Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/tsuru/tsuru-autoscale
auto scale as a service
https://github.com/tsuru/tsuru-autoscale
Last synced: 3 months ago
JSON representation
auto scale as a service
- Host: GitHub
- URL: https://github.com/tsuru/tsuru-autoscale
- Owner: tsuru
- License: bsd-3-clause
- Created: 2015-02-21T22:05:37.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2019-02-28T13:15:33.000Z (over 5 years ago)
- Last Synced: 2024-01-08T01:45:11.670Z (6 months ago)
- Language: Go
- Homepage:
- Size: 920 KB
- Stars: 32
- Watchers: 14
- Forks: 7
- Open Issues: 9
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Lists
- awesome-tsuru - autoscale - Autoscale as a service (Extensions / Services)
README
# tsuru-autoscale
[](https://travis-ci.org/tsuru/tsuru-autoscale)
[](https://goreportcard.com/report/github.com/tsuru/tsuru-autoscale)
[](https://codecov.io/gh/tsuru/tsuru-autoscale)
## Features
* HTTP api to manage auto scale configuration
* integration with [tsuru](https://tsuru.io)
* Wizard that makes easy configure an auto scale for [tsuru](https://tsuru.io)
applications
* Support HTTP based data sources
* Support HTTP based actions
* Support alarm scripts in javascript
## Concepts
The `tsuru-autoscale` is based on three elements: `data sources`, `actions` and `alarms`.
### Data sources
Data source is a http endpoint that provide the data to an alarm. Is based on the
data source data that the alarm will execute an action.
### Actions
Action is a http endpoint that is called when the alarm expression result is `true`.
### Alarms
Alarm is composed by data sources, actions and by an expression. When the expression result is `true` the actions will be executed.
### Wizard
Wizard is an easy way to use autoscale with `tsuru`. Wizard creates the alarms
for scale up and scale down, based on simple inputs like: ``
## Install as tsuru application
### Create tsuru app using Go platform
```
tsuru app-create autoscale go
```
### Configuring MongoDB
We should use environment variables to configure the database:
```
tsuru env-set "MONGODB_URL=mongodb://172.17.0.1:27017/tsuru_autoscale" -a autoscale
```
### Deploy the applications
```
tsuru env-set GO_PKG_PATH=github.com/tsuru/tsuru-autoscale -a autoscale
```
```
tsuru app-deploy . -a autoscale
```
## API Reference
### list data sources
```
curl /datasource
```
### add data source
```
curl -XPOST -d '{}' -H "Content-Type: application/json" /datasource
```
### remove a data source
```
curl -XDELETE /datasource/{name}
```
### list actions
```
curl /action
```
### add an action
```
curl -XPOST -d '{}' -H "Content-Type: applicaiton/json" /action
```
### remove an action
```
curl -XDELETE /action/{name}
```
### list alarms
```
curl /alarm
```
### add an alarm
```
curl -XPOST -d '{}' -H "Content-Type: application/json" /alarm
```
### remove an alarm
```
curl -XDELETE /alarm/{name}
```
## Configuring Wizard to works with tsuru
To `wizard` works fine with `tsuru` it is necessary to configure some data sources
and the actions to scale up and scale down.
### Add the scale up action
```
curl -XPOST -d '{"name": "scale_up", "url": "http:///apps/{app}/units", "method": "PUT", "body": "units={step}&process={process}", "headers": {"Authorization": "bearer ", "Content-Type": "application/x-www-form-urlencoded"}}' -H "Content-Type: application/json" /action
```
### Add the scale down action
```
curl -XPOST -d '{"name": "scale_down", "url": "http:///apps/{app}/units?units={step}&process={process}", "method": "DELETE", "headers": {"Authorization": "bearer ", "Content-Type": "application/x-www-form-urlencoded"}}' -H "Content-Type: application/json" /action
```
### Add data source to get the number of units
```
curl -XPOST -d '{"name": "units", "url": "http:///apps/{app}", "method": "GET", "headers" : {"Authorization": "bearer "}}' -H "Content-Type: application/json" /datasource
```
### Add data source to get cpu data from ElasticSearch
Only configure it if you are using ElasticSearch as tsuru metrics backend.
```bash
curl -XPOST -d '{"name": "cpu", "url": "http:////cpu_max/_search", "method": "POST", "body" : "{\"size\":0, \"query\": {\"filtered\": {\"filter\": {\"bool\": {\"must\": [{\"range\": {\"value\": {\"lt\": 500}}},{ \"term\": {\"app.raw\": \"{app}\"}}, {\"term\": {\"process.raw\": \"{process}\"}}]}}}}, \"aggs\": {\"range\": {\"date_range\": {\"field\": \"@timestamp\", \"ranges\": [{\"from\": \"now-5m/m\", \"to\": \"now\"}]}, \"aggs\": {\"date\": {\"date_histogram\": {\"field\": \"@timestamp\", \"interval\": \"1m\"}, \"aggs\": {\"max\": {\"max\": {\"field\": \"value\"}}, \"avg\": {\"avg\": {\"field\": \"value\"}}}}}}}}", "public": true}' -H "Content-Type: application/json" /datasource
```
### Add data source to get cpu data from Prometheus
Only configure it if you are using Prometheus as tsuru metrics backend.
```bash
curl -XPOST -d '{"name": "cpu", "url": "/api/v1/query?query=max(irate(container_cpu_system_seconds_total{container_label_tsuru_process_name=\"{process}\",container_label_tsuru_app_name=\"{app}\"}[1m]))*100", "public": true}' -H "Content-Type: application/json" /datasource
```