Add changes to run darwin locally

UiPath · Jan 16, 2025 · 2d7d824 · 2d7d824
1 parent eb3cd1a
commit 2d7d824
Show file tree

Hide file tree

Showing 4 changed files with 55 additions and 36 deletions.
diff --git a/backend/danswer/document_index/vespa/app_config/hosts-dev.xml b/backend/danswer/document_index/vespa/app_config/hosts-dev.xml
@@ -1,36 +1,7 @@
 <?xml version="1.0" encoding="utf-8" ?>
+<!-- Copyright Vespa.ai. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root. -->
 <hosts>
-  <host name='vespa-configserver-0.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>config0</alias>
+  <host name="index">
+    <alias>danswer-node</alias>
   </host>
-  <host name='vespa-configserver-1.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>config1</alias>
-  </host>
-  <host name='vespa-configserver-2.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>config2</alias>
-  </host>
-  <host name='vespa-admin-0.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>admin0</alias>
-  </host>
-  <host name='vespa-feed-container-0.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>feed0</alias>
-  </host>
-  <host name='vespa-feed-container-1.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>feed1</alias>
-  </host>
-  <host name='vespa-query-container-0.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>query0</alias>
-  </host>
-  <host name='vespa-query-container-1.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>query1</alias>
-  </host>
-  <host name='vespa-content-0.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>content0</alias>
-  </host>
-  <host name='vespa-content-1.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>content1</alias>
-  </host>
-  <host name='vespa-content-2.vespa-internal.darwin-dev.svc.cluster.local'>
-    <alias>content2</alias>
-  </host>
-</hosts>
+</hosts>
diff --git a/backend/danswer/document_index/vespa/app_config/services-dev.xml b/backend/danswer/document_index/vespa/app_config/services-dev.xml
@@ -0,0 +1,36 @@
+<?xml version="1.0" encoding="utf-8" ?>
+<services version="1.0">
+    <container id="default" version="1.0">
+        <document-api/>
+        <search/>
+        <http>
+            <server id="default" port="8081"/>
+        </http>
+        <nodes>
+            <node hostalias="danswer-node" />
+        </nodes>
+    </container>
+    <content id="danswer_index" version="1.0">
+        <redundancy>1</redundancy>
+        <documents>
+            <!-- <document type="danswer_chunk" mode="index" /> -->
+            DOCUMENT_REPLACEMENT
+        </documents>
+        <nodes>
+            <node hostalias="danswer-node" distribution-key="0" />
+        </nodes>
+        <tuning>
+            <resource-limits>
+                <!-- Default is 75% but this can be increased for Dockerized deployments -->
+                <!-- https://docs.vespa.ai/en/operations/feed-block.html -->
+                <disk>0.75</disk>
+            </resource-limits>
+        </tuning>
+        <config name="vespa.config.search.summary.juniperrc">
+            <max_matches>3</max_matches>
+            <length>750</length>
+            <surround_max>350</surround_max>
+            <min_length>300</min_length>
+        </config>
+    </content>
+</services>
diff --git a/backend/danswer/document_index/vespa/index.py b/backend/danswer/document_index/vespa/index.py
@@ -815,12 +815,13 @@ def ensure_indices_exist(
             os.getcwd(), "danswer", "document_index", "vespa", "app_config"
         )
         schema_file = os.path.join(vespa_schema_path, "schemas", "danswer_chunk.sd")
-        services_file = os.path.join(vespa_schema_path, "services.xml")
         overrides_file = os.path.join(vespa_schema_path, "validation-overrides.xml")
         if ENVIRONMENT in "DEV":
             hosts_file = os.path.join(vespa_schema_path, "hosts-dev.xml")
+            services_file = os.path.join(vespa_schema_path, "services-dev.xml")
         else:
             hosts_file = os.path.join(vespa_schema_path, "hosts.xml")
+            services_file = os.path.join(vespa_schema_path, "services.xml")
 
         with open(services_file, "r") as services_f:
             services_template = services_f.read()

diff --git a/deployment/docker_compose/docker-compose.dev.yml b/deployment/docker_compose/docker-compose.dev.yml
@@ -17,9 +17,10 @@ services:
     ports:
       - "8080:8080"
     environment:
+      - ENVIRONMENT=${ENVIRONMENT:-DEV} 
       # Auth Settings
       - AUTH_TYPE=${AUTH_TYPE:-disabled}
-      - SESSION_EXPIRE_TIME_SECONDS=${SESSION_EXPIRE_TIME_SECONDS:-86400}
+      - SESSION_EXPIRE_TIME_SECONDS=${SESSION_EXPIRE_TIME_SECONDS:-86400}   
       - ENCRYPTION_KEY_SECRET=${ENCRYPTION_KEY_SECRET:-}
       - VALID_EMAIL_DOMAINS=${VALID_EMAIL_DOMAINS:-}
       - GOOGLE_OAUTH_CLIENT_ID=${GOOGLE_OAUTH_CLIENT_ID:-}
@@ -72,6 +73,7 @@ services:
       # Other services
       - POSTGRES_HOST=relational_db
       - VESPA_HOST=index
+      - VESPA_FEED_PORT=${VESPA_FEED_PORT:-}
       - WEB_DOMAIN=${WEB_DOMAIN:-}  # For frontend redirect auth purpose
       # Don't change the NLP model configs unless you know what you're doing
       - DOCUMENT_ENCODER_MODEL=${DOCUMENT_ENCODER_MODEL:-}
@@ -119,6 +121,7 @@ services:
       - indexing_model_server
     restart: always
     environment:
+      - ENVIRONMENT=${ENVIRONMENT:-DEV}
       - ENCRYPTION_KEY_SECRET=${ENCRYPTION_KEY_SECRET:-}
       # Gen AI Settings (Needed by DanswerBot)
       - GEN_AI_MODEL_PROVIDER=${GEN_AI_MODEL_PROVIDER:-}
@@ -157,6 +160,9 @@ services:
       - POSTGRES_PASSWORD=${POSTGRES_PASSWORD:-}
       - POSTGRES_DB=${POSTGRES_DB:-}
       - VESPA_HOST=index
+      - VESPA_PORT=${VESPA_PORT:-}
+      - VESPA_FEED_PORT=${VESPA_FEED_PORT:-}
+      - VESPA_FEED_HOST=${VESPA_FEED_HOST:-}
       - WEB_DOMAIN=${WEB_DOMAIN:-}  # For frontend redirect auth purpose for OAuth2 connectors
       # Don't change the NLP model configs unless you know what you're doing
       - DOCUMENT_ENCODER_MODEL=${DOCUMENT_ENCODER_MODEL:-}
@@ -231,6 +237,7 @@ services:
       - api_server
     restart: always
     environment:
+      - ENVIRONMENT=${ENVIRONMENT:-DEV}
       - INTERNAL_URL=http://api_server:8080
       - WEB_DOMAIN=${WEB_DOMAIN:-}
       - THEME_IS_DARK=${THEME_IS_DARK:-}
@@ -240,7 +247,7 @@ services:
 
 
   inference_model_server:
-    image: danswer/danswer-model-server:latest
+    image: danswer/danswer-model-server:v0.4.28
     build:
       context: ../../backend
       dockerfile: Dockerfile.model_server
@@ -253,6 +260,7 @@ services:
       fi"
     restart: on-failure
     environment:
+      - ENVIRONMENT=${ENVIRONMENT:-DEV}
       - MIN_THREADS_ML_MODELS=${MIN_THREADS_ML_MODELS:-}
       # Set to debug to get more fine-grained logs
       - LOG_LEVEL=${LOG_LEVEL:-info}
@@ -280,6 +288,7 @@ services:
       fi"
     restart: on-failure
     environment:
+      - ENVIRONMENT=${ENVIRONMENT:-DEV}
       - MIN_THREADS_ML_MODELS=${MIN_THREADS_ML_MODELS:-}
       - INDEXING_ONLY=True
       # Set to debug to get more fine-grained logs
@@ -297,6 +306,7 @@ services:
     image: postgres:15.2-alpine
     restart: always
     environment:
+      - ENVIRONMENT=${ENVIRONMENT:-DEV}
       - POSTGRES_USER=${POSTGRES_USER:-postgres}
       - POSTGRES_PASSWORD=${POSTGRES_PASSWORD:-password}
     ports:
@@ -330,6 +340,7 @@ services:
       - api_server
       - web_server
     environment:
+      - ENVIRONMENT=${ENVIRONMENT:-DEV}
       - DOMAIN=localhost
     ports:
       - "80:80"