Amazon announced the creation of the OpenSearch project, which created a fork of the Elasticsearch search, analysis and data storage platform, as well as the Kibana web interface associated with the platform. The code is distributed under the Apache 2.0 license. In the future, we plan to rename the Amazon Elasticsearch Service to Amazon OpenSearch Service.
OpenSearch is forked from the Elasticsearch 7.10.2 codebase. Officially, work on the fork began on January 21, after which the forked code was cleaned of components not distributed under the Apache 2.0 license, and elements of the Elasticsearch brand were replaced with OpenSearch. In its current form, the code is still at the alpha testing stage, and the first beta release is expected in a few weeks. It is planned to stabilize the code base and make OpenSearch ready for use in production systems by mid-2021.
OpenSearch will be developed as a collaborative project developed with the participation of the community. It is noted that Amazon is currently the curator of the project, but in the future, together with the community, an optimal strategy for managing, making decisions and interacting with the participants involved in the development will be developed.
Companies such as Red Hat, SAP, Capital One and Logz.io have already joined the work on OpenSearch. It is noteworthy that Logz.io previously tried to develop its own fork of Elasticsearch, but joined the work on a common project. Participation in the development of OpenSearch does not require the signing of an agreement on the transfer of property rights (CLA, Contributor License Agreement), and the rules for using the OpenSearch trademark are permissive and allow you to use this name when promoting your products.
The reason for creating the fork was the transfer of the original Elasticsearch project to the non-free SSPL license (Server Side Public License) and the cessation of publishing changes under the old Apache 2.0 license. The SSPL is recognized by the OSI (Open Source Initiative) as non-compliant with the Open Source criteria due to discriminatory requirements. In particular, despite the fact that the SSPL license is based on AGPLv3, the text contains additional requirements for the delivery under the SSPL license not only of the application code itself, but also of the source codes of all components involved in the provision of the cloud service.
The fork is motivated by the intention to keep Elasticsearch and Kibana as open source projects and to provide a complete open source solution developed with the participation of the community. The OpenSearch project will also continue the independent development of the Open Distro for Elasticsearch distribution, which was previously developed jointly with Expedia Group and Netflix as an add-on to Elasticsearch and included additional features that replace the paid components of Elasticsearch, such as machine learning tools, SQL support, notification generation, cluster performance diagnostics mechanisms, authentication via Active Directory, Kerberos, SAML and OpenID, implementation of a single point of entry (SSO), traffic encryption support, a system of differentiation role-based access (RBAC), detailed audit logging.
Source: opennet.ru