cody/docker-swarm-autoscaler
1
0
Fork 0
mirror of https://github.com/jcwimer/docker-swarm-autoscaler synced 2026-03-24 23:04:43 +00:00
Code Issues Releases Wiki Activity

Changed naming

Browse Source
This commit is contained in:
Jacob Cody Wimer
2018-10-29 07:43:34 -04:00
parent 33c3b17a09
commit f4db193d57
6 changed files with 21 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
README.md
View File

@@ -1,18 +1,18 @@
# autoscale-docker-swarm
# docker-swarm-autoscaler
This project is intended to bring auto service staling to Docker Swarm. This script uses prometheus paired with cadvisor metrics to determine cpu usage. It then uses a manager node to determine if a service wants to be autoscaled and uses a manager node to scale the service.
Currently the project only uses cpu to autoscale. If cpu usage reaches 85% the service will scale up, if it reaches 25% it will scale down.
## Usage
1. You can deploy prometheus, cadvisor, and docker-swarm-autoscale by running `docker stack deploy -c swarm-autoscale-stack.yml`.
* You can also utilize an already deploy prometheus and cadvisor by specifying the `PROMETHEUS_URL` in docker-swarm-autoscale environment. `swarm-autoscale-stack.yml` shows an example of this.
* docker-swarm-autoscale needs a placement contstraint to deploy to a manager. `swarm-autoscale-stack.yml` shows an example of this.
2. For services you want to autoscale you will need a deploy label `cpu.autoscale=true`.
1. You can deploy prometheus, cadvisor, and docker-swarm-autoscaler by running `docker stack deploy -c swarm-autoscaler-stack.yml autoscaler`.
* You can also utilize an already deploy prometheus and cadvisor by specifying the `PROMETHEUS_URL` in docker-swarm-autoscaler environment. `swarm-autoscaler-stack.yml` shows an example of this.
* docker-swarm-autoscale needs a placement contstraint to deploy to a manager. `swarm-autoscaler-stack.yml` shows an example of this.
2. For services you want to autoscale you will need a deploy label `swarm.autoscaler=true`.
```
deploy:
labels:
- "cpu.autoscale=true"
- "swarm.autoscaler=true"
```
This is best paired with resource constraints limits. This is also under the deploy key.
@@ -31,6 +31,6 @@ deploy:
## Configuration
| Setting | Value | Description |
| --- | --- | --- |
| `cpu.autoscale` | `true` | Required. This enables autoscaling for a service. Anything other than `true` will not enable it |
| `cpu.autoscale.minimum` | Integer | Optional. This is the minimum number of replicas wanted for a service. The autoscaler will not downscale below this number |
| `cpu.autoscale.maximum` | Integer | Optional. This is the maximum number of replicas wanted for a service. The autoscaler will not scale up past this number |
| `swarm.autoscaler` | `true` | Required. This enables autoscaling for a service. Anything other than `true` will not enable it |
| `swarm.autoscaler.minimum` | Integer | Optional. This is the minimum number of replicas wanted for a service. The autoscaler will not downscale below this number |
| `swarm.autoscaler.maximum` | Integer | Optional. This is the maximum number of replicas wanted for a service. The autoscaler will not scale up past this number |