Elasticsearch to OpenSearch migration
Move Elasticsearch to OpenSearch run in our cloud, your cloud account or your data center, using the methods the OpenSearch project documents.
- ✓Operated by AceMQ
- ✓160+ customers
An Elasticsearch to OpenSearch migration has three documented methods: remote reindex, which needs no downtime but is slower; snapshot and restore, which needs downtime or a change-capture tool; and Migration Assistant, which can capture live traffic. The Elasticsearch version that created your indexes decides which methods work.
Which migration method fits
The OpenSearch project lists the trade-offs of each method itself. Remote reindex copes with large version jumps, snapshot and restore handles large datasets, and Migration Assistant automates the most and can run with zero downtime.
| Method | What it does | Trade-off | Source |
|---|---|---|---|
| Remote reindex | Reindexes data from the old cluster into the new OpenSearch cluster with the Reindex API. Remote hosts must be on an allow list in opensearch.yml. | No downtime and supports large version jumps. Slower and more resource intensive, and it can degrade the source cluster. | OpenSearch docs |
| Snapshot and restore | A snapshot of the current cluster is restored into a new OpenSearch cluster. The original cluster is untouched. | Supports large datasets and allows reversion. Needs downtime or a change-capture solution, a new cluster, and sometimes manual reindexing. | OpenSearch docs |
| Migration Assistant | A Kubernetes-based tool for moving data, metadata and live traffic. It offers a backfill-only mode with planned downtime and a zero-downtime mode with live-traffic capture and replay. | The most automated path. It handles multi-version hops. | OpenSearch docs |
Sources: OpenSearch documentation, linked in each row and on the migration and upgrade page; checked 30 Oct 2026.
How the methods were chosen: these are the migration methods the OpenSearch documentation lists for moving from Elasticsearch. The fourth method on that page, rolling upgrade, upgrades a cluster in place and supports only adjacent major versions, so it is not used for a move between products.
The Elasticsearch version decides the method
OpenSearch was forked from Elasticsearch 7.10.2. Indexes created by Elasticsearch 7.11 or later cannot be restored from a snapshot, and need Migration Assistant or a reindex.
| Index created by | What OpenSearch reads | Method |
|---|---|---|
| Elasticsearch 6.0 to 7.10.2 | OpenSearch 1.x reads these. OpenSearch 2.x reads 7.0 to 7.10.2. | Snapshot and restore works within these ranges. |
| Elasticsearch 7.11 and later | OpenSearch cannot restore these indexes, even on the same Lucene major version. | Use Migration Assistant, or remote reindex. |
Source: the OpenSearch index compatibility reference on the migration and upgrade page; checked 30 Oct 2026.
- Tools around the cluster.
- If Logstash, Beats or other services connect to the cluster, the OpenSearch tools compatibility matrices show whether they need to change (OpenSearch docs).
- Plugins.
- Plugin compatibility is reviewed before the move, and configuration files are backed up (OpenSearch docs).
- Before the first copy.
- The OpenSearch guidance is to take a snapshot, back up configuration files and stop nonessential indexing before the move (OpenSearch docs).
Sources: OpenSearch documentation, linked above; checked 30 Oct 2026.
The migration in phases
The OpenSearch cluster is built and loaded while Elasticsearch keeps serving search. Traffic only moves once the data and queries are verified on the new cluster.
Review
Elasticsearch version, index versions, mappings, plugins, index size, write rate and the clients and pipelines that connect.
Build the new cluster
Version, node roles, sizing and plugins come from the design.
Copy the data
Remote reindex, snapshot restore or Migration Assistant backfill loads the indexes, depending on the version and the downtime allowed.
Catch up
Live traffic is captured and replayed with Migration Assistant, or changes since the copy are reindexed in a second pass.
Verify and switch
Queries and result quality are compared, and traffic switches to the target once the new side is verified.
The timeline and runbook for Elasticsearch to OpenSearch migrations are written in the design step, from your data volume, traffic and change windows, and you approve them before the cutover is scheduled.
Cutover and rollback
Migration Assistant has a documented replay and switch workflow for live traffic. With reindex or snapshots, the design decides how the gap since the last copy is closed before traffic moves.
- The backfill is complete and failed documents, if any, are tracked and fixed.
- Captured traffic is replayed until the target is in step, or a second reindex pass closes the gap.
- Query results and relevance are checked against the old cluster.
- Traffic switches to OpenSearch. Elasticsearch stays untouched until the new side has run normally.
Source: the Migration Assistant workflow pages (OpenSearch docs); checked 30 Oct 2026. Remote reindex copies documents as they are when the call runs, so later changes need the second pass.
How OSStend runs the move
We run the migration and own the cutover, with both sides live until you switch. Zero downtime is the plan.
Managed OpenSearch from OSStend is operated by AceMQ engineers in our cloud, your cloud account or your data center. The product differences are covered in OpenSearch vs Elasticsearch, the reasons teams leave a hosted Elastic service are on the Elastic Cloud alternatives page, and the migration hub lists the other moves.
AceMQ names a customer only with that customer's written permission, so most of this record stays anonymous. Ask for a similar workload in the 45-minute review.
When this move fits, and when it does not
The move fits when the licence, cost or hosting place is the reason to change and the queries and mappings carry over. It does not when the search stack depends on Elastic-only features.
Contact us if
- ✓Licence policy, cost or hosting place is the reason to move to OpenSearch.
- ✓The queries, mappings and plugins carry over after a compatibility review.
- ✓Search must stay up during the move.
Not a fit if
- ✕The stack depends on Elastic-only features that OpenSearch does not offer.
- ✕The cluster is small and a snapshot restore in a quiet window is enough.
- ✕You want to run the move yourselves: the OpenSearch methods above work without OSStend.
Questions about Elasticsearch to OpenSearch
Can Elasticsearch be migrated to OpenSearch without downtime?
Yes, with two of the documented methods. Remote reindex needs no downtime, and Migration Assistant has a zero-downtime mode with live-traffic capture and replay. Snapshot and restore needs downtime or a change-capture solution (OpenSearch docs, checked 30 Oct 2026).
Which Elasticsearch versions can OpenSearch restore from a snapshot?
OpenSearch 1.x reads indexes created by Elasticsearch 6.0 to 7.10.2, and OpenSearch 2.x reads 7.0 to 7.10.2. It cannot restore indexes created by Elasticsearch 7.11 or later (OpenSearch docs, checked 30 Oct 2026).
Does remote reindex slow the source cluster?
It can. The OpenSearch documentation lists slower speed, higher resource use and possible degradation of the source cluster as the cost of remote reindex (OpenSearch docs, checked 30 Oct 2026). Throttling is available on the Reindex API.
Related pages
- OpenSearchElasticsearch as a service on managed OpenSearchElasticsearch as a service on OpenSearch: dedicated master and data nodes, hybrid keyword and vector search and snapshots, sized for your index and run 24/7.
- ComparisonElasticsearch vs OpenSearch in 2026: which to runElasticsearch vs OpenSearch in 2026: licence, vector search, migration and when each is the better choice. Sourced and dated, by a team that runs OpenSearch.
- AlternativeElastic Cloud pricing and alternatives in 2026Elastic Cloud pricing, the licence history and the alternative: managed OpenSearch under Apache 2.0, tailored to your workload and run where you choose.
- OverviewZero-downtime migration, owned by our engineersZero-downtime migration guides for Amazon MQ, RDS, Redis, Atlas, Oracle, MySQL, Confluent and Elasticsearch, using each engine's documented method.
Plan your OpenSearch move with an engineer
45 minutes with an engineer: the current setup is reviewed against the documented method and the cutover is sketched. No ticket queue.
- ✓A design sketch for your workload, not a plan menu
- ✓A zero-downtime cutover path for what you run today
- ✓A price for that design, after the review