| mapped_pages | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| applies_to |
|
||||||||||||
| products |
|
AutoOps diagnoses issues in {{es}} by analyzing hundreds of metrics, providing root-cause analysis and accurate resolution paths. With AutoOps, customers can prevent and resolve issues, cut down administration time, and optimize resource utilization.
:::{image} /deploy-manage/images/cloud-autoops-overview-page.png :screenshot: :alt: Screenshot showing the Overview page in AutoOps :::
-
Real-time root-cause analysis for hundreds of issues.
-
Accurate resolution paths and customized recommendations.
-
Insight into what occurred and detailed views into nodes, index, shards, and templates.
-
Wide range of insights, including:
- Cluster status, node failures, and shard sizes.
- High CPU, memory, disk usage, and other resource-related metrics.
- Misconfigurations and possible optimizations.
- Insights on data structure, shards, templates, and mapping optimizations.
- Unbalanced loads between nodes.
- Indexing latency, rejections, search latency, high index/search queues, and slow queries.
- Resource utilization.
-
Multi-deployment dashboard to quickly spot issues across all clusters.
-
Possibility to customize event triggers and connect to different notification services such as PagerDuty, Slack, MS Teams, and webhooks.
-
Coming soon: Long-term reports for sustained evaluation.
Depending on your deployment type, AutoOps has been rolled out across various regions. More regions and CSPs are coming soon.
:::{include} /deploy-manage/monitor/_snippets/autoops-availability.md :::
Currently, you can make specific configuration choices to securely use AutoOps in air-gapped environments, but a locally deployable version is planned for a future release.
AutoOps is available for free across all subscription levels in {{ech}} deployments and {{serverless-short}} projects.
New: AutoOps is now also available for free across all self-managed license types in ECE, ECK, and self-managed clusters.
Using AutoOps does not consume ECUs.
AutoOps currently has a 10 day retention period.
AutoOps currently only monitors {{es}}, not the entire {{stack}}. AutoOps is compatible with supported {{es}} versions(7.17.x and above). We plan to expand AutoOps monitoring to the entire stack in the future.
In this section, you'll find the following information:
- How to access AutoOps in your {{ech}} deployments.
- How to use AutoOps in your {{serverless-full}} projects.
- How to connect your ECE, ECK, or self-managed clusters to AutoOps.
- Regions where AutoOps is available.
- Which views AutoOps offers to gain insight into your deployment.
- What AutoOps events are and how you can configure event settings and notifications.
- Frequently asked questions about AutoOps.