fix: make logs persistent + other fixes (#140)

- remove root logs dir
- revert CI env var in embedding server proc
- remove unsupported vectordb configs
- update readme: remove beta status
- update readme for logs
- fix: better handling of embedding server failure

---------

Signed-off-by: Anupam Kumar <kyteinsky@gmail.com>

Author: kyteinsky

.github/workflows/integration-test.yml

@@ -163,7 +163,7 @@ jobs:
           pip install -r requirements.txt
           cp example.env .env
           echo "NEXTCLOUD_URL=http://localhost:8080" >> .env
-          python3 -u ./main.py > logs/backend_logs 2>&1 &
+          python3 -u ./main.py > backend_logs 2>&1 &
           echo $! > ../pid.txt          # Save the process ID (PID)
 
       - name: Register backend
@@ -233,7 +233,9 @@ jobs:
       - name: Show logs
         if: always()
         run: |
-          tail data/nextcloud.log
+          cat data/nextcloud.log
           echo '--------------------------------------------------'
-          tail -v -n +1 context_chat_backend/logs/* || echo "No logs in logs directory"
+          cat context_chat_backend/backend_logs || echo "No backend logs"
+          echo '--------------------------------------------------'
+          tail -v -n +1 context_chat_backend/persistent_storage/logs/* || echo "No logs in logs directory"
           

.gitignore

@@ -5,4 +5,3 @@ __pycache__/
 .env
 persistent_storage/*
 .vscode/
-logs/

Dockerfile

@@ -40,9 +40,6 @@ RUN python3 -m pip install --no-cache-dir https://github.com/abetlen/llama-cpp-p
 RUN sed -i '/llama_cpp_python/d' requirements.txt
 RUN python3 -m pip install --no-cache-dir -r requirements.txt && python3 -m pip cache purge
 
-# Create an empty logs dir
-RUN mkdir logs
-
 # Copy application files
 COPY context_chat_backend context_chat_backend
 COPY main.py .

README.md

@@ -7,8 +7,6 @@
 [![REUSE status](https://api.reuse.software/badge/github.com/nextcloud/context_chat_backend)](https://api.reuse.software/info/github.com/nextcloud/context_chat_backend)
 
 > [!NOTE]
-> This is a beta software. Expect breaking changes.
->
 > Be mindful to install the backend before the Context Chat php app (Context Chat php app would sends all the user-accessible files to the backend for indexing in the background. It is not an issue even if the request fails to an uninitialised backend since those files would be tried again in the next background job run.)
 >
 > The HTTP request timeout is 50 minutes for all requests and can be changed with the `request_timeout` app config for the php app `context_chat` using the occ command (`occ config:app:set context_chat request_timeout --value=3000`, value is in seconds). The same also needs to be done for docker socket proxy. See [Slow responding ExApps](https://github.com/cloud-py-api/docker-socket-proxy?tab=readme-ov-file#slow-responding-exapps)
@@ -101,6 +99,11 @@ volumes:
 -v /var/run/docker.sock:/var/run/docker.sock:ro
 ```
 
+## Logs
+Logs are stored in the `logs/` directory in the persistent directory. In a docker container, it should be at `/nc_app_context_chat_backend/logs/`. The log file is named `ccb.log` and is set to otate at 20 MB with 10 backups. These logs are in JSONL format, i.e. each line is a valid JSON object.
+Now only warning and above logs are printed to the console. All the debug logs are written to the log file if `debug` is set to `true` in the config file.
+The logs of the embedding server are written to `logs/embedding_server_[date].log` in the persistent directory, it rotates with date change and is not in JSONL format, just raw stdout and stderr from the embedding server's process.
+
 ## Configuration
 Configuration resides inside the persistent storage as `config.yaml`. The location is `$APP_PERSISTENT_STORAGE`. By default it would be at `/nc_app_context_chat_backend_data/config.yaml` inside the container.
 

config.cpu.yaml

@@ -14,17 +14,6 @@ vectordb:
   pgvector:
     # 'connection' overrides the env var 'CCB_DB_URL'
 
-  chroma:
-    is_persistent: true
-    # chroma_server_host:
-    # chroma_server_http_port:
-    # chroma_server_ssl_enabled:
-    # chroma_server_api_default_path:
-
-  weaviate:
-    # auth_client_secret:
-    # url: http://localhost:8080
-
 embedding:
   protocol: http
   host: localhost

config.gpu.yaml

@@ -14,17 +14,6 @@ vectordb:
   pgvector:
     # 'connection' overrides the env var 'CCB_DB_URL'
 
-  chroma:
-    is_persistent: true
-    # chroma_server_host:
-    # chroma_server_http_port:
-    # chroma_server_ssl_enabled:
-    # chroma_server_api_default_path:
-
-  weaviate:
-    # auth_client_secret:
-    # url: http://localhost:8080
-
 embedding:
   protocol: http
   host: localhost

context_chat_backend/controller.py

@@ -351,6 +351,8 @@ def _(sources: list[UploadFile]):
 
 	try:
 		added_sources = exec_in_proc(target=embed_sources, args=(vectordb_loader, app.extra['CONFIG'], sources))
+	except (DbException, EmbeddingException) as e:
+		raise e
 	except Exception as e:
 		raise DbException('Error: failed to load sources') from e
 	finally:

context_chat_backend/dyn_loader.py

@@ -11,6 +11,7 @@
 import signal
 import subprocess
 from abc import ABC, abstractmethod
+from datetime import datetime
 from time import sleep, time
 from typing import Any
 
@@ -22,7 +23,7 @@
 
 from .models.loader import init_model
 from .network_em import NetworkEmbeddings
-from .types import LoaderException, TConfig
+from .types import EmbeddingException, LoaderException, TConfig
 from .vectordb.base import BaseVectorDB
 from .vectordb.loader import get_vector_db
 from .vectordb.types import DbException
@@ -43,7 +44,11 @@ def offload(self):
 class EmbeddingModelLoader(Loader):
 	def __init__(self, config: TConfig):
 		self.config = config
-		self.logfile = open('logs/embedding_server.log', 'a+')
+		logfile_path = os.path.join(
+			os.environ['EM_SERVER_LOG_PATH'],
+			f'embedding_server_{datetime.now().strftime("%Y-%m-%d")}.log',
+		)
+		self.logfile = open(logfile_path, 'a+')
 
 	def load(self):
 		global pid
@@ -84,8 +89,7 @@ def load(self):
 				try_ += 1
 				sleep(3)
 
-		logger.error('Error: failed to start the embedding server')
-		os.kill(os.getpid(), signal.SIGTERM)
+		raise EmbeddingException('Error: the embedding server is not responding')
 
 	def offload(self):
 		global pid

context_chat_backend/logger.py

@@ -9,6 +9,7 @@
 import logging
 import logging.config
 import logging.handlers
+import os
 from time import gmtime
 
 from ruamel.yaml import YAML
@@ -88,6 +89,19 @@ def get_logging_config() -> dict:
 		try:
 			yaml = YAML(typ='safe')
 			config: dict = yaml.load(f)
+
+			persistent_storage = os.getenv('APP_PERSISTENT_STORAGE', 'persistent_storage')
+			if (config.get('handlers', {}).get('file_json', {}).get('filename')):
+				if (
+					not config['handlers']['file_json']['filename'].startswith(persistent_storage)
+					and not config['handlers']['file_json']['filename'].startswith('/')
+				):
+					config['handlers']['file_json']['filename'] = os.path.join(
+						persistent_storage,
+						config['handlers']['file_json']['filename'],
+					)
+				# create logs directory if it doesn't exist
+				os.makedirs(os.path.dirname(config['handlers']['file_json']['filename']), exist_ok=True)
 		except Exception as e:
 			raise AssertionError('Error: could not load config from logger_config.yaml file') from e
 

context_chat_backend/setup_functions.py

@@ -44,9 +44,14 @@ def setup_env_vars():
 
 	config_path = os.path.join(persistent_storage, 'config.yaml')
 
+	em_server_log_path = os.path.join(persistent_storage, 'logs')
+	if not os.path.exists(em_server_log_path):
+		os.makedirs(em_server_log_path, 0o750, True)
+
 	os.environ['APP_PERSISTENT_STORAGE'] = persistent_storage
 	os.environ['VECTORDB_DIR'] = vector_db_dir
 	os.environ['MODEL_DIR'] = model_dir
 	os.environ['SENTENCE_TRANSFORMERS_HOME'] = os.getenv('SENTENCE_TRANSFORMERS_HOME', model_dir)
 	os.environ['HF_HOME'] = os.getenv('HF_HOME', model_dir)
 	os.environ['CC_CONFIG_PATH'] = os.getenv('CC_CONFIG_PATH', config_path)
+	os.environ['EM_SERVER_LOG_PATH'] = os.getenv('EM_SERVER_LOG_PATH', em_server_log_path)