Environment Configuration
Copy the example env file and edit it with your connection details:
Environment Variable Reference
Connection
| Variable |
Type |
Default |
Description |
OPENSEARCH_HOST |
string |
localhost |
OpenSearch server hostname or IP |
OPENSEARCH_PORT |
integer |
9200 |
OpenSearch server port |
OPENSEARCH_USE_SSL |
boolean |
false |
Enable SSL/TLS connection |
OPENSEARCH_VERIFY_CERTS |
boolean |
false |
Verify SSL certificates |
Authentication
| Variable |
Type |
Default |
Description |
OPENSEARCH_USERNAME |
string |
— |
Username for authentication |
OPENSEARCH_PASSWORD |
string |
— |
Password for authentication |
OPENSEARCH_AWS_AUTH |
boolean |
false |
Use AWS IAM authentication |
OPENSEARCH_AWS_REGION |
string |
us-east-1 |
AWS region (when using AWS auth) |
Engine
| Variable |
Type |
Default |
Description |
OPENSEARCH_ENGINE |
string |
faiss |
KNN engine (faiss, lucene, nmslib) |
OPENSEARCH_ALGORITHM |
string |
hnsw |
Search algorithm (hnsw, ivf) |
OPENSEARCH_SPACE_TYPE |
string |
l2 |
Distance metric (l2, cosine, inner_product) |
OPENSEARCH_VECTOR_DIMENSION |
integer |
384 |
Vector embedding dimension |
| Variable |
Type |
Default |
Description |
OPENSEARCH_INDEX_NAME |
string |
docpipe_test |
Name of the OpenSearch index |
OPENSEARCH_DOC_ID_COLUMN |
string |
doc_id_hash |
Column name for document IDs |
OPENSEARCH_EMBEDDINGS_COLUMN |
string |
embeddings |
Column name for vector embeddings |
OPENSEARCH_BATCH_SIZE |
integer |
100 |
Documents per batch |
OPENSEARCH_CREATE_INDEX |
boolean |
true |
Create index if it doesn’t exist |