# Prometheus Published articles for Prometheus. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Building an Internal Developer Platform with Artificial Intelligence DevFeed: [Building an Internal Developer Platform with Artificial Intelligence]() Original publisher: [Read original article]() Author: Ben Linders Published: 2026-09-17T11:11:00Z Content type: news Language: en Sources: [InfoQ]() Topics: [internal developer platform](), [AI Development](), [AI search](), [agent observability](), [OpenTelemetry](), [telemetry]() Tags: [agent](), [agents](), [artificial-intelligence](), [culture-methods](), [developer-experience](), [developer-platform](), [distributed-tracing](), [guardrails](), [internal-developer-platform](), [logging](), [logs](), [metrics](), [news](), [opentelemetry](), [platform-artificial-intelligence](), [platform-engineering](), [prometheus](), [traces]() ### AI overview This news article covers a KubeCon presentation about using AI agents as an internal developer platform. It discusses semantic search across sources such as Git, Slack, Jira, repositories, pull requests, and wiki pages; guardrails for controlling actions; and logs, metrics, and traces for understanding agent behavior. The speakers also describe OpenTelemetry conventions for GenAI and related observability tools. ### Source excerpt Agents are becoming the new developer platform, using semantic search with data from tools like Git, Slack, and Jira for context. Things to consider are setting guardrails to block or allow things, and using logs, metrics, and traces to understand agent behavior. By Ben Linders ## ClickHouse Cloud Announces Private Preview of PromQL Support and Time-Series Table Engine DevFeed: [ClickHouse Cloud Announces Private Preview of PromQL Support and Time-Series Table Engine]() Original publisher: [Read original article]() Author: James Cunningham Published: 2026-09-15T14:00:00Z Content type: release Language: en Sources: [ClickHouse Blog]() Topics: [clickhouse](), [Prometheus](), [Time Series](), [Cloud](), [Grafana](), [telemetry](), [Traces]() Tags: [clickhouse](), [cloud](), [grafana](), [logs](), [metrics](), [prometheus](), [time-series](), [traces]() ### AI overview ClickHouse announces a private preview of PromQL support and a time-series table engine in ClickHouse Cloud, allowing Prometheus metrics to be stored in ClickHouse and queried with existing PromQL. ### Source excerpt ClickHouse PromQL support lets you store Prometheus metrics in ClickHouse Cloud, query them using familiar PromQL, and bring metrics together with your logs and traces without rewriting queries in SQL. ## Agoda Replaces 72-Shard SQL Server Price Cache with DragonflyDB DevFeed: [Agoda Replaces 72-Shard SQL Server Price Cache with DragonflyDB]() Original publisher: [Read original article]() Author: Leela Kumili Published: 2026-09-14T13:48:00Z Content type: news Language: en Sources: [InfoQ]() Topics: [Cache](), [Caching](), [migration](), [sql-server](), [Latency](), [Prometheus]() Tags: [a-b-testing](), [agoda-price-cache-dragonflydb](), [architecture-design](), [cache](), [caching](), [development](), [devops](), [distributed-cache](), [distributed-data](), [distributed-systems](), [latency](), [microservices](), [migration](), [news](), [prometheus](), [redis](), [sql-server]() ### AI overview Agoda migrated its 1.5 TB hotel Price Cache from 72 Microsoft SQL Server shards to DragonflyDB. The staged migration used workload testing, dual reads, parity metrics, and gradual traffic shifting; Agoda reported substantially lower P99 read latency. ### Source excerpt Agoda migrated its 1.5 TB hotel Price Cache from 72 SQL Server shards to DragonflyDB to handle growing read and write volumes. The migration used staged dual reads, parity validation, gradual traffic shifting, and decentralized failover detection. Agoda reports an approximately eightfold reduction in P99 read latency, with two DragonflyDB clusters providing high availability. By Leela Kumili ## How to monitor Cypress tests with Grafana Cloud DevFeed: [How to monitor Cypress tests with Grafana Cloud]() Original publisher: [Read original article]() Author: Rajesh Mahalingaswamy Published: 2026-09-12T11:22:06.456390Z Content type: tutorial Language: en Sources: [Grafana Labs blog on Grafana Labs]() Topics: [Grafana](), [Node.js](), [ci]() Tags: [analytics](), [cypress](), [grafana](), [grafana-alloy](), [grafana-cloud](), [grafana-cloud-metrics](), [how-to](), [logs](), [metrics](), [monitoring](), [node-js](), [observability](), [prometheus](), [traces]() ### AI overview A tutorial for monitoring Cypress test results by converting them to Prometheus metrics, pushing them through a Pushgateway, scraping them with Alloy, and forwarding them to Grafana Cloud Metrics. ### Source excerpt If your Cypress suite has tests that fail more often or run slower, you know it can be hard to figure out the pattern from a single job. It could be one spec that slowed down, or a single test that fails, or maybe the entire suite is trending slower. The root cause could be a bug in the app, or a flaky test, or something else. Your terminal output and CI log will tell you what happened on a single run, but that doesn't help you spot any larger trends--especially since you lose that data as soon as the job finishes. Thankfully, Cypress, a front-end automated test framework built for web applications, already exposes everything you need through its plugin hooks. After each spec finishes, Cypress hands you a results object with pass and fail counts, per-test durations, and states. You just need to turn that into metrics and ship it somewhere durable. In this post, you'll learn how to monitor your Cypress tests by converting those results into Prometheus metrics inside a Cypress hook, pushing them to a Prometheus Pushgateway, and letting Alloy scrape the gateway and forward everything to Grafana Cloud Metrics--using nothing but the free tiers. By the end you'll have a pipeline running with the following architecture: What you'll need This walkthrough runs everything alongside your existing Cypress project. Before you start, make sure you have: A Cypress project (this example uses Cypress 14.x) with a cypress.config.js you can edit A Prometheus Pushgateway. Cypress runs are short-lived batch jobs, so they can't be scraped directly--the Pushgateway holds the metrics between runs so a scraper can pick them up. Set up a Prometheus Pushgateway in your infrastructure Alloy, our open source collector we use to scrape the Pushgateway and remote-write to Grafana Cloud A Grafana Cloud account. The free tier includes Grafana Cloud Metrics and a Prometheus remote-write endpoint. If you don't have an account, you can sign up here Your Grafana Cloud remote-write URL, numeric user ID, an ## How to scale Alloy as a central telemetry gateway: capacity planning, load testing, and production lessons DevFeed: [How to scale Alloy as a central telemetry gateway: capacity planning, load testing, and production lessons]() Original publisher: [Read original article]() Author: Fatjon Nebiu Published: 2026-09-12T11:22:06.456390Z Content type: tutorial Language: en Sources: [Grafana Labs blog on Grafana Labs]() Topics: [Grafana]() Tags: [analytics](), [auth](), [best-practices](), [cloud](), [cost](), [data](), [deployment](), [enterprise](), [grafana](), [grafana-alloy](), [grafana-cloud](), [how-to](), [kubernetes](), [logs](), [metrics](), [monitoring](), [observability](), [opentelemetry](), [platform](), [production](), [prometheus](), [scale](), [telemetry](), [testing](), [traces]() ### AI overview A practical guide to scaling Grafana Alloy as a centralized telemetry gateway. It covers capacity planning, load testing, and production considerations for collecting metrics, logs, and traces and forwarding them to Grafana Cloud. ### Source excerpt Running Alloy as a single-instance sidecar is simple. Running it as a centralized gateway that absorbs the full telemetry stream of an enterprise platform--tens of millions of active series, terabytes of logs per day, and tens of thousands of trace spans per second--is a different challenge altogether. To get it right, you need deliberate capacity planning, honest load testing, and a monitoring setup that doesn't rely on the very thing you're testing. As part of the Professional Services team here at Grafana Labs, we've seen this firsthand working with customers. In this post, we'll walk you through the best practices we follow to help them find success, and we'll do so using real, anonymized data from a recent engagement. We'll cover how we sized and load tested a production Alloy central collector deployment on Kubernetes, what the numbers looked like under real stress, and how the cluster behaves today handling the full production telemetry workload for a large enterprise platform. By the end, you should have a better sense for how you can create your own central gateway for collecting telemetry in Grafana Cloud. Why a central gateway? Before diving into numbers, it's worth explaining the pattern. In a central gateway setup, all telemetry from application teams--metrics, logs, and traces--flows to a shared Alloy fleet via OTLP or native Prometheus/Loki write protocols. Alloy buffers, processes, batches, and forwards everything to Grafana Cloud. This gives you several things that per-team sidecar deployments struggle to provide: A single control plane: Auth, rate limiting, and routing in one place so application teams don't need to manage Grafana Cloud credentials Centralized buffering: Ensure a transient Grafana Cloud slowdown doesn't immediately cause data loss at the source Cost visibility: Configure the gateway to only accept telemetry data containing the label or attribute that is mandatory for cost-attribution Protocol normalization: Send OTLP, Prometheus Remote ## Kubernetes v1.37: Native Histograms Graduates to Beta DevFeed: [Kubernetes v1.37: Native Histograms Graduates to Beta]() Original publisher: [Read original article]() Author: Richa Banker Published: 2026-09-11T18:30:00Z Content type: release Language: en Sources: [Kubernetes Blog]() Topics: [telemetry]() Tags: [kubernetes](), [latency](), [metrics](), [observability](), [prometheus](), [storage](), [telemetry](), [time-series]() ### AI overview Kubernetes v1.37 enables Prometheus native histograms by default at Beta. The article explains how dynamic exponential buckets improve metric resolution and quantile accuracy while reducing time-series, scraping, and storage overhead compared with classic histograms. ### Source excerpt I'm excited to announce that native histogram support for Kubernetes metrics is graduating to Beta and is enabled by default in Kubernetes v1.37! Native histograms (previously introduced as Alpha in Kubernetes v1.36 under KEP-5808) bring high-resolution, low-cardinality observability to Kubernetes metrics. By adopting Prometheus Native Histograms, Kubernetes components now expose latency and duration metrics with far greater accuracy while significantly reducing telemetry storage and scraping overhead. Why move beyond classic histograms? Since the early days of Kubernetes observability, duration and latency metrics (such as API server request latencies or scheduling durations) have relied on classic Prometheus histograms. Classic histograms require metric authors to define a static list of cumulative bucket boundaries (le labels), such as 0.005, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1, 2.5, 5, 10. While familiar, this approach introduces three major challenges: The Bucket Guessing Game: If a workload's latency profile changes, for example, shifting into microsecond ranges or experiencing long-tail tail latencies beyond the highest bucket, the histogram loses visibility. Specifying bucket boundaries upfront requires knowing the distribution before observing it High Cardinality & Storage Cost: With classic histograms, each bucket boundary is exported as a separate time series (_bucket{le="..."}). A histogram with 10 buckets across multiple labels multiplies the number of time series by 10, increasing memory consumption in Prometheus and inflating time series database (TSDB) storage costs Interpolation Error in Quantiles: Calculating percentiles using histogram_quantile() relies on linear interpolation between static bucket boundaries. When bucket spans are coarse, quantile calculations can suffer from significant estimation error What are Prometheus native histograms? Prometheus Native Histograms replace static user-defined buckets with dynamic, exponential buckets. Inst ## Security updates for Thursday DevFeed: [Security updates for Thursday]() Original publisher: [Read original article]() Author: jzb Published: 2026-09-10T13:14:35Z Content type: news Language: en Sources: [LWN.net]() Topics: [Security, Privacy and Abuse Prevention](), [.NET](), [Boot Process, passwd, systemd](), [Grafana]() Tags: [ansible](), [grafana](), [kernel](), [linux](), [net](), [opentelemetry](), [oracle](), [prometheus](), [python](), [redis](), [rust](), [security](), [systemd](), [ubuntu](), [updates]() ### AI overview A roundup of security updates issued by several Linux distributions for a range of packages. ### Source excerpt Security updates have been issued by AlmaLinux (389-ds-base, ansible-core, buildah, expat, glib2, gpsd, gpsd-minimal, gzip, kernel, kernel-rt, opentelemetry-collector, osbuild-composer, perl-DBI, python-lxml, python3.12-lxml, qt5-qtbase, thunderbird, valkey, vim, and xz), Debian (pyasn1), Fedora (darktable, freeipa, freerdp2, gdk-pixbuf2, GitPython, libsoup3, openssl, perl-Net-DNS, rust-ppmd-rust, samba, and valkey), Mageia (ceph, firefox, nss, perl-DBI, thunderbird, and wget), Oracle (389-ds-base, buildah, expat, git-lfs, glib2, glibc, gpsd, gpsd-minimal, grafana-pcp, kernel, libssh, nginx, perl-GD, python3.14-cryptography, redis:7, skopeo, thunderbird, valkey, xmlrpc-c, and xz), Slackware (xz), SUSE (bzip2, cpio, curl, dracut, fuse-overlayfs, golang-github-vpenso-prometheus_slurm_exporter, helm, java-1_8_0-ibm, kbfs, kernel, kernel-devel, libopenslide-devel, libsoup, libssh2_org, libusb-1_0, libvirt, libzypp, zypper, mcphost, multipath-tools, NetworkManager, opensc, openssl-3, perl-Net-DNS, python-aiohttp, python-Authlib, python-pip, python-sqlparse, python313-dnspython, python313-idna, rpcbind, sssd, strongswan, systemd, tomcat11, ucode-intel, and wget), and Ubuntu (dotnet8, dotnet10, ffmpeg, flatpak, netty, and perl). ## Whose GPUs are these, anyway? Secure, self-service metrics for multi-tenant Kubernetes DevFeed: [Whose GPUs are these, anyway? Secure, self-service metrics for multi-tenant Kubernetes]() Original publisher: [Read original article]() Author: Bingi Narasimha Karthik (Golden Kubestronaut, Adobe) and Ramkumar Nagaraj (Golden Kubestronaut, Adobe) Published: 2026-09-09T11:27:00Z Content type: article Language: en Sources: [Cloud Native Computing Foundation]() Topics: [Kubernetes](), [GPU](), [telemetry](), [SIEM, Security, Observability]() Tags: [blog](), [cost](), [gpu](), [kubernetes](), [metrics](), [prometheus](), [security]() ### AI overview The article describes providing Kubernetes tenants with secure, self-service access to their own GPU metrics without exposing a central Prometheus instance or allowing noisy queries to affect the wider fleet. ### Source excerpt The question that stopped the meeting It was a routine cost review. The slide showed the month's GPU spend, the biggest line on the whole infrastructure bill, and someone asked a five-word question: "Are we using... ## Kubernetes v1.37: Scale Workloads to Zero with HorizontalPodAutoscaler DevFeed: [Kubernetes v1.37: Scale Workloads to Zero with HorizontalPodAutoscaler]() Original publisher: [Read original article]() Author: Johannes Würbach Published: 2026-09-02T18:30:00Z Content type: release Language: en Sources: [Kubernetes Blog]() Topics: [Kubernetes](), [autoscaling]() Tags: [autoscaling](), [deployment](), [kubernetes](), [metrics](), [prometheus]() ### AI overview Kubernetes v1.37 adds beta, default-enabled HPA support for scaling eligible workloads to zero and back using object or external metrics. It explains queue-based scaling, cold-start trade-offs, and the metrics adapter setup required for an external Prometheus metric. ### Source excerpt Kubernetes v1.37 includes API support for horizontal autoscaling of workloads down to zero replicas. This feature is now Beta and enabled by default. A HorizontalPodAutoscaler (HPA) that uses a suitable object metric or external metric can now scale a workload to zero replicas, then bring it back when the metric changes. Before v1.37, you needed an add-on or external component, or you had to enable the Alpha feature gate, to scale from zero. It is now part of core Kubernetes. Scaling to zero removes the last idle Pod from workloads such as queue consumers and batch processors. The savings are largest when each Pod reserves expensive resources, including dedicated CPUs or GPUs. The trade-off is cold-start time: the HPA must observe the metric, schedule a Pod, and start the application. This works well when work can wait in a durable queue. Kubernetes Services do not buffer requests while no Pods are ready, so HTTP and other request-driven workloads need a separate buffering layer. Why scaling from zero needs a different metric The HPA commonly scales on CPU or memory usage. Both metrics come from running Pods. Once the replica count reaches zero, there are no Pods left to measure and no signal that can tell the HPA to scale back up. Object and external metrics do not have that limitation. A queue length, for example, exists independently of the workers that consume it. The HPA can continue reading the queue length while no workers are running. The following example scales a queue consumer to and from zero using an external metric. Configure an external metric The following example uses a Prometheus metric named queue_consumer_lag. It assumes that Prometheus already collects a series similar to this one: queue_consumer_lag{namespace="default",name="worker_tasks"} Kubernetes needs a metrics adapter to make that value available through the External Metrics API. One implementation is the Prometheus Adapter, which can expose the series using an externalRules entry: extern ## pgwatch v6: dashboards reimagined, and a reaper that doesn't choke DevFeed: [pgwatch v6: dashboards reimagined, and a reaper that doesn't choke]() Original publisher: [Read original article]() Author: Pavlo Golub Published: 2026-08-28T03:00:58Z Content type: release Language: en Sources: [CYBERTEC PostgreSQL | Services & Support]() Topics: [Grafana](), [PostgreSQL](), [Prometheus](), [incident]() Tags: [blog](), [clusters](), [dashboards](), [features](), [grafana](), [incident](), [linux](), [network](), [news](), [patroni](), [pgwatch](), [postgresql](), [production](), [prometheus](), [release]() ### AI overview This release article describes pgwatch v6.0.0-beta, including a Grafana dashboard overhaul, first-class dashboards for Prometheus sources, Patroni cluster views, four new metrics, and fixes for collection hangs caused by production infrastructure failures. ### Source excerpt This is an extended blog, and talks about the features of pgwatch v6.0.0 beta in more details. Read to know and contribute. The post pgwatch v6: dashboards reimagined, and a reaper that doesn't choke appeared first on CYBERTEC PostgreSQL | Services & Support. ## pgwatch v6: Prometheus becomes a source, not just a sink DevFeed: [pgwatch v6: Prometheus becomes a source, not just a sink]() Original publisher: [Read original article]() Author: Pavlo Golub Published: 2026-08-26T07:43:28Z Content type: release Language: en Sources: [CYBERTEC PostgreSQL | Services & Support]() Topics: [Prometheus](), [PostgreSQL](), [Replication](), [dashboards]() Tags: [metrics](), [news](), [password](), [patroni](), [pgwatch](), [postgresql](), [prometheus](), [release](), [tls]() ### AI overview This blog post introduces pgwatch v6.0.0-beta, whose headline feature allows pgwatch to scrape Prometheus exporters directly as first-class sources alongside PostgreSQL sources. It explains source configuration, metric and label handling, and support for Basic Auth and TLS. ### Source excerpt This blog explains all feature highlights related to PGWatch 6.0.0 beta. You can read to understand and test the features yourself. The post pgwatch v6: Prometheus becomes a source, not just a sink appeared first on CYBERTEC PostgreSQL | Services & Support. ## nixos-telemetry: A NixOS flake for opt-in observability pipelines DevFeed: [nixos-telemetry: A NixOS flake for opt-in observability pipelines]() Original publisher: [Read original article]() Author: palo Published: 2026-08-23T08:27:36Z Content type: article Language: en Sources: [Announcements - NixOS Discourse]() Topics: [telemetry](), [OpenTelemetry](), [Tooling](), [log management](), [Prometheus](), [Grafana](), [Grafana Alloy]() Tags: [announcements](), [collector](), [config](), [grafana](), [grafana-alloy](), [loki](), [opentelemetry](), [prometheus](), [telemetry]() ### AI overview A discussion introduces nixos-telemetry, a NixOS flake that uses an OpenTelemetry collector to connect opt-in scrapers, storage, and visualization components. It supports local stacks, remote forwarding, and fan-out to multiple sinks through a unified option tree. ### Source excerpt Hi everyone, I'd like to share nixos-telemetry, a NixOS flake that makes setting up observability in your infrastructure easy. An OpenTelemetry collector sits at the center of each machine. Scrapers, storage, and visualization are all opt-in. You enable what you need; the collector wires the pipeline together automatically. Why? Wiring up telemetry in NixOS today means gluing together Telegraf, Prometheus, Loki, Grafana, Alloy, each with its own config format, ports, and inter-service dependencies. nixos-telemetry puts all of that behind a single telemetry.* option tree: Turn on the system with telemetry.enable = true. Every app is opt-in. Nothing starts that you didn't ask for. The collector starts automatically once a complete pipeline exists, a matching source and sink for the same signal type. No sink? It waits. Forward to a remote collector, run a full local stack, or both. Fan-out to multiple sinks is supported. What it looks like Full local stack on one machine: { telemetry.enable = true; telemetry.telegraf.enable = true; # host metrics telemetry.alloy.enable = true; # journald logs telemetry.prometheus.enable = true; # metrics storage telemetry.loki.enable = true; # logs storage telemetry.grafana.enable = true; # visualization (datasources auto-provisioned) } Forward to a remote collector: # machine 1 { telemetry.enable = true; telemetry.telegraf.enable = true; telemetry.opentelemetry.exporter.endpoints.remote = "100.64.0.1:4317"; } # machine 2 { telemetry.enable = true; telemetry.opentelemetry.receiver.endpoint = "0.0.0.0:4317"; } Full option reference: OPTIONS.md Thanks! Regarding Discourse LLM Policy : I must disclose that substantial parts of this project are llm generated. And because I coppied parts of the README in this post, substantial parts of this announcmement too. 27 posts - 7 participants Read full topic ## What else runs on your Postgres server, and how do we stop it from taking the database down? DevFeed: [What else runs on your Postgres server, and how do we stop it from taking the database down?]() Original publisher: [Read original article]() Author: Kaushik Iska Published: 2026-08-21T14:11:30Z Content type: article Language: en Sources: [ClickHouse Blog]() Topics: [Database](), [Availability](), [Processes](), [Server](), [Monitoring](), [Go Language](), [systemd]() Tags: [availability](), [backup](), [cpu](), [database](), [go](), [logging](), [memory](), [metrics](), [monitoring](), [postgres](), [process](), [processes](), [prometheus](), [server](), [systemd]() ### AI overview ClickHouse Managed Postgres isolates supporting services with runtime budgets and cgroup v2 limits so leaks, CPU saturation, and memory pressure do not compromise database availability. ### Source excerpt ClickHouse Managed Postgres uses runtime budgets, cgroup limits, and disk-full session exemptions to keep supporting services from compromising database availability. ## Adobe Firefly: Simplified observability with Amazon Managed Prometheus DevFeed: [Adobe Firefly: Simplified observability with Amazon Managed Prometheus]() Original publisher: [Read original article]() Author: Dev Arora Published: 2026-08-13T00:14:19Z Content type: article Language: en Sources: [AWS Architecture Blog]() Topics: [observability](), [GPU](), [Amazon EKS](), [Amazon Web Services](), [Monitoring](), [telemetry](), [migration](), [Training AI Models]() Tags: [adobe](), [amazon](), [amazon-eks](), [amazon-elastic-kubernetes-service](), [amazon-managed-service-for-prometheus](), [amazon-web-services-aws](), [aws](), [compute](), [cpu](), [customer-solutions](), [data](), [deployment](), [efficiency](), [gpu](), [infrastructure](), [jobs](), [kubernetes](), [metrics](), [observability](), [performance](), [prometheus]() ### AI overview Adobe Firefly migrated critical GPU infrastructure metrics from self-managed Prometheus to Amazon Managed Service for Prometheus. The article describes the observability challenges of large-scale model training on Amazon EKS, including high-cardinality GPU, compute, memory, and network telemetry, and reports 28x faster GPU metric queries with improved reliability and operational efficiency. ### Source excerpt Learn how Adobe Firefly achieved 28x faster GPU metric queries by migrating from self-managed Prometheus to Amazon Managed Service for Prometheus, with improvements in query performance, infrastructure reliability, and operational efficiency. ## Building open-source observability together: The OpenSearch Observability TAG turns one DevFeed: [Building open-source observability together: The OpenSearch Observability TAG turns one]() Original publisher: [Read original article]() Author: Dotan Horovits Published: 2026-08-11T17:01:57Z Content type: article Language: en Sources: [OpenSearch]() Topics: [observability](), [Open Source](), [OpenTelemetry](), [Prometheus](), [agent observability]() Tags: [agent-observability](), [blog](), [cncf](), [community](), [metrics](), [observability](), [open-source](), [opensearch](), [opentelemetry](), [prometheus]() ### AI overview The OpenSearch Observability Technical Advisory Group marks its first anniversary after a year of community-driven guidance, RFC reviews, architecture decisions, and collaboration across organizations. The article highlights work involving OpenTelemetry, Prometheus, Perses, KEDA, Kubernetes, and the OpenSearch Observability Stack, and explains how contributors can participate in the group's second year. ### Source excerpt The OpenSearch Observability TAG's first year: 10+ orgs shipped OTLP ingestion, Prometheus metrics, an open-source Observability Stack, and AI agent observability. The post Building open-source observability together: The OpenSearch Observability TAG turns one appeared first on OpenSearch. ## Whats new in ClickStack - June + July DevFeed: [Whats new in ClickStack - June + July]() Original publisher: [Read original article]() Author: The ClickStack Team Published: 2026-08-10T17:53:42Z Content type: release Language: en Sources: [ClickHouse Blog]() Topics: [releases](), [dashboards]() Tags: [clickhouse](), [dashboards](), [metrics](), [observability](), [opentelemetry](), [prometheus](), [releases](), [traces]() ### AI overview ClickStack's June-July update adds trace-navigation improvements, exponential-histogram quantiles, Prometheus-compatible endpoint connections, and dashboard enhancements. It also uses ClickHouse text indexes for faster autocomplete and filter dropdowns where available. ### Source excerpt Explore ClickStack's latest upgrades, from richer trace navigation and Prometheus connectivity to smarter dashboards, quieter alerts, faster filtering and support for exponential histogram metrics. ## Beszel: monitorización ligera para Docker DevFeed: [Beszel: monitorización ligera para Docker]() Original publisher: [Read original article]() Author: Antonio Pérez Published: 2026-08-06T00:00:00Z Content type: tutorial Language: es Sources: [tengoping.com]() Topics: [Docker](), [dashboards](), [Homelab](), [Open Source](), [Grafana](), [Prometheus]() Tags: [dashboards](), [docker](), [grafana](), [homelab](), [open-source](), [prometheus]() ### AI overview This article explains Beszel, a lightweight, self-hosted, open-source server monitoring platform for CPU, memory, disk, network, and Docker container metrics. It describes the hub-and-agent architecture, deployment options, suitable homelab and small VPS use cases, and trade-offs versus Prometheus, Grafana, and Zabbix. ### Source excerpt Instala Beszel, monitorización ligera con arquitectura hub-and-agent para vigilar CPU, RAM, disco y contenedores Docker sin montar un stack pesado. ## Use of VRF in dual WAN setup DevFeed: [Use of VRF in dual WAN setup]() Original publisher: [Read original article]() Published: 2026-08-03T20:58:23Z Content type: tutorial Language: en Sources: [Personal blog of Anurag Bhatia]() Topics: [networking](), [MikroTik](), [Prometheus](), [Containers]() Tags: [containers](), [mikrotik](), [prometheus](), [router](), [routing](), [routing-table](), [vrf]() ### AI overview This article describes migrating a MikroTik home router from multiple routing tables to a VRF-based dual-WAN design. It explains how shared routing domains caused router-bound traffic to return through the wrong ISP during partial outages and made ISP-specific routing for measurement containers difficult. VRFs provide separate layer-3 routing domains for the WAN links, while Prometheus and Semaphore handle packet-loss-based switchover. ### Source excerpt Over the weekend I migrated my home router from multiple routing tables to a VRF-based design, placing each WAN uplink into its own VRF. While multiple routing tables worked for basic policy routing, they have some limitations and that led to several edge cases that became increasingly difficult to work around. Issues with the setup: If the active default route pointed to ISP1, traffic arriving on ISP1 naturally returned via ISP1. However, traffic terminating on the router and arriving via ISP2 would also be replied to through ISP1, since both uplinks still shared the same routing domain. This did not cause issues for traffic on devices below the router but was bad for traffic terminating on the router interface itself. Due to the above reason, I recently lost access to my home router while I was out of the country because ISP1 had a partial outage (their transit went down, peering stayed up) & due to distributed tooling, the auto switch trigger could not happen either to take care of it. Packets from ISP 2 were being returned via the ISP1 route & thus blackholed. I have a special case where I want most of the devices on a redundant setup but some devices (containers) on specific ISP only. These are measurement containers running blackbox exporter behind a specific ISP as well as RIPE Atlas. I don't want these to switch over for accuracy of measurement. Without VRF it was ugly config-wise, as ISP1 failure will lead to ISP2 routing even when the specific routing table did not have that route. Old Setup My old setup was running multiple routing table pairs: ISP1 only & ISP1 as primary (ISP2 as secondary) ISP2 only & ISP2 as primary (ISP1 as secondary) All WAN interfaces, LAN interfaces and routing tables were in the same single "Main" VRF. This by default takes care of full outage events, fibre cuts, etc. For packet loss-based switchover, I have set up Prometheus + Semaphore as covered in this post last year. Understanding VRF VRF is Virtual Routing and Forwarding. It ## AWS Weekly Roundup: Price reduction of GPT models in Bedrock, CloudWatch managed collectors for Prometheus metrics, and more (August 3, 2026) DevFeed: [AWS Weekly Roundup: Price reduction of GPT models in Bedrock, CloudWatch managed collectors for Prometheus metrics, and more (August 3, 2026)]() Original publisher: [Read original article]() Author: Micah Walter Published: 2026-08-03T16:12:30Z Content type: news Language: en Sources: [AWS News Blog]() Topics: [AI, ML & Data Engineering](), [telemetry](), [networking](), [Amazon EC2](), [Amazon Elastic Kubernetes Service](), [Amazon OpenSearch Service]() Tags: [amazon-bedrock](), [amazon-cloudwatch](), [amazon-s3-tables](), [aws](), [aws-iam-identity-center](), [bedrock](), [cloud-networking](), [gpt](), [inference](), [observability](), [openai](), [prometheus](), [week-in-review]() ### AI overview AWS weekly roundup covering lower Amazon Bedrock pricing for OpenAI GPT-5.6 models, managed Prometheus metric collection in CloudWatch, and Oracle Cloud Infrastructure connectivity through AWS Interconnect. ### Source excerpt Last week I had the joy of participating in Amazon's "Bring Your Kids to Work Day" with my 7 year old son. We commuted together into the New York City office, his first real rush hour train ride, and spent the day exploring how Amazon uses AI, machine learning, and robotics to deliver packages to [...] ## How to Self-Host a Validated AI Coding Assistant with NVIDIA NeMo Guardrails DevFeed: [How to Self-Host a Validated AI Coding Assistant with NVIDIA NeMo Guardrails]() Original publisher: [Read original article]() Author: Tanya Lenz Published: 2026-07-29T16:46:45Z Content type: tutorial Language: en Sources: [NVIDIA Developer](), [NVIDIA Technical Blog]() Topics: [AI Chat](), [ai observability](), [Grafana]() Tags: [agentic-ai-generative-ai](), [ai](), [ai-agent](), [ai-coding](), [audit-trail](), [ci](), [developer-tools-techniques](), [docker](), [featured](), [git](), [gpu](), [grafana](), [how-to](), [metrics](), [nemo](), [nim](), [nvidia](), [policy](), [prometheus](), [python](), [supply-chain](), [tutorial]() ### AI overview Tutorial for self-hosting a validated AI coding assistant on NVIDIA infrastructure. It combines a StarCoder2-7B NIM endpoint, NeMo Guardrails, CI dependency checks, commit traceability, and a Prometheus/Grafana metrics loop. ### Source excerpt Deploying an AI coding assistant in a regulated, sovereign, or source-sensitive environment, often comes with challenges. Three common issues are: the source... ## How to Connect Prometheus Alerts to an Event-Driven AI Agent for Initial Investigation DevFeed: [How to Connect Prometheus Alerts to an Event-Driven AI Agent for Initial Investigation]() Original publisher: [Read original article]() Author: Pramodh Kumar M Published: 2026-07-23T15:00:31Z Content type: tutorial Language: en Sources: [Kubernetes - KodeKloud Blog | DevOps, Cloud, Kubernetes, AI Tutorials & More]() Topics: [AI Agent](), [Prometheus](), [event driven](), [Kubernetes](), [Slack](), [Deployment]() Tags: [agents](), [ai](), [ai-agents](), [aiops](), [alert-fatigue](), [alert-manager](), [alert-triage](), [architecture](), [auto-remediation](), [automated-incident-response](), [automation](), [devaiops](), [event-driven](), [event-driven-ai-agents-with-prometheus-alerts](), [guide](), [how-to](), [kubernetes](), [kubernetes-ai-agent](), [prometheus](), [prometheus-alert-rules](), [prometheus-alertmanager-webhook](), [root-cause-analysis](), [slack](), [sre](), [sre-automation]() ### AI overview This guide explains how to connect Prometheus and Alertmanager to an event-driven AI agent that investigates alerts before a human responds. It covers the architecture, read-only investigation tools, alert-rule annotations, safety guardrails, and a progression toward guarded remediation. ### Source excerpt Every page interrupts a human, yet most alerts end in the same ten investigation steps. Here is how event driven AI agents catch Prometheus alerts and do that first pass before you even look at your phone. ## Diagnosing a 28,232-Connection SSE Limit in a Go Leaderboard DevFeed: [Diagnosing a 28,232-Connection SSE Limit in a Go Leaderboard]() Original publisher: [Read original article]() Author: Pranshu Raj Published: 2026-07-23T00:00:00Z Content type: tutorial Language: en Sources: [Pranshu Raj - blog on backend systems, performance and sidequests]() Topics: [Server-sent events (SSE)](), [Go Language](), [Docker Compose](), [Redis](), [Grafana](), [Prometheus]() Tags: [docker-compose](), [go](), [grafana](), [prometheus](), [redis](), [sse]() ### AI overview A tutorial on testing and diagnosing the SSE connection limit of a Go leaderboard service. It describes the service architecture, Docker Compose deployment, and monitoring with Prometheus and Grafana. ### Source excerpt Why my Go leaderboard's capped at exactly 28,232 SSE connections, how Docker networking got me to 150k, and the path toward millions. ## Broadcasting to 28K SSE Clients (and learning about Go channels) DevFeed: [Broadcasting to 28K SSE Clients (and learning about Go channels)]() Original publisher: [Read original article]() Author: Pranshu Raj Published: 2026-07-23T00:00:00Z Content type: tutorial Language: en Sources: [Pranshu Raj - blog on backend systems, performance and sidequests]() Topics: [Go Language](), [Server-sent events (SSE)](), [Concurrency](), [backpressure](), [real-time](), [observability](), [Redis](), [Grafana](), [Prometheus]() Tags: [architecture](), [backpressure](), [concurrency](), [go](), [grafana](), [observability](), [prometheus](), [real-time](), [redis](), [sse]() ### AI overview The article describes building a real-time leaderboard in Go that reached 28,232 concurrent SSE connections before exposing a broken broadcast design. It explains the fix using centralized polling, deduplication, fan-out, backpressure handling, and observability with Prometheus and Grafana. ### Source excerpt A post on progressively getting my server to 28k SSE connections, finding a bug in my implementation of fan out using Go channels and the resulting mental model update. ## Building a Custom Metrics Exporter for Kubernetes DevFeed: [Building a Custom Metrics Exporter for Kubernetes]() Original publisher: [Read original article]() Author: Victor David Effiok Published: 2026-07-14T18:00:00Z Content type: tutorial Language: en Sources: [Kubernetes Blog]() Topics: [Kubernetes](), [autoscaling](), [Grafana Cloud Metrics](), [Time Series](), [Go Language](), [WebSocket]() Tags: [autoscaling](), [building](), [code](), [go](), [http](), [kubernetes](), [metrics](), [prometheus](), [server](), [time-series]() ### AI overview A tutorial on building a custom Kubernetes metrics exporter that exposes application or external state for Prometheus, enabling queries, alerts, and HorizontalPodAutoscaler scaling. It covers exporter architecture, direct instrumentation, metric formatting, and choosing counters, gauges, or histograms. ### Source excerpt Kubernetes ships with built-in awareness of CPU and memory, but most real-world scaling decisions depend on signals that live entirely outside that narrow window: how many messages are waiting in a queue, how long the last batch job took, how many active WebSocket connections a pod is holding. When the built-in metrics are not enough, a metrics exporter bridges that gap. This post walks through writing one from scratch, packaging it as a container, and wiring it into a cluster so that Prometheus -- and ultimately the HorizontalPodAutoscaler -- can consume it. What a metrics exporter actually does An exporter is a small HTTP server with a single responsibility: expose application state as text on a /metrics endpoint. Prometheus scrapes that endpoint on a regular interval, stores the time-series data, and makes it available for queries, alerts, and autoscaling rules. In some cases you can instrument your application directly -- embedding the Prometheus client library and exposing /metrics from within the same process -- rather than running a separate exporter. A standalone exporter makes more sense when the data source is external to your application or when you do not control the application code. The format Prometheus expects is plain text -- one metric per line, with a name, optional labels, and a numeric value. Client libraries handle the serialization for you, so in practice you only need to decide what to measure and call the right function when that value changes. Choosing what to measure Before writing any code, it helps to decide what kind of signal you are dealing with. The Prometheus data model has three main types: Counters only ever increase. They are the right tool for totals: requests served, jobs processed, errors encountered. Never use a counter for a value that can go down. Gauges represent a current snapshot of a value that can rise and fall freely. Queue depth, active connections, and cache size are all gauges. Histograms record the distribution of obse [Next page]()