled 70 surgeries in Belgium. Yesterday. AZ Monica Hospital shut down every server at 6:32 AM. All scheduled surgeries? Cancelled. Emergency department? Reduced capacity. Ambulances? Diverted to other hospitals. Seven critical care patients had to be transferred by the Red Cross. Because their conditions were too unstable to risk. The hospital can't access a single patient record. Not one. Here's the part that should terrify every healthcare executive... Healthcare organizations now face an average of 2,620 cyberattacks per week. That makes it the most targeted sector in Belgium. And attacks on Belgian organizations have doubled since 2020. The problem? Hospitals run on legacy systems. They can't afford to take them offline for updates. They're dealing with life-and-death situations 24/7. So they skip patches. And hackers know it. Ransomware gangs used to have "ethical guidelines" about attacking hospitals. That ended years ago. Now they specifically target healthcare because they know ‣ Hospitals will pay to save lives ‣ Patient data sells for more on the dark web ‣ Downtime isn't an option when people are dying Yesterday in Belgium, someone's surgery got cancelled because of a ransomware gang. Today, it's happening somewhere else. Every hospital in the world is a target. And most of them are running systems older than the doctors treating patients. The next breach won't just cost money. It might cost lives.

42

9 Comments

In France, I saw a clear shift with NIS2. Companies realized : “It’s not just about technical teams… who’s going to own governance? Who will turn regulation into action?” That’s when hybrid profiles (able to talk to the board, lead audits, and still understand technical challenges) became the most sought-after. Now that I’m focusing on the Belgian market, I can feel the same momentum building. NIS2 registrations are booming, companies are mobilizing… and the shortage of cybersecurity talent is already visible. To those of you active in Belgium : Do you also see this shift, from pure technical skills to governance and leadership?

10

💭 ARM, Bicep, or Terraform — What’s the right choice for Azure IaC in 2025? ARM templates? Outdated and clunky. Bicep came in as Microsoft’s cleaner, more manageable solution — essentially a native alternative to Terraform. But in practice? 🔹 Azure-native teams often go with Bicep 🔹 Multi-cloud & enterprise environments still prefer Terraform for its flexibility and maturity Both have their place, it depends on your setup and priorities. What’s your team using and why? #Azure #Bicep #Terraform #DevOps #CloudArchitecture #IaC

4

🔐 Belgium Blue Team Cyber Security Market Update – 2026 Having recently started focusing on the Blue Team Cyber Security market in Belgium, a few trends are standing out already: 1️⃣Cloud security skills are becoming a baseline expectation Many Security Operations roles now require experience with AWS/Azure environments alongside traditional SIEM monitoring. 2️⃣Threat Detection Engineering is growing fast Companies are increasingly building out teams focused on detection logic, threat hunting, and SIEM optimisation rather than purely reactive SOC work. 3️⃣Language requirements still shape hiring In Belgium, the need for Dutch or French alongside English can significantly reduce the available talent pool. 📈 What this means: • Organisations need to move faster in hiring processes • Strong candidates often receive multiple offers • Internal progression is becoming critical to retain SOC talent I’m currently speaking with SOC Analysts, Incident Responders, Threat Hunters, and Detection Engineers across Belgium. 💬 If you’re hiring in this space or open to discussing the market, feel free to reach out! #cybersecurity #blueteam #contract #freelance #Belgium

32

Over the years, both personally and now together with my colleagues at ITcare, I have worked daily with many infrastructures. Every network has partners, peers, upstreams and vendors. Every time you open a ticket with a partner NOC or vendor support, the experience is different. Sometimes it is great. Sometimes it is surprising. And very often lately, it is simply sad. Today’s post is about those sad experiences. Recently, during a project, our client was using a Tier 1 ISP to transport traffic via EVPL toward an NNI. On paper this should be very simple. Transparent transport from A to B. It took almost four weeks! We were not receiving any traffic, while they kept saying they were sending it. After weeks of emails, we jumped on a call and shared packet captures and real evidence. The responses were always the same. “On our side it is transparent, we were using the same config for other customers and it was working” and “the problem is on your side, please involve JTAC to support you with config.” Really? 🤦‍♂️ We were showing that we were simply trunking the same VLANs on both sides and sharing pcaps that clearly showed they were sending traffic with an additional outer VLAN tag on top of what we were sending. In the end, after escalation, a more experienced engineer fixed it in a couple of hours. The real problem? Everything was done by templates, without thinking or validation. Another case was an EVPL from a different Tier 1 ISP. Nobody could tell us if the service was tagged or untagged, point to point or multipoint. The NOC pointed us to a portal with almost no technical details. It was simply useless... We ended up hunting for the person who originally provisioned the service. And then there was a DIA circuit that could only be activated by calling a phone number, navigating an IVR and typing the circuit ID. No emails. No scheduled turn up. Just a phone call. 🙃 I am curious, what other “funny” stories did you have when working with big providers? Why do you think this happens? In my next post, I will share the opposite side. The positive experiences where support teams were truly impressive and did a great job.

32

2 Comments

🔐 Physical security and IT infrastructure go hand in hand. Just signed a new contract with Telenet to manage access control for the building. One more step in the full structuring of our IT infrastructure — covering everything from: the network, to workstations, Microsoft 365, and now physical access security. Because PME don’t need a technician. They need a trusted IT partner who understands the real business challenges. Lots of work ahead — but each brick matters. #IT #Security #AccessControl #Telenet #ITPartner #Microsoft365 #Infrastructure #SMB #PME

6

Something I have been working on for a while goes live on Monday. If you are an IT Director or Head of IT at a professional services firm — and Microsoft 365 Copilot is either already deployed or being discussed — it is probably worth a look. More on Monday.

5

And another one. An exposed Elasticsearch server open for anyone to see, accessible for three months. Openprovider, a Netherlands-based domain registrar, has leaked over 100GB of sensitive customer and internal data, including domain transfer authentication codes, risking domain hijacking. “The exposed server contained logs detailing domain registration data, customer actions, internal response payloads, and even domain transfer authentication codes (authCode). These codes act like passwords for transferring domains, and attackers could abuse them to hijack website addresses. Notably, the exposure included sensitive metadata such as usernames, reseller IDs, WHOIS privacy status, and raw domain provisioning records,” Diachenko said. The records include registrants’ names, addresses, phone numbers, and email addresses, even if they paid for domain privacy services, which redact this info from public WHOIS records.” https://lnkd.in/ec3CjMnY

7

1 Comment