https://github.com/rishang/aws-autoscale-warmup
Aws custom one time shedules for autoscaleing ec2 and ecs services
https://github.com/rishang/aws-autoscale-warmup
automation autoscaling aws ec2 fargate serverless
Last synced: about 2 months ago
JSON representation
Aws custom one time shedules for autoscaleing ec2 and ecs services
- Host: GitHub
- URL: https://github.com/rishang/aws-autoscale-warmup
- Owner: Rishang
- Created: 2022-01-12T13:39:17.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2022-02-20T03:32:04.000Z (over 4 years ago)
- Last Synced: 2026-04-09T00:28:10.494Z (2 months ago)
- Topics: automation, autoscaling, aws, ec2, fargate, serverless
- Language: Python
- Homepage:
- Size: 25.4 KB
- Stars: 0
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# autoscaling warmup for ec2/fargate
---
# Problem statement
---
There are 2 parts in order to perform this process
1. `rulegen` serverless code to shedule cloudwatch events for autoscaling.
2. `rulegen-scale` serverless code which gets invoked by `rulegen` cloudwatch rules to perform autoscaling on defined autoscaling groups.
## logical flow
- sheduled events for multiple resource and time, autoscaling will be defined by use in for of json as the syntax defined in [rulegen](https://github.com/Rishang/onetime-aws-autoscale/tree/main/event_rule_gen)
- This json will be uploaded to s3 bucket, before uploading bucket-arn and file name has to be defined in `serverless.yml` file as defined in [rulegen](https://github.com/Rishang/onetime-aws-autoscale/tree/main/event_rule_gen)
- On s3 bucket put event for that file `rulegen` lambda will generate cloudwatch event rules based on shedules and minimum count defined in json file by the user.
- This rules will trigger next lambda [rulegen-scale](https://github.com/Rishang/onetime-aws-autoscale/tree/main/scale) which will perform task of modifing autoscaling minimum count of ec2/fargate and after updating autoscaling count, event rule will be removed as it was been made to run once only.
## Flow Diagram
## prerequisite
1. [Nodejs](https://nodejs.org/en/)
2. [npm](http://npmjs.org/install.sh)
To install nodejs [click here](https://github.com/nodesource/distributions)
---
## Steps for implementation
1. Install serverless on your local device.
- `sudo npm install -G serverless`
- Now inside git repo directory run: `npm i -D serverless-dotenv-plugin`
2. Create a s3 bucket on your aws account where you will upload your sheduled scaling json file.
3. provide required example variables values at `.env.dev` file as follows
ENV_AWS_REGION=""
ENV_AWS_PROFILE=""
BUCKET_NAME="shedule-event-ec2"
BUCKET_FILE="test.json"
4. run `serverless deploy`
This now in you aws account in s3 bucket `shedule-event-ec2` you can put scaling time shedules in file [example/shedule.json](https://github.com/Rishang/onetime-aws-autoscale/blob/main/example/shedule.json)
same can be written in yaml as well like: [example/shedule.yml](https://github.com/Rishang/onetime-aws-autoscale/blob/main/example/shedule.yml)
like follows:
```yaml
---
- service_type: fargate
scaling_group_name: service/cluster-wp/fastapi
shedule_table:
date: 30/11/2021
shedule:
- start: '16:40:00'
end: '16:57:00'
count: 4
- start: '16:55:00'
end: '17:57:00'
count: 6
- service_type: ec2
scaling_group_name: scl
shedule_table:
date: 30/11/2021
shedule:
- start: '16:10:00'
end: '17:37:00'
count: 4
- start: '17:35:00'
end: '18:57:00'
count: 6
```