Doccano gb connection

README.md

@@ -226,7 +226,31 @@ Doing this in us-east-1 - Virginia and used the base name `doccano`, so for inst
 ### Populate Secrets needed by the EC2
 - add useful secrets to secrets manager:
   - quay_io_creds (quay.io login creds)
-  - doccano_creds (all the information needed in the .env file, mostly doccano and DB credentials)
+    ```
+    {
+      "user":"",
+      "password":""
+    }
+    ```
+  - doccano_creds
+    ```
+    {
+      "ADMIN_PASSWORD":"",
+      "RABBITMQ_DEFAULT_PASS":"",
+      "POSTGRES_PASSWORD":"",
+      "FLOWER_BASIC_AUTH":"",
+      "POSTGRES_HOST":""
+    }
+    ```
+  - gearbox_creds
+    ```
+    {
+      "FENCE_CLIENT_ID":"",
+      "FENCE_CLIENT_SECRET":"",
+      "FENCE_TOKEN_URL":"https://gearbox-dev.pedscommons.org/user/oauth2/token?grant_type=client_credentials&scope=openid",
+      "GEARBOX_RAW_CRITERIA_URL":"https://gearbox-dev.pedscommons.org/gearbox/raw-criteria"
+    }
+    ```
 
 ### Create VPC
 Select the following options:
@@ -292,7 +316,7 @@ and add the ec2 created previously
 - add listener to listen to 80 and redirect (redirect to url) 301 to 443 - Redirect to HTTPS://#{host}:443/#{path}?#{query}
 
 ### Deploy new version
-- Update the ec2_user_data.sh file with the new tag
+- Update the docker-compose.prod.yml file with the new tag/s
 - Repeat the step `Create EC2 instance`
 - Repeat the step `Add instance/s to the target group`
 - Remove old instance from the target group

backend/api/gearbox_client.py

@@ -30,5 +30,9 @@ def submit_to_gearbox(jsonl_content: bytes, filename: str = "annotations.jsonl")
         files={"file": ("annotations.zip", zip_buffer, "application/zip")},
         timeout=30,
     )
-    response.raise_for_status()
+    try:
+        response.raise_for_status()
+    except requests.exceptions.HTTPError as e:
+        logger.error("Gearbox returned %s: %s", e.response.status_code, e.response.text[:500])
+        raise
     return response

backend/examples/celery_tasks.py

@@ -39,6 +39,9 @@ def _transform_to_gearbox_format(jsonl_bytes: bytes) -> bytes:
         entities = []
         pre_annotated = []
         for entry in raw_labels:
+            if len(entry) < 3:
+                logger.warning("Skipping malformed label entry for example: %s", entry)
+                continue
             start_offset, end_offset, label_name = entry[0], entry[1], entry[2]
             meta = entry[3] if len(entry) > 3 else {}
             entities.append(
@@ -92,6 +95,11 @@ def submit_example_to_gearbox(self, example_id, project_id):
             example.gearbox_status = "failed"
             example.save(update_fields=["gearbox_status"])
         raise
+    except Exception:
+        example.gearbox_status = "failed"
+        example.save(update_fields=["gearbox_status"])
+        logger.exception("Unexpected error submitting example %s to gearbox", example.pk)
+        raise
 
     example.gearbox_status = "success"
     example.save(update_fields=["gearbox_status"])

docker/docker-compose.prod.yml

@@ -2,7 +2,7 @@ version: "3.7"
 services:
 
   backend:
-    image: quay.io/pcdc/doccano:be_20250618
+    image: quay.io/pcdc/doccano:be_20260701
     volumes:
       - static_volume:/backend/staticfiles
       - media:/backend/media
@@ -27,7 +27,7 @@ services:
       - network-frontend
 
   celery:
-    image: quay.io/pcdc/doccano:be_20250618
+    image: quay.io/pcdc/doccano:be_20260701
     volumes:
       - media:/backend/media
       - tmp_file:/backend/filepond-temp-uploads
@@ -48,7 +48,7 @@ services:
       - network-backend
 
   flower:
-    image: quay.io/pcdc/doccano:be_20250618
+    image: quay.io/pcdc/doccano:be_20260701
     entrypoint: ["/opt/bin/prod-flower.sh"]
     environment:
       PYTHONUNBUFFERED: "1"
@@ -75,7 +75,7 @@ services:
       - network-backend
 
   nginx:
-    image: quay.io/pcdc/doccano:fe_20250618
+    image: quay.io/pcdc/doccano:fe_20260701
     command: >
       /bin/sh -c
       "envsubst '

ec2_user_data.sh

@@ -25,6 +25,14 @@ POSTGRES_HOST="$(echo $doccano_secrets | jq -r '.POSTGRES_HOST')"
 ENCODED_POSTGRES_PASSWORD=$(jq -rn --arg pwd $POSTGRES_PASSWORD '$pwd|@uri')
 
 
+gearbox_secrets_str=$(aws secretsmanager get-secret-value --secret-id gearbox_creds)
+gearbox_secrets="$(echo $gearbox_secrets_str | jq '.SecretString' | jq '. | fromjson')"
+FENCE_CLIENT_ID="$(echo $gearbox_secrets | jq -r '.FENCE_CLIENT_ID')"
+FENCE_CLIENT_SECRET="$(echo $gearbox_secrets | jq -r '.FENCE_CLIENT_SECRET')"
+FENCE_TOKEN_URL="$(echo $gearbox_secrets | jq -r '.FENCE_TOKEN_URL')"
+GEARBOX_RAW_CRITERIA_URL="$(echo $gearbox_secrets | jq -r '.GEARBOX_RAW_CRITERIA_URL')"
+
+
 #define the template.
 env_file=$(cat  << EOF
 # platform settings
@@ -44,6 +52,12 @@ POSTGRES_DB=postgres
 
 # Flower settings
 FLOWER_BASIC_AUTH='$FLOWER_BASIC_AUTH'
+
+# GEARBOx settings
+FENCE_CLIENT_ID='$FENCE_CLIENT_ID'
+FENCE_CLIENT_SECRET='$FENCE_CLIENT_SECRET'
+FENCE_TOKEN_URL='$FENCE_TOKEN_URL'
+GEARBOX_RAW_CRITERIA_URL='$GEARBOX_RAW_CRITERIA_URL'
 EOF
 )
 echo "$env_file" > ./.env
@@ -54,13 +68,3 @@ docker login quay.io -p "$(echo $quay_secrets | jq '.SecretString' | jq '. | fro
 sudo docker-compose -f docker/docker-compose.prod.yml --env-file .env up -d
 
 
-
-#old command
-# sed -i s/password/D4cGTech/g .env
-# sed -i s/admin/admin/g .env
-# sed -i s/FLOWER_BASIC_AUTH=""/FLOWER_BASIC_AUTH="admin:D4cGTech"/g .env
-# sed -i s/FLOWER_BASIC_AUTH=""/FLOWER_BASIC_AUTH=\"admin:D4cGTech\"/g .env
-# sed -i s/FLOWER_BASIC_AUTH=\"\"/FLOWER_BASIC_AUTH=\"admin:D4cGTech\"/g .env
-
-
-