/
SC-2190 : ES scaling - reindexing Org index

SC-2190 : ES scaling - reindexing Org index

About

The org index mapping has been reviewed and optimised to reduce the index size without limiting any of the search capabilities.

Steps to re-index the data with the new mapping:

  1. First check the number of number_of_shards & number_of_replicas from old org index by running below curl

    curl --location --request GET 'http://localhost:9200/_cat/indices/%2A?v=&s=index:desc'

 

2. Create new org index with name as orgv{x} // x is version number and update the org mapping using ES Mapping jenkins job {OpsAdministration/job/dev/job/Core/job/ESMapping/}

3. After updating org mapping, run the reindex curl

curl --location --request POST 'localhost:9200/_reindex?pretty' \ --header 'Content-Type: application/json' \ --data-raw '{ "source": { "index": "{old_org_index_name}" }, "dest": { "index": "{new_org_index_name}" } }'

4. After updating replica , add alias to the new index with alias name as org_alias , by below curl

curl --location --request PUT 'localhost:9200/{new_index_name}/_alias/org_alias?pretty'

Note : If alias name changed other than org_alias, update the learner env for org_index_alias

To remove alias binding from old index run the below curl

 

5. Deploy the learner and check the disk usage , replica number for the new index by running the below curl

Related content

Elasticsearch mapping update job steps
Elasticsearch mapping update job steps
More like this
Sunbird Elasticsearch migration to multi index
Sunbird Elasticsearch migration to multi index
More like this
Mapping type removal in Elasticsearch
Mapping type removal in Elasticsearch
More like this
Elasticsearch upgrade in Sunbird Platform
Elasticsearch upgrade in Sunbird Platform
More like this
Elasticsearch Type Mapping
Elasticsearch Type Mapping
More like this
Core Elasticsearch upgrade
Core Elasticsearch upgrade
More like this