24AUG
Fix Elasticsearch Unassigned Shards & Cluster Red
A step-by-step guide to diagnosing unassigned shards and a red Elasticsearch cluster with the allocation explain API, plus safe fixes and prevention tips.
TECHNICAL BLOG
OFFICIAL ELASTIC PARTNER
Elasticsearch & OpenSearch Engineering Notes
Practical production lessons on search infrastructure, architecture, performance, migrations, and relevance.
22JUL
How Budget Thuis Tamed Elasticsearch CPU, Extended
When your observability cluster spends its life at 100% CPU, dashboards feel sluggish, ingest backs up, and every outage hunt starts with “maybe Elasticsearch is the problem again.” For Budget
20JUL
DataDome Elasticsearch: Stabilizing After RED & ILM
How DataDome stabilized a petabyte-scale Elasticsearch cluster after RED incidents and ILM storms—with a 24-minute recovery playbook.
10APR
Kibana CSV Reporting for Domainsbot’s Pandalytics Workload
If you run analytics on hundreds of millions of domain datapoints, your Elasticsearch cluster is not “just search.” It is a production data plane for product teams, analysts, and sometimes entire
27MAR
Snap! Raise: 86% OpenSearch Cost Reduction
This case study shows how Snap! Raise achieved an 86% OpenSearch cost reduction while scaling safely through orphan-index cleanup, shard discipline, and proactive alerting.
17DEC
Arcade Player Lifetime with Elasticsearch
LAI Games needed trustworthy player lifetime and active segments from messy arcade events. We locked the rules into ingest pipelines and continuous transforms—no invented cost percentages.
30MAY
ImportYeti: OpenSearch Health Check at 500M Documents
How a slowlog-first OpenSearch health check at ImportYeti exposed GC pressure, bucket limits, and aggregation-heavy queries behind 70M bills of lading.
24JUN
myaihub.ai: Hybrid Search on Elasticsearch with eland
myaihub.ai combined kNN semantic search (multilingual-e5-small via eland) with English-analyzer lexical filtering to fix irrelevant AI-tool results.
12JUN
musth.io: Email Ratio Dashboards on OpenSearch
musth.io needed dynamic email open/sent ratios per campaign offer on managed OpenSearch. TSVB Filter Ratio, a Vega table, and a 19-field mapping cleanup made it work.
07JUN
Elasticsearch for Vehicle Telemetry: Diff, Enrich, ILM
A field guide to vehicle/IoT telemetry on Elasticsearch: calculate diffs between counter events, join lookup data with enrich processors, and manage the time series with rollover + ILM.
14NOV
Keeping Elasticsearch Fresh: Sync, Deletes, and Refresh
An AI startup told me 'refreshing the index is the problem.' Here is how to untangle Elasticsearch refresh from data freshness—deletes, sync, and alias swaps.
15SEP
MetadataWorks: Metadata Quality Scores in Elasticsearch
MetadataWorks needed a per-dataset metadata quality score. We built it inside Elasticsearch: nine chained ingest pipelines, Painless presence checks, a weighted overall score, and Kibana on top.
01SEP
Natrx: Elastic Observability for a 3-Site WASP Printer Fleet
Natrx centralized three WASP printer sites on Elastic—production/staging fields, 25GB/30d ILM, Fleet HEALTHY, and 5-minute no-data alerts for edge silence.
15JUL
Advanced ChemBlocks: Magento Search on Elasticsearch
Chemistry e-commerce search on Magento 2: Elasticsearch dying at boot with a JNA error, version constraints, and a captured query that shows why field-level symptoms live in the index.
09JUN
Index per Tenant in Elasticsearch: 7,000 Shards Later
A fintech platform ran 7,000 indices as 7,000 shards on 400 GB—one index per account. Why that works on day one, where it breaks, and what to monitor.
01MAY
Shipping Laravel Logs to Elasticsearch with Logstash
A field guide to shipping Laravel application logs into the ELK stack: grok patterns for Laravel's log format, multiline stack traces, and a local Vagrant lab to test it all.
08FEB
Elasticsearch Indexing Slow? Fix Write Backpressure
When Elasticsearch segment creation outruns merges you get 429 rejections and throttled writes. Here is how I diagnose and fix indexing backpressure.
13JAN
Elasticsearch post_filter for Faceted Search
Filter in the query and your facets collapse. The post_filter + filtered-aggs pattern from a real health e-commerce build, plus the edge_ngram trap that matched everything.
11NOV
Dais Technology: Postgres JSON to Elasticsearch Search
Insurance sales data in Postgres JSON, search in Elasticsearch on AWS—diagnosing indexing latency from 1-50 ngrams, refresh-per-write, and per-query mapping reads.
17OCT
Autocomplete on Every Field with Elasticsearch edge_ngram
A field-tested pattern: one dynamic template, one edge_ngram analyzer, one multi_match query — and an 8-character UUID fragment finds a deeply nested SOC case.
30AUG
PostgreSQL to Elasticsearch with Logstash JDBC Input
A field-tested walkthrough for syncing PostgreSQL into Elasticsearch with Logstash JDBC input—complete config, incremental sync with sql_last_value, and the honest truth about deletes.
29JUN
ElastAlert2 Setup: Rules, Email Alerts, systemd Service
A field-tested ElastAlert2 guide: a real frequency rule, readable email alert templates, a systemd unit for 24/7 operation, and when to pick it over Kibana alerting.
10MAY
Elasticsearch Nested Fields: Aggregations and inner_hits
A field guide to Elasticsearch nested fields: when nested beats object, how nested aggregations really work, and how inner_hits reveals which nested element matched.