feat: couchbase integration (#6165)
Co-authored-by: crazywoola <427733928@qq.com> Co-authored-by: Elliot Scribner <elliot.scribner@couchbase.com> Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com> Co-authored-by: Bowen Liang <bowenliang@apache.org>
This commit is contained in:
@@ -375,7 +375,7 @@ SUPABASE_URL=your-server-url
|
||||
# ------------------------------
|
||||
|
||||
# The type of vector store to use.
|
||||
# Supported values are `weaviate`, `qdrant`, `milvus`, `myscale`, `relyt`, `pgvector`, `pgvecto-rs`, `chroma`, `opensearch`, `tidb_vector`, `oracle`, `tencent`, `elasticsearch`, `analyticdb`, `vikingdb`.
|
||||
# Supported values are `weaviate`, `qdrant`, `milvus`, `myscale`, `relyt`, `pgvector`, `pgvecto-rs`, `chroma`, `opensearch`, `tidb_vector`, `oracle`, `tencent`, `elasticsearch`, `analyticdb`, `couchbase`, `vikingdb`.
|
||||
VECTOR_STORE=weaviate
|
||||
|
||||
# The Weaviate endpoint URL. Only available when VECTOR_STORE is `weaviate`.
|
||||
@@ -414,6 +414,14 @@ MYSCALE_PASSWORD=
|
||||
MYSCALE_DATABASE=dify
|
||||
MYSCALE_FTS_PARAMS=
|
||||
|
||||
# Couchbase configurations, only available when VECTOR_STORE is `couchbase`
|
||||
# The connection string must include hostname defined in the docker-compose file (couchbase-server in this case)
|
||||
COUCHBASE_CONNECTION_STRING=couchbase://couchbase-server
|
||||
COUCHBASE_USER=Administrator
|
||||
COUCHBASE_PASSWORD=password
|
||||
COUCHBASE_BUCKET_NAME=Embeddings
|
||||
COUCHBASE_SCOPE_NAME=_default
|
||||
|
||||
# pgvector configurations, only available when VECTOR_STORE is `pgvector`
|
||||
PGVECTOR_HOST=pgvector
|
||||
PGVECTOR_PORT=5432
|
||||
|
||||
4
docker/couchbase-server/Dockerfile
Normal file
4
docker/couchbase-server/Dockerfile
Normal file
@@ -0,0 +1,4 @@
|
||||
FROM couchbase/server:latest AS stage_base
|
||||
# FROM couchbase:latest AS stage_base
|
||||
COPY init-cbserver.sh /opt/couchbase/init/
|
||||
RUN chmod +x /opt/couchbase/init/init-cbserver.sh
|
||||
44
docker/couchbase-server/init-cbserver.sh
Executable file
44
docker/couchbase-server/init-cbserver.sh
Executable file
@@ -0,0 +1,44 @@
|
||||
#!/bin/bash
|
||||
# used to start couchbase server - can't get around this as docker compose only allows you to start one command - so we have to start couchbase like the standard couchbase Dockerfile would
|
||||
# https://github.com/couchbase/docker/blob/master/enterprise/couchbase-server/7.2.0/Dockerfile#L88
|
||||
|
||||
/entrypoint.sh couchbase-server &
|
||||
|
||||
# track if setup is complete so we don't try to setup again
|
||||
FILE=/opt/couchbase/init/setupComplete.txt
|
||||
|
||||
if ! [ -f "$FILE" ]; then
|
||||
# used to automatically create the cluster based on environment variables
|
||||
# https://docs.couchbase.com/server/current/cli/cbcli/couchbase-cli-cluster-init.html
|
||||
|
||||
echo $COUCHBASE_ADMINISTRATOR_USERNAME ":" $COUCHBASE_ADMINISTRATOR_PASSWORD
|
||||
|
||||
sleep 20s
|
||||
/opt/couchbase/bin/couchbase-cli cluster-init -c 127.0.0.1 \
|
||||
--cluster-username $COUCHBASE_ADMINISTRATOR_USERNAME \
|
||||
--cluster-password $COUCHBASE_ADMINISTRATOR_PASSWORD \
|
||||
--services data,index,query,fts \
|
||||
--cluster-ramsize $COUCHBASE_RAM_SIZE \
|
||||
--cluster-index-ramsize $COUCHBASE_INDEX_RAM_SIZE \
|
||||
--cluster-eventing-ramsize $COUCHBASE_EVENTING_RAM_SIZE \
|
||||
--cluster-fts-ramsize $COUCHBASE_FTS_RAM_SIZE \
|
||||
--index-storage-setting default
|
||||
|
||||
sleep 2s
|
||||
|
||||
# used to auto create the bucket based on environment variables
|
||||
# https://docs.couchbase.com/server/current/cli/cbcli/couchbase-cli-bucket-create.html
|
||||
|
||||
/opt/couchbase/bin/couchbase-cli bucket-create -c localhost:8091 \
|
||||
--username $COUCHBASE_ADMINISTRATOR_USERNAME \
|
||||
--password $COUCHBASE_ADMINISTRATOR_PASSWORD \
|
||||
--bucket $COUCHBASE_BUCKET \
|
||||
--bucket-ramsize $COUCHBASE_BUCKET_RAMSIZE \
|
||||
--bucket-type couchbase
|
||||
|
||||
# create file so we know that the cluster is setup and don't run the setup again
|
||||
touch $FILE
|
||||
fi
|
||||
# docker compose will stop the container from running unless we do this
|
||||
# known issue and workaround
|
||||
tail -f /dev/null
|
||||
@@ -110,6 +110,11 @@ x-shared-env: &shared-api-worker-env
|
||||
QDRANT_CLIENT_TIMEOUT: ${QDRANT_CLIENT_TIMEOUT:-20}
|
||||
QDRANT_GRPC_ENABLED: ${QDRANT_GRPC_ENABLED:-false}
|
||||
QDRANT_GRPC_PORT: ${QDRANT_GRPC_PORT:-6334}
|
||||
COUCHBASE_CONNECTION_STRING: ${COUCHBASE_CONNECTION_STRING:-'couchbase-server'}
|
||||
COUCHBASE_USER: ${COUCHBASE_USER:-Administrator}
|
||||
COUCHBASE_PASSWORD: ${COUCHBASE_PASSWORD:-password}
|
||||
COUCHBASE_BUCKET_NAME: ${COUCHBASE_BUCKET_NAME:-Embeddings}
|
||||
COUCHBASE_SCOPE_NAME: ${COUCHBASE_SCOPE_NAME:-_default}
|
||||
MILVUS_URI: ${MILVUS_URI:-http://127.0.0.1:19530}
|
||||
MILVUS_TOKEN: ${MILVUS_TOKEN:-}
|
||||
MILVUS_USER: ${MILVUS_USER:-root}
|
||||
@@ -475,6 +480,39 @@ services:
|
||||
environment:
|
||||
QDRANT_API_KEY: ${QDRANT_API_KEY:-difyai123456}
|
||||
|
||||
# The Couchbase vector store.
|
||||
couchbase-server:
|
||||
build: ./couchbase-server
|
||||
profiles:
|
||||
- couchbase
|
||||
restart: always
|
||||
environment:
|
||||
- CLUSTER_NAME=dify_search
|
||||
- COUCHBASE_ADMINISTRATOR_USERNAME=${COUCHBASE_USER:-Administrator}
|
||||
- COUCHBASE_ADMINISTRATOR_PASSWORD=${COUCHBASE_PASSWORD:-password}
|
||||
- COUCHBASE_BUCKET=${COUCHBASE_BUCKET_NAME:-Embeddings}
|
||||
- COUCHBASE_BUCKET_RAMSIZE=512
|
||||
- COUCHBASE_RAM_SIZE=2048
|
||||
- COUCHBASE_EVENTING_RAM_SIZE=512
|
||||
- COUCHBASE_INDEX_RAM_SIZE=512
|
||||
- COUCHBASE_FTS_RAM_SIZE=1024
|
||||
hostname: couchbase-server
|
||||
container_name: couchbase-server
|
||||
working_dir: /opt/couchbase
|
||||
stdin_open: true
|
||||
tty: true
|
||||
entrypoint: [""]
|
||||
command: sh -c "/opt/couchbase/init/init-cbserver.sh"
|
||||
volumes:
|
||||
- ./volumes/couchbase/data:/opt/couchbase/var/lib/couchbase/data
|
||||
healthcheck:
|
||||
# ensure bucket was created before proceeding
|
||||
test: [ "CMD-SHELL", "curl -s -f -u Administrator:password http://localhost:8091/pools/default/buckets | grep -q '\\[{' || exit 1" ]
|
||||
interval: 10s
|
||||
retries: 10
|
||||
start_period: 30s
|
||||
timeout: 10s
|
||||
|
||||
# The pgvector vector database.
|
||||
pgvector:
|
||||
image: pgvector/pgvector:pg16
|
||||
|
||||
Reference in New Issue
Block a user