DevOps News and Headlines
DevOps News and Headlines
DevOps Pulse
EMAIL DIGESTS
Daily
Every morning
Weekly
Tuesday recap
EMAIL DIGESTS
Daily
Every morning
Weekly
Tuesday recap
Microsoft has released Patch 2 for Azure DevOps Server on March 13 2026, addressing a defect that could deactivate group memberships. The update applies to on‑premises installations that were deployed before the re‑published release and completes remediation for customers who previously ran the mitigation script. New installations or upgrades using the March 13 2026 release are exempt. The patch can be downloaded from the official Azure DevOps Server page and verified with a simple command‑line check.
Grafana Assistant, an AI agent built into Grafana Cloud, now automates cloud cost optimization by translating natural‑language prompts into telemetry queries. It delivers 30‑day waste analyses, actionable recommendations, and transparent data without requiring PromQL expertise. Integrated with Model Context Protocol...
Grafana’s native PagerDuty integration dumps every alert label and annotation into the incident details, creating unreadable payloads. By adding a custom key named "firing" in the contact point’s Details section, users can override the default template and send only essential...
Docker now enables developers to run Claude Code locally, connect it to external tools, and sandbox its actions. Using Docker Model Runner, Claude Code accesses an Anthropic‑compatible API, giving full control over data, infrastructure, and spending. The Docker MCP Toolkit...
Kubernetes 1.36 is slated for release on 22 April 2026, continuing the CNCF’s three‑times‑a‑year cadence. The update emphasizes security, bolstering Linux user namespaces to improve container isolation and refining the WatchCache for faster API queries. It also retires the Ingress‑nginx controller, positioning...
NanoClaw has partnered with Docker to run its open‑source AI agent platform inside Docker Sandboxes, providing enterprise‑grade isolation for autonomous agents. The integration leverages MicroVM‑based sandboxes, allowing agents to install packages, modify files, and access external systems without exposing the...
EXANTE replaced its manual Saturday‑only CRM deployments with a fully automated pipeline that now serves over 30 services across multiple jurisdictions. The new flow triggers on a Git tag, builds images, creates Jira tickets, posts to Slack, and uses Flux...
Buffer discovered seven background jobs running on Amazon SQS for up to five years despite providing no value. A recent repository consolidation allowed engineers to map queues and identify the orphaned workers, leading to their incremental removal. The cleanup eliminated...
Observability Day, a co-located event at KubeCon + CloudNativeCon Europe 2026, brings together CNCF observability project maintainers and practitioners. The program expands beyond traditional monitoring, highlighting AI-driven trace analysis, cost‑efficiency strategies, and large‑scale telemetry engineering. Featuring two parallel tracks, the...
A recent analysis highlights that Kubernetes reliability failures stem from the sheer velocity of machine‑speed control loops rather than tool or skill deficiencies. Deployments, autoscalers, and GitOps reconciliations can trigger cascading alerts that outpace human on‑call response, turning single incidents...
The article demonstrates how to use the sqlpackage command‑line utility to detect schema drift between Azure SQL databases by comparing a DACPAC file against a target database and generating a delta script. It outlines a lightweight, scriptable workflow that avoids...
FluidCloud, a Pleasanton‑based startup, unveiled its Large Infrastructure Model (LIM), an AI engine that generates, translates, and validates Terraform code for multicloud environments. Unlike typical fine‑tuned LLMs, LIM combines a front‑end language parser with custom foundation models trained on synthetic...
Vehicle software updates are shifting from a single release mindset to staged rollouts that serve as safety evidence. Emerging regulations such as UN Regulation 156 and ISO 24089 require a software update management system, and a progressive rollout with measurable health gates...
The article outlines how AI‑driven testing tools are reshaping web application quality assurance in 2026. It highlights core AI techniques—NLP, computer vision, reinforcement learning—that enable self‑healing, semantic element recognition, and visual regression detection. Leading platforms now integrate with CI/CD pipelines,...
PagerDuty announced its Spring 2026 release of the next‑generation Operations Cloud, centering on an autonomous SRE Agent that acts as a virtual responder. The platform now embeds full‑lifecycle incident management directly into Slack and other developer environments via the Model...
Percona Server for MongoDB 8.0.16‑5 uncovered a Docker‑specific failure in the tcmalloc_set_parameters_test, where the allocator reported zero capacity due to misreading the cgroup v2 "max" limit. The root cause was a code change that swapped Mongo’s NumberParser for the standard atoll...
Global App Testing highlights that AI‑driven test generation accelerates coverage but cannot replace human judgment. AI tools can produce large test suites, detect anomalies, and flag surface‑level defects, yet they often miss contextual, regulatory, and edge‑case issues. Integrating human‑in‑the‑loop testing...
Kubernetes operators often face vague symptoms when etcd degrades, leading to slow API responses or total cluster outages. VMware’s open‑source etcd‑diagnosis tool consolidates health, disk I/O, network latency, and resource‑pressure metrics into a single report, turning guesswork into actionable insight....
Richard Yen discovered that AWS RDS Proxy was pinning thousands of client connections after asyncpg temporarily disabled PostgreSQL JIT during type introspection. The JIT toggle altered session state, which RDS Proxy could not track, causing each connection to become permanently...
The Kubernetes community announced the retirement of the Ingress NGINX controller, with best‑effort maintenance ending in March 2026 and no further security updates thereafter. Fastly Security customers must decide between short‑term mitigations—such as a commercial ingress controller or Chainguard’s maintained fork—and...
GitLab’s Container Virtual Registry acts as a pull‑through cache for Docker Hub, Docker Hardened Images (dhi.io), Microsoft Container Registry, Quay and internal registries. It consolidates authentication to GitLab, caches images on the first pull and serves subsequent pulls locally, cutting...
Red Hat introduced a Model Context Protocol (MCP) server for Satellite as a Technology Preview, allowing large language models (LLMs) to query Satellite’s RHEL inventory via REST APIs. The server runs as a container, authenticates with a personal access token, and...
The article contrasts command‑line interfaces (CLIs) and Model Context Protocol (MCP) servers as AI‑native tooling, positioning CLIs for the fast inner development loop and MCPs for the structured outer loop. It highlights the token‑budget penalty of loading full MCP schemas...
Australian Unity has made SonarQube Cloud its enterprise‑wide static application security testing platform, extending it to every codebase and infrastructure‑as‑code project. By enforcing automated quality gates at the earliest stage of development, the firm shifts left on security and quality...
Tetrate has introduced Built on Envoy, a free, open‑source marketplace that bundles ready‑to‑use Envoy extensions. The platform addresses common adoption hurdles such as security integration, authentication, and AI governance by providing pre‑built modules for WAF, OAuth2, SAML, and content‑safety checks....
Manufact, a YC‑backed startup, announced a $6.3 million seed round led by Peak XV to build infrastructure for the Model Context Protocol (MCP), the emerging “USB‑C” standard for AI agents. The company’s open‑source mcp‑use SDK has already logged five million downloads and...
Jenkins’ controller‑agent model hits performance limits as organizations add pipelines, agents, and plugins, often forcing multi‑controller deployments that increase operational overhead. The article contrasts this with TeamCity’s server‑agent architecture, where a single server centralizes configuration while agents scale horizontally. TeamCity...
Microsoft unveiled a new suite of AI‑driven agents that automate end‑to‑end application and infrastructure modernization on Azure. The Azure Copilot migration agent and GitHub Copilot modernization agent entered public preview, enabling continuous discovery, assessment, planning, code transformation, and deployment. By...
Honeycomb announced a suite of AI‑enhanced observability features and the general availability of Honeycomb Metrics. New Agent Skills let Claude Code, Cursor and other AI agents migrate telemetry, create boards and set SLOs autonomously. Automated Investigations, a Slackbot with chain‑of‑thought...
In a recent enterprise round‑table, PostgreSQL experts shifted from feature talk to a trust‑centered dialogue about upgrades, monitoring, and operational control. The discussion highlighted that modern PostgreSQL platforms are judged by how calmly they evolve, not just by raw capabilities....
Developer workflow fragmentation is causing a hidden factory of rework, draining roughly 12 hours per week per engineer and inflating mean time to recovery. The lack of standardized CI/CD and environment provisioning leads to a 30 % capacity loss and up to...
A PostgreSQL production cluster was killed by the OOM killer after a single query consumed 2 TB of RAM, despite work_mem being set to only 2 MB. The investigation revealed that the query’s ExecutorState memory context retained hundreds of thousands of work_mem‑sized...
Pulumi has added a version selector to its Registry, letting users view API documentation for previous major releases of first‑party providers. The dropdown displays the current version plus the latest releases of the two prior major versions, eliminating the need...
Cohesity, together with ServiceNow and Datadog, is launching a recoverability service that can detect and roll back damage caused by agentic AI in enterprise environments. The solution leverages immutable snapshots and API‑driven restorations to return files, databases, vector stores, and...
Developers are using Claude‑powered agents to generate code autonomously, but lack reliable verification. Traditional code reviews are overwhelmed as agents produce dozens of pull requests weekly, prompting a need for automated testing. The author proposes a TDD‑style workflow: write precise...
Tricentis unveiled its Enterprise Agentic Quality Engineering Platform, powered by the new Tricentis AI Workspace, to orchestrate AI agents across testing, automation, performance, and quality intelligence. The platform promises up to 60% regression test automation, 90‑95% faster performance testing, and...
The article walks through deploying a private AI server on Debian or Ubuntu using Ollama and Docker. It starts by adding the user to the sudo and Docker groups, then installs Ollama, pulls the llama3.2 model, and configures it for...
HashiCorp announced that HCP Vault Dedicated is now available in four new cloud regions—AWS Stockholm and Paris, and Azure Australia East and Australia Central. The expansion broadens the service’s global footprint, giving customers the ability to locate Vault clusters closer...
Opsera announced the launch of its AI‑powered AppSec agents, a new suite designed to embed security, compliance, and architectural validation directly into AI‑assisted development workflows. The agents operate as autonomous pre‑commit guards, automatically scanning AI‑generated code, enforcing SOC 2, HIPAA, PCI‑DSS...
Amazon announced that senior engineers must now sign off on any AI‑assisted code changes after a series of high‑impact outages. The incidents, affecting both its retail platform and AWS services, were linked to generative AI tools used without established safeguards....
GitOps and Argo CD provide a “green checkmark” that a cluster matches the Git repo, but that sync alone cannot guarantee the safety of the configuration. Traditional diff and lint tools only catch syntax or schema errors, leaving temporal and dependency...
Enterprises embracing cloud and AI still perform most NetOps tasks manually, creating scalability and error‑prone challenges. Network automation promises to cut human error, improve security, and lower operating costs, serving as the foundation for the emerging agentic NetOps model. By...
Kubernetes excels at infrastructure orchestration but was never meant to be a developer’s primary interface, leading to growing operational friction as organizations scale. Internal developer platforms (IDPs) introduce abstraction layers—golden paths, service catalogs, and self‑service APIs—that shield developers from cluster‑level...
.png)
The Model Context Protocol (MCP) lets AI assistants run Kubernetes commands through a local server while Rafay’s Zero Trust Kubectl Access (ZTKA) supplies a secure, token‑less kubeconfig. This architecture places the MCP server on the admin workstation, routes traffic via...
Rafay’s platform lets GPU cloud operators provision and manage thousands of GPU‑backed Jupyter notebooks for hackathons through a declarative API and templated SKUs. By batching parallel API calls and using an inventory‑aware scheduler, operators can spin up 1,000 environments in...
.png)
Rafay’s zero‑trust kubectl lets operators run commands inside pods on remote GPU‑enabled Kubernetes clusters without exposing the API or using bastion hosts. Using this workflow, they open an exec session to the nvidia‑dcgm‑exporter pod and execute nvidia‑smi to verify driver,...
F5 Distributed Cloud and Red Hat AI have released a joint AI quickstart that secures LLM inference endpoints. The modular blueprint integrates F5’s API security services with Red Hat’s AI platform and can be deployed in under 90 minutes. It adds schema...
AI agents often fail in production because they lack environmental parity, a mismatch known as the data context gap. Providing agents with a production‑identical sandbox—including live schema, services, and data—eliminates this blind spot. Modern platforms achieve this through metadata‑level cloning...
Developers are moving LLM routing logic to the edge to avoid downtime, latency spikes, and scattered code. A proof‑of‑concept built on Fastly Compute acts as an AI gateway that classifies each request with a lightweight model and forwards it to...
Anthropic has launched Claude Code Review, a beta feature that adds AI‑driven agents to automatically analyze pull requests for bugs and security issues. Internal testing shows substantive review comments rose from 16% to 54%, effectively tripling the amount of useful...
