IT System Monitoring Tools

Over the last year, I’ve seen many people fall into the same trap: They launch an AI-powered agent (chatbot, assistant, support tool, etc.)… But only track surface-level KPIs — like response time or number of users. That’s not enough. To create AI systems that actually deliver value, we need 𝗵𝗼𝗹𝗶𝘀𝘁𝗶𝗰, 𝗵𝘂𝗺𝗮𝗻-𝗰𝗲𝗻𝘁𝗿𝗶𝗰 𝗺𝗲𝘁𝗿𝗶𝗰𝘀 that reflect: • User trust • Task success • Business impact • Experience quality    This infographic highlights 15 𝘦𝘴𝘴𝘦𝘯𝘵𝘪𝘢𝘭 dimensions to consider: ↳ 𝗥𝗲𝘀𝗽𝗼𝗻𝘀𝗲 𝗔𝗰𝗰𝘂𝗿𝗮𝗰𝘆 — Are your AI answers actually useful and correct? ↳ 𝗧𝗮𝘀𝗸 𝗖𝗼𝗺𝗽𝗹𝗲𝘁𝗶𝗼𝗻 𝗥𝗮𝘁𝗲 — Can the agent complete full workflows, not just answer trivia? ↳ 𝗟𝗮𝘁𝗲𝗻𝗰𝘆 — Response speed still matters, especially in production. ↳ 𝗨𝘀𝗲𝗿 𝗘𝗻𝗴𝗮𝗴𝗲𝗺𝗲𝗻𝘁 — How often are users returning or interacting meaningfully? ↳ 𝗦𝘂𝗰𝗰𝗲𝘀𝘀 𝗥𝗮𝘁𝗲 — Did the user achieve their goal? This is your north star. ↳ 𝗘𝗿𝗿𝗼𝗿 𝗥𝗮𝘁𝗲 — Irrelevant or wrong responses? That’s friction. ↳ 𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗗𝘂𝗿𝗮𝘁𝗶𝗼𝗻 — Longer isn’t always better — it depends on the goal. ↳ 𝗨𝘀𝗲𝗿 𝗥𝗲𝘁𝗲𝗻𝘁𝗶𝗼𝗻 — Are users coming back 𝘢𝘧𝘵𝘦𝘳 the first experience? ↳ 𝗖𝗼𝘀𝘁 𝗽𝗲𝗿 𝗜𝗻𝘁𝗲𝗿𝗮𝗰𝘁𝗶𝗼𝗻 — Especially critical at scale. Budget-wise agents win. ↳ 𝗖𝗼𝗻𝘃𝗲𝗿𝘀𝗮𝘁𝗶𝗼𝗻 𝗗𝗲𝗽𝘁𝗵 — Can the agent handle follow-ups and multi-turn dialogue? ↳ 𝗨𝘀𝗲𝗿 𝗦𝗮𝘁𝗶𝘀𝗳𝗮𝗰𝘁𝗶𝗼𝗻 𝗦𝗰𝗼𝗿𝗲 — Feedback from actual users is gold. ↳ 𝗖𝗼𝗻𝘁𝗲𝘅𝘁𝘂𝗮𝗹 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 — Can your AI 𝘳𝘦𝘮𝘦𝘮𝘣𝘦𝘳 𝘢𝘯𝘥 𝘳𝘦𝘧𝘦𝘳 to earlier inputs? ↳ 𝗦𝗰𝗮𝗹𝗮𝗯𝗶𝗹𝗶𝘁𝘆 — Can it handle volume 𝘸𝘪𝘵𝘩𝘰𝘶𝘵 degrading performance? ↳ 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆 — This is key for RAG-based agents. ↳ 𝗔𝗱𝗮𝗽𝘁𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗦𝗰𝗼𝗿𝗲 — Is your AI learning and improving over time? If you're building or managing AI agents — bookmark this. Whether it's a support bot, GenAI assistant, or a multi-agent system — these are the metrics that will shape real-world success. 𝗗𝗶𝗱 𝗜 𝗺𝗶𝘀𝘀 𝗮𝗻𝘆 𝗰𝗿𝗶𝘁𝗶𝗰𝗮𝗹 𝗼𝗻𝗲𝘀 𝘆𝗼𝘂 𝘂𝘀𝗲 𝗶𝗻 𝘆𝗼𝘂𝗿 𝗽𝗿𝗼𝗷𝗲𝗰𝘁𝘀? Let’s make this list even stronger — drop your thoughts 👇

💥 This Linux Automation script will save you hours. Ever SSH into a server and think: “What is even going on here?” That’s why I made this: ✅ 1 Linux script to rule them all – Checks CPU, RAM, Disk, Network, Services — everything. Just copy → paste → run. Instant system health check. No more digging through 10 commands. ✅ Here’s what it does: • Shows top processes eating your RAM • Lists open ports & network interfaces • Highlights disk space issues • Prints running services • And more… #!/bin/bash echo "===== SYSTEM HEALTH CHECK =====" echo "Hostname: $(hostname)" echo "Uptime: $(uptime -p)" echo echo "== CPU Load ==" uptime echo echo "== Memory Usage ==" free -h echo echo "== Disk Usage ==" df -h --total echo echo "== Top Processes ==" ps -eo pid,ppid,cmd,%mem,%cpu --sort=-%mem | head -10 echo echo "== Network Interfaces ==" ip a echo echo "== Listening Ports ==" ss -tuln echo echo "== Running Services (systemd) ==" systemctl list-units --type=service --state=running | head -20 echo "===== END =====" 💬 What’s your go-to script? Drop it below! #Linux #ShellTips #DevTools #Productivity #alias

𝗔𝗻𝗸𝗶𝘁𝗮: You know 𝗣𝗼𝗼𝗷𝗮, last Monday our new data pipeline was live in cloud and it failed terribly. Literally had an exhaustive week fixing the critical issues. 𝗣𝗼𝗼𝗷𝗮: Ohh, so don’t you use Cloud monitoring for data pipelines? From my experience always start by tracking these four key metrics: latency, traffic, errors, and saturation. It helps you to check your pipeline health, if it's running smoothly or if there’s a bottleneck somewhere.. 𝗔𝗻𝗸𝗶𝘁𝗮: Makes sense. What tools do you use for this? 𝗣𝗼𝗼𝗷𝗮: Depends on the cloud platform. For AWS, I use CloudWatch—it lets you set up dashboards, track metrics, and create alarms for failures or slowdowns. On Google Cloud, Cloud Monitoring (formerly Stackdriver) is awesome for custom dashboards and log-based metrics. For more advanced needs, tools like Datadog and Splunk offer real-time analytics, anomaly detection, and distributed tracing across service. 𝗔𝗻𝗸𝗶𝘁𝗮: And what about data lineage tracking? How do you track when something goes wrong, it's always a nightmare trying to figure out which downstream systems are affected. 𝗣𝗼𝗼𝗷𝗮: That's where things get interesting. You could simply implement custom logging to track data lineage and create dependency maps. If the customer data pipeline fails, you’ll immediately know that the segmentation, recommendation, and reporting pipelines might be affected. 𝗔𝗻𝗸𝗶𝘁𝗮: And what about logging and troubleshooting? 𝗣𝗼𝗼𝗷𝗮: Comprehensive logging is key. I make sure every step in the pipeline logs events with timestamps and error details. Centralized logging tools like ELK stack or cloud-native solutions help with quick debugging. Plus, maintaining data lineage helps trace issues back to their source. 𝗔𝗻𝗸𝗶𝘁𝗮: Any best practices you swear by? 𝗣𝗼𝗼𝗷𝗮: Yes, here’s what’s my mantra to ensure my weekends are free from pipeline struggles - Set clear monitoring objectives—know what you want to track. Use real-time alerts for critical failures. Regularly review and update your monitoring setup as the pipeline evolves. Automate as much as possible to catch issues early. 𝗔𝗻𝗸𝗶𝘁𝗮: Thanks, 𝗣𝗼𝗼𝗷𝗮! I’ll set up dashboards and alerts right away. Finally, we'll be proactive instead of reactive when it comes to pipeline issues! 𝗣𝗼𝗼𝗷𝗮: Exactly. No more finding out about problems from angry business users. Monitoring will catch issues before they impact anyone downstream. In data engineering, a well-monitored pipeline isn’t just about catching errors—it’s about building trust in every insight you deliver. #data #engineering #reeltorealdata #cloud #bigdata

🚨 New Episode of "Observe & Resolve" is Live! 🚨 🎙️ eBPF: Powerful, but Handle with Care in Kubernetes Hey folks! 👋 i just released a new episode of  "Observe & Resolve," I’m diving into something I absolutely love—#eBPF. It’s one of the most powerful technologies we have today for observability and security in cloud-native environments. But here’s the catch: 🔍 If we’re not careful, eBPF can silently overload our Kubernetes clusters. Running too many or poorly optimized probes can stress the kernel, while kubelet remains blissfully unaware. The result? Unresponsive nodes and a lot of head-scratching. That’s why in this episode, I’ll show you how to monitor and report the resource usage of your eBPF probes—before they become a problem. 🛠️ What you’ll learn: - How to use #InspektorGadget to manage eBPF programs - How to collect metrics and logs with the OpenTelemetry Collector - How Dynatrace helps visualize and alert on eBPF resource usage I’ll walk you through two practical solutions—one using #bpfstats and another using #topebpf— so you can choose what fits your setup best. 📊 By the end, you’ll be able to: - Track CPU and memory usage of your eBPF programs - Build dashboards to spot high consumers - Adjust your security policies based on real usage data 💬 This one’s for anyone who loves eBPF but wants to use it responsibly. Let’s make our clusters smarter, not slower. 👉 https://lnkd.in/dZSSg436 #Kubernetes #eBPF #Observability #CloudNative #OpenTelemetry #Dynatrace #InspektorGadget #DevOps #SRE #K8sMonitoring #ObserveAndResolve

ACSC, in cooperation with international partners, has released Best Practices for Event Logging and Threat Detection, and it contains a lot of useful advice, with a particular focus on detecting Living Off The Land (#LOTL) techniques, including a case study on PRC-attributed threat group Volt Typhoon. At a high level, the advice is to: 1. Have an enterprise-approved event logging policy 2. Store logs centrally to enable correlation 3. Ensure logs are stored securely and that log integrity is assured 4. Have a detection strategy This document has some great checklists of sources of logs and fields to collect, and advice for consistent timestamps (ISO 8601, but the RFC 3339 compatible subset, down to millisecond precision). Also included is a great set of references - well worth a read!

💭 Ever faced the challenge of keeping your data consistent across regions, clouds, and systems — in real time? A few years ago, I worked on a global rollout where CRM operations spanned three continents, each with its own latency, compliance, and data residency needs. The biggest question: 👉 How do we keep Dataverse and Azure SQL perfectly in sync, without breaking scalability or data integrity? That challenge led us to design a real-time bi-directional synchronization framework between Microsoft Dataverse and Azure SQL — powered by Azure’s event-driven backbone. 🔹 Key ideas that made it work: Event-driven architecture using Event Grid + Service Bus for reliable data delivery. Azure Functions for lightweight transformation and conflict handling. Dataverse Change Tracking to detect incremental updates. Geo-replication in Azure SQL to ensure low latency and disaster recovery. What made this special wasn’t just the technology — it was the mindset: ✨ Think globally, sync intelligently, and architect for resilience, not just performance. This pattern now helps enterprises achieve near real-time visibility across regions — no more stale data, no more integration chaos. 🔧 If you’re designing large-scale systems on the Power Platform + Azure, remember: Integration is not about moving data. It’s about orchestrating trust between systems. #MicrosoftDynamics365 #Dataverse #AzureIntegration #CloudArchitecture #PowerPlatform #AzureSQL #EventDrivenArchitecture #DigitalTransformation #CommonManTips

In today’s always-on world, downtime isn’t just an inconvenience — it’s a liability. One missed alert, one overlooked spike, and suddenly your users are staring at error pages and your credibility is on the line. System reliability is the foundation of trust and business continuity and it starts with proactive monitoring and smart alerting. 📊 𝐊𝐞𝐲 𝐌𝐨𝐧𝐢𝐭𝐨𝐫𝐢𝐧𝐠 𝐌𝐞𝐭𝐫𝐢𝐜𝐬: 💻 𝐈𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞: 📌CPU, memory, disk usage: Think of these as your system’s vital signs. If they’re maxing out, trouble is likely around the corner. 📌Network traffic and errors: Sudden spikes or drops could mean a misbehaving service or something more malicious. 🌐 𝐀𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧: 📌Request/response counts: Gauge system load and user engagement. 📌Latency (P50, P95, P99):  These help you understand not just the average experience, but the worst ones too. 📌Error rates: Your first hint that something in the code, config, or connection just broke. 📌Queue length and lag: Delayed processing? Might be a jam in the pipeline. 📦 𝐒𝐞𝐫𝐯𝐢𝐜𝐞 (𝐌𝐢𝐜𝐫𝐨𝐬𝐞𝐫𝐯𝐢𝐜𝐞𝐬 𝐨𝐫 𝐀𝐏𝐈𝐬): 📌Inter-service call latency: Detect bottlenecks between services. 📌Retry/failure counts: Spot instability in downstream service interactions. 📌Circuit breaker state: Watch for degraded service states due to repeated failures. 📂 𝐃𝐚𝐭𝐚𝐛𝐚𝐬𝐞: 📌Query latency: Identify slow queries that impact performance. 📌Connection pool usage: Monitor database connection limits and contention. 📌Cache hit/miss ratio: Ensure caching is reducing DB load effectively. 📌Slow queries: Flag expensive operations for optimization. 🔄 𝐁𝐚𝐜𝐤𝐠𝐫𝐨𝐮𝐧𝐝 𝐉𝐨𝐛/𝐐𝐮𝐞𝐮𝐞: 📌Job success/failure rates: Failed jobs are often silent killers of user experience. 📌Processing latency: Measure how long jobs take to complete. 📌Queue length: Watch for backlogs that could impact system performance. 🔒 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲: 📌Unauthorized access attempts: Don’t wait until a breach to care about this. 📌Unusual login activity: Catch compromised credentials early. 📌TLS cert expiry: Avoid outages and insecure connections due to expired certificates. ✅𝐁𝐞𝐬𝐭 𝐏𝐫𝐚𝐜𝐭𝐢𝐜𝐞𝐬 𝐟𝐨𝐫 𝐀𝐥𝐞𝐫𝐭𝐬: 📌Alert on symptoms, not causes. 📌Trigger alerts on significant deviations or trends, not only fixed metric limits. 📌Avoid alert flapping with buffers and stability checks to reduce noise. 📌Classify alerts by severity levels – Not everything is a page. Reserve those for critical issues. Slack or email can handle the rest. 📌Alerts should tell a story : what’s broken, where, and what to check next. Include links to dashboards, logs, and deploy history. 🛠 𝐓𝐨𝐨𝐥𝐬 𝐔𝐬𝐞𝐝: 📌 Metrics collection: Prometheus, Datadog, CloudWatch etc. 📌Alerting: PagerDuty, Opsgenie etc. 📌Visualization: Grafana, Kibana etc. 📌Log monitoring: Splunk, Loki etc. #tech #blog #devops #observability #monitoring #alerts

Recent academic research has shown that tree-based Machine Learning (ML) models like XGBoost and Random Forest outperform deep learning (DL) methods on supervised learning tasks involving tabular data with heterogeneous features (i.e., categorical and continuous). On homogeneous numerical tabular data, however, DL methods tend to outperform tree-based models. Today, enterprises and organisations across various industries broadly rely on data (in tabular form) to drive business growth and gain a competitive edge. Despite the sophistication of the available analytics tools, the collected data often contains errors. Real-world data frequently has heterogeneous error profiles that may emerge during data collection or transfer. Common data quality problems include missing values, duplicates, numerical outliers, inconsistencies, and violations of business and integrity rules. Minimising errors in the collected data is essential to ensure the accuracy and reliability of data-driven applications. This means that data cleaning plays a vital role in addressing these issues before model training. However, identifying dirty data instances can be a cumbersome and time-consuming process. There are numerous automated error detection tools for tabular data that have been introduced. Nevertheless, these tools suffer from several shortcomings, encompassing the necessity for domain-specific expertise and substantial time requirements. To address these shortcomings, the authors of [1] introduced a novel error detection tool, denoted as SAGED (Software AG Error Detection), which leverage meta-learning principles. SAGED exploits an ensemble of pre-trained models derived from historical datasets to facilitate error detection in new data with limited labelled instances. The SAGED method consists of two key phases:  • Knowledge extraction: The ML models are trained to distinguish erroneous instances within historical datasets, thereby accumulating valuable insights. • Error detection: The pretrained base models, chosen through rigorous matching, generate a comprehensive feature vector based on predictions, facilitating the role of a meta-classifier in pinpointing errors efficiently. They conducted experiments to compare SAGED against 10 state-of-the-art (SOTA) error detection tools, employing a set of 14 real-world datasets. The findings revealed the superior performance of SAGED in error detection tasks, with limited user intervention, compared to the baselines. The links to their paper [1] and #Python code [2] are shared in the comments.