top of page

How to be familiar with Lambda - Create error notification



Intro: First step to AWS Lambda


Not many people choose Lambda as their new service

Lambda is a compute service that lets you run without provisioning or managing servers.

It is an essential service for automating AWS environments

I learned AWS Lambda with little experiences about code

But the advantage of being able to customize the various environments of AWS for automation and integration, motivated and challenged me to use lambda



 


Load CloudWatch Metric Index through Lambda


[Summary]

We worked on cloning Lambda for the first time

CloudWatch is the most used AWS service for monitoring AWS infra environment

We synced this service and lambda for delivering monitoring index to community messenger called slack



[Purposes]

  • Deliver CloudWatch metric index to slack channel by running Lambda code

  • If we achieve that goal, we don’t need to check CloudWatch monitoring information by accessing user account console.

  • We can implement the system that can directly requests to slack about CloudWatch information of various accounts.

  • It detects disabilities and enhances the readability by loading each essential CloudWatch metric index.


[Implementation]

<Set up modules to use Lambda>

1. pip Package management system

  • Install pip package to set up modules for training

C:\Users\user>curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
C:\Users\user>python get-pip.py
C:\Users\user>python -m pip install –upgrade pip

2. Essential modules Installation

C:\Users\user>pip install boto3
                          certifi
                          chardet
                          idna
                          json
                          requests

3. Compress(.zip) modules to use Lambda

C:\Users\your account\AppData\Local\Programs\Python\Python37\Lib\site-packages



<Create IAM roles to use Lambda>

1. Create IAM policies

  • IAM - Policy - Create Policy

  • Paste Jason policy code

"Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "VisualEditor0", 
            "Effect": "Allow", // 
            "Action": [ 
                "cloudwatch:DescribeAlarmHistory",
                "cloudwatch:GetDashboard",
                "cloudwatch:GetMetricData",
                "cloudwatch:DescribeAlarmsForMetric",
                "cloudwatch:DescribeAlarms",
                "cloudwatch:GetMetricStatistics",
                "cloudwatch:GetMetricWidgetImage"
            ],
            "Resource": "*" 
        }
    ]
}

2. Create IAM roles

  • IAM - roles - create roles

  • Lambda - choose created policies - create roles


<Write Lambda functions>

1. Create Lambda functions

  • Lambda - functions - create functions

  • write new - function names - runtime - running roles - existing roles

- Create random function names

- Choose the language and version when you write functions

- Apply roles that you created


2. Slack Settings

  • Slack - Apps - Create Bots - Copy API Token


3. Write Lambda functions

  • CloudWatch - Index - Choose Index - Copy sources

"metrics": [
        [ "AWS/EC2", "CPUUtilization", "InstanceId", "i-0cf3d15721cedbf7e" ]
    ],
    "view": "timeSeries",
    "stacked": False, 
    "stat": "Maximum",
    "period": 300,
    "width": 914,
    "height": 250,
    "start": "-PT3H",
    "end": "P0D",
    "timezone": "+0900"
}

- false → False to correct the grammar

  • function codes - work - modules(.zip)upload


  • Create files - Write sample codes

import json
import boto3
import requests
 
cloudwatch = boto3.client('cloudwatch')
metric_sources = [
{
        "metrics":
        [ "AWS/EC2", "CPUUtilization"
        ],
        "view": "timeSeries",
        "stacked": False,
        "region": "ap-northeast-2",
        "timezone": "+0900"
    }
]
 
def metric(event, context):
    for metric_data in metric_sources :
        metric_data = json.dumps(metric_data)
         
        #print(metric_data)
        image_data = cloudwatch.get_metric_widget_image(MetricWidget=metric_data)
        #print(image_data)

        slack_params = {
            "filename":"test.png",
            "token":"SLACK BOT의 token 값을 넣어 주시면 됩니다",
            "channels":['#aws-db-slowquery-noti <-와 같이 채널 명칭을 넣어주세요.']
        }
        image = {'file': image_data['MetricWidgetImage']}
         
        requests.post("https://slack.com/api/files.upload", params=slack_params, files=image)

-Paste CloudWatch metric sources to metric_sources

-Paste Token value of token Slack Bots

-Type channels name (ex. #awschannels)



4. Lambda functions test

  • Lambda - Choose functions - Test

-Set-up CloudWatch metric information will be printed out when functions run



 


Review: Potentials of Lambda


With this training process, We sync AWS services to trigger functions automatically to use AWS Lambda more flexibly and we will implement methods to use AWS Lambda more easily.

Lambda is an essential AWS service to implement the automation environment.

Our final goal is to get custom abilities for the automation environment and we need to get used to Lambda by practicing it.




References



72 views0 comments

Comments


bottom of page