Add http request timeout, improve logging (#77)

* logging, timeout

* makefile, aws template

* shipper

* makefile, aws template

* shipper, template

* gitignore, remove git-secrets workflow, add pre commit config

* update cw makefile

* readme

* reamde bump version in links

* fix template, readme

.gitignore

@@ -1 +1,4 @@
-.prettierignore
+.prettierignore
+.idea
+.DS_Store
+*/*/dist

.pre-commit-config.yaml

@@ -0,0 +1,8 @@
+repos:
+  - repo: https://github.com/cycodehq-public/cycode-cli
+    rev: 0.1.6
+    hooks:
+      - id: cycode
+        language_version: python3
+        stages:
+          - commit

python3/cloudwatch/Makefile

@@ -1,9 +1,11 @@
 build:
 	mkdir -p dist/python3/shipper
+	mkdir -p dist/python3/custom_logger
 	cp src/lambda_function.py dist
 	cp ../shipper/shipper.py dist/python3/shipper
+	cp ../custom_logger/custom_logger.py dist/python3/custom_logger
 
-	(cd dist/ && zip -r logzio-cloudwatch-log-shipper.zip lambda_function.py python3/shipper/*)
+	(cd dist/ && zip -r logzio-cloudwatch-log-shipper.zip lambda_function.py python3/shipper/* python3/custom_logger/*)
 
 clean:
 	rm -rf dist/

python3/cloudwatch/sam-template.yaml

@@ -5,7 +5,15 @@ Parameters:
   logzioListener:
     Type: "String"
     Description: "The Logz.io listener URL for your region. You can find explanations here: https://docs.logz.io/user-guide/accounts/account-region.html" 
-    Default: ""
+    Default: 'https://listener.logz.io:8071'
+    AllowedValues:
+      - 'https://listener.logz.io:8071'
+      - 'https://listener-au.logz.io:8071'
+      - 'https://listener-ca.logz.io:8071'
+      - 'https://listener-eu.logz.io:8071'
+      - 'https://listener-nl.logz.io:8071'
+      - 'https://listener-uk.logz.io:8071'
+      - 'https://listener-wa.logz.io:8071'
 
   logzioToken:
     Type: "String"
@@ -45,6 +53,23 @@ Parameters:
     Description: "CloudWatch Log Group name from where you want to send logs."
     Default: ""
 
+  shipperLogLevel:
+    Type: "String"
+    Description: "Log level of the shipping Lambda function."
+    AllowedValues:
+      - "DEBUG"
+      - "INFO"
+      - "WARNING"
+      - "ERROR"
+      - "CRITICAL"
+    Default: "INFO"
+
+  requestTimeout:
+    Type: "Number"
+    Description: "Timeout in seconds for each http request for sending logs into logz.io"
+    MinValue: 0
+    Default: 15
+
 Outputs:
     LogzioCloudwatchLogsLambda:
       Description: "Logz.io CW logs lambda ARN"
@@ -54,12 +79,12 @@ Resources:
   LogzioCloudwatchLogsLambda:
     Type: 'AWS::Serverless::Function'
     Properties:
-      Runtime: python3.7
+      Runtime: python3.9
       Handler: lambda_function.lambda_handler
       CodeUri:
         Bucket: logzio-aws-integrations-<<REGION>>
         Key: cloudwatch-auto-deployment/<<VERSION>>/logzio-cloudwatch.zip
-      FunctionName: logzio-cloudwatch-log-shipper
+      FunctionName: !Ref AWS::StackName
       MemorySize: 512
       Timeout: 60
       Events:
@@ -75,4 +100,6 @@ Resources:
           TYPE: !Ref logzioType
           FORMAT: !Ref logzioFormat
           COMPRESS: !Ref logzioCompress
-          ENRICH: !Ref logzioEnrich
+          ENRICH: !Ref logzioEnrich
+          SHIPPER_LOG_LEVEL: !Ref shipperLogLevel
+          REQUEST_TIMEOUT: !Ref requestTimeout

python3/cloudwatch/src/lambda_function.py

@@ -1,12 +1,12 @@
 import base64
 import gzip
 import json
-import logging
 from dateutil import parser
 import os
 from io import BytesIO
 
 from python3.shipper.shipper import LogzioShipper
+from python3.custom_logger import custom_logger
 
 KEY_INDEX = 0
 VALUE_INDEX = 1
@@ -37,8 +37,7 @@
 
 
 # set logger
-logger = logging.getLogger()
-logger.setLevel(logging.INFO)
+logger = custom_logger.get_logger(__name__)
 
 
 def _extract_aws_logs_data(event):
@@ -171,7 +170,9 @@ def _get_additional_logs_data(aws_logs_data, context):
 def lambda_handler(event, context):
     # type (dict, 'LambdaContext') -> None
 
+    logger.debug(f'Handling event: {event}')
     aws_logs_data = _extract_aws_logs_data(event)
+    logger.debug(f'Logs data: {aws_logs_data}')
     additional_data = _get_additional_logs_data(aws_logs_data, context)
     shipper = LogzioShipper()
 

python3/custom_logger/custom_logger.py

@@ -0,0 +1,17 @@
+import logging
+import os
+
+LOG_LEVEL_ENV = 'SHIPPER_LOG_LEVEL'
+DEFAULT_LEVEL = 'INFO'
+
+
+def get_logger(name):
+    level_str = os.getenv(LOG_LEVEL_ENV, DEFAULT_LEVEL)
+    logger = logging.getLogger(name)
+    try:
+        level = logging.getLevelName(level_str)
+        logger.setLevel(level)
+    except Exception as e:
+        logger.setLevel(logging.INFO)
+        logger.warning(f'Could not set logger to level {level_str}, setting to default level {DEFAULT_LEVEL}: {e}')
+    return logger

python3/shipper/shipper.py

@@ -1,16 +1,16 @@
 import gzip
 import io
 import json
-import logging
 import os
 import sys
 import time
 import urllib
 import urllib.request
 
+from python3.custom_logger import custom_logger
+
 # set logger
-logger = logging.getLogger()
-logger.setLevel(logging.INFO)
+logger = custom_logger.get_logger(__name__)
 
 
 class MaxRetriesException(Exception):
@@ -125,6 +125,8 @@ class LogzioShipper(object):
     URL_ENV = 'LISTENER_URL'
     BASE_URL = "https://listener.logz.io:8071"
     region = None
+    TIMEOUT_ENV = 'REQUEST_TIMEOUT'
+    default_timeout = 15  # seconds
 
     def __init__(self):
         self._logzio_url = self.BASE_URL
@@ -152,6 +154,21 @@ def __init__(self):
             if self._compress \
             else StringLogRequest(self.MAX_BULK_SIZE_IN_BYTES)
 
+        self.timeout = self._get_timeout()
+        logger.debug(f'Request timeout is set to: {self.timeout} seconds')
+
+    def _get_timeout(self):
+        timeout_str = os.getenv(self.TIMEOUT_ENV)
+        if timeout_str != '':
+            try:
+                timeout = int(timeout_str)
+                if timeout > 0:
+                    return timeout
+                logger.warning(f'Timeout input from user is invalid, reverting to default value {self.default_timeout}')
+            except TypeError:
+                logger.warning(f'Could not parse timeout input {timeout_str}, reverting to default value {self.default_timeout}')
+        return self.default_timeout
+
     def get_base_api_url(self):
         return self.BASE_URL.replace("listener.", "listener{}.".format(self.get_region_code()))
 
@@ -221,9 +238,10 @@ def _send_to_logzio(self):
         def do_request():
             self._logs.close()
             logs = self._logs.bytes()
+            logger.info(f'About to send {len(logs)} bytes')
             request = urllib.request.Request(self._logzio_url, data=self._logs.bytes(),
                                              headers=self._logs.http_headers())
-            return urllib.request.urlopen(request)
+            return urllib.request.urlopen(request, timeout=self.timeout)
 
         try:
             do_request()