Blogsby GlyphDashboard activity logs: Find out which Kibana dashboards get used
Kibana now logs who viewed, edited or deleted each dashboard, how long it took and what failed, so you can catch a broken dashboard before anyone reports it.
15 of 800 posts
Mon Aug 311 post
Fri Aug 281 post
Migrating 1,100 files to Redux Toolkit v2 without freezing the Kibana monorepo
Kibana gave Redux Toolkit v2 the default package name and pushed v1 onto an explicit alias, which inverts the usual migration order. Webpack externals, yarn resolutions and an ESLint rule keep React Redux v7 and v9 out of each other's way.
Thu Aug 271 post
Know your facts: How Elasticsearch AI Indices let agents skip the reading and keep the answer
A technical walkthrough of precomputing facts into an Elasticsearch AI Index, so agents answer from a single ES|QL query instead of reading whole documents, with fewer tokens and lower latency.
Wed Aug 261 post
Taming PUNKs: How ES|QL queries Elasticsearch fields it was never told about
In Elasticsearch 9.5, ES|QL can query unmapped fields. It reads them from _source or returns nulls, so a query keeps working when a field drops out of the mapping and you avoid a reindex that takes hours.
Tue Aug 251 post
Three SLOs every search team needs: monitoring search latency, availability and quality with OpenTelemetry
Your OpenTelemetry search spans already carry the signals for SLOs, burn rate alerts, anomaly detection and incident response, and this post shows how to build all four in Elastic Observability.
Fri Aug 211 post
Let the big model think, let the small model work: Splitting LLM costs in Elastic Workflows
Build an Elastic workflow that sends a data sample to a large model to propose classification labels. A human signs off, then a smaller model applies them across the full corpus.
Thu Aug 201 post
One setting for production vector search: How vectordb_document mode tunes Elasticsearch automatically
Benchmarks across four datasets show how one index setting applies bfloat16 vector quantization, cache preloading and parallel merges to improve vector search throughput and decrease storage.
Wed Aug 191 post
ES95: Adaptive Compression for Elasticsearch Time-Series Metrics
ES95 is Elasticsearch 9.5's new adaptive time series codec that cuts @timestamp storage by 92% and floating point fields by up to 74%, with zero configuration.
Tue Aug 181 post
Skip the mapping explosion: ES|QL queries schemaless JSON keys without dynamic mapping
Flattened fields turn Elasticsearch into a schema-on-read store where you index schemaless data under one mapping, then use ES|QL's FIELD_EXTRACT to pull out any JSON key you need to filter, group or join on, with predicates pushed into...
Mon Aug 171 post
Ask the source: Scaling code search to a billion lines with Elasticsearch and Elastic Agent Builder
Sourcerer matches Claude Code and Codex on code retrieval quality and searches up to thousands of times faster than grep. Every answer links back to the exact files and lines across repos and versions.
Fri Aug 142 posts
Skip the stateful OTel Collector: Elasticsearch 9.5 natively stores both metric temporalities
Ingest cumulative and delta OpenTelemetry metrics under the same metric name while ES|QL and PromQL queries auto-detect temporality per series, with no new syntax or conversion pipelines required.
Search relevance from click streams: Using Learn To Rank and behavioral signals with OpenTelemetry
Learn how to turn click streams and behavioral signals from OpenTelemetry search analytics into judgment lists, rank features and Learn To Rank models that make search relevance improve over time.
Thu Aug 131 post
Your AI agent doesn't need your API key: OAuth 2.1 for Elasticsearch MCP server authentication
OAuth 2.1 lets you connect AI agents to the Elasticsearch MCP server with a browser sign-in instead of an API key. Your agent gets a short-lived token tied to your permissions that you can revoke any time.
Wed Aug 122 posts
Two lines of JSON to replace your ILM policy: data stream lifecycle adds frozen tier support
In Elasticsearch 9.5, frozen_after in data stream lifecycle moves indices to searchable snapshots on object storage on their own, keeping them queryable alongside downsampling and retention.
0.35% trained, 100% competitive: the frozen-tower architecture behind jina-embeddings-v5-omni
The latest jina embeddings model generates multimodal embeddings for text, images, video and audio, competing with models nearly 6x its size on vector search while training just 0.35% of the weights.