Giter Site home page Giter Site logo

swipswaps / faasd Goto Github PK

View Code? Open in Web Editor NEW

This project forked from openfaas/faasd

0.0 1.0 0.0 5.23 MB

A lightweight & portable faas engine

Home Page: https://openfaas.com/blog/introducing-faasd/

License: MIT License

Go 92.81% Makefile 2.44% Shell 4.75%

faasd's Introduction

faasd - a lightweight & portable faas engine

Build Status License: MIT OpenFaaS Downloads

faasd is OpenFaaS reimagined, but without the cost and complexity of Kubernetes. It runs on a single host with very modest requirements, making it fast and easy to manage. Under the hood it uses containerd and Container Networking Interface (CNI) along with the same core OpenFaaS components from the main project.

When should you use faasd over OpenFaaS on Kubernetes?

  • You have a cost sensitive project - run faasd on a 5-10 USD VPS or on your Raspberry Pi
  • When you just need a few functions or microservices, without the cost of a cluster
  • When you don't have the bandwidth to learn or manage Kubernetes
  • To deploy embedded apps in IoT and edge use-cases
  • To shrink-wrap applications for use with a customer or client

faasd does not create the same maintenance burden you'll find with maintaining, upgrading, and securing a Kubernetes cluster. You can deploy it and walk away, in the worst case, just deploy a new VM and deploy your functions again.

About faasd

  • is a single Golang binary
  • uses the same core components and ecosystem of OpenFaaS
  • is multi-arch, so works on Intel x86_64 and ARM out the box
  • can be set-up and left alone to run your applications

demo

Demo of faasd running in KVM

Tutorials

Run locally on MacOS, Linux, or Windows with multipass

DigitalOcean tutorial with Terraform and TLS

The terraform can be adapted for any IaaS provider:

See also: Build a Serverless appliance with faasd and cloud-init

Get started on armhf / Raspberry Pi

You can run this tutorial on your Raspberry Pi, or adapt the steps for a regular Linux VM/VPS host.

Terraform for DigitalOcean

Automate everything within < 60 seconds and get a public URL and IP address back. Customise as required, or adapt to your preferred cloud such as AWS EC2.

Operational concerns

A note on private repos / registries

To use private image repos, ~/.docker/config.json needs to be copied to /var/lib/faasd/.docker/config.json.

If you'd like to set up your own private registry, see this tutorial.

Beware that running docker login on MacOS and Windows may create an empty file with your credentials stored in the system helper.

Alternatively, use you can use the registry-login command from the OpenFaaS Cloud bootstrap tool (ofc-bootstrap):

curl -sLSf https://raw.githubusercontent.com/openfaas-incubator/ofc-bootstrap/master/get.sh | sudo sh

ofc-bootstrap registry-login --username <your-registry-username> --password-stdin
# (the enter your password and hit return)

The file will be created in ./credentials/

Note for the GitHub container registry, you should use ghcr.io Container Registry and not the previous generation of "Docker Package Registry". See notes on migrating

Logs for functions

You can view the logs of functions using journalctl:

journalctl -t openfaas-fn:FUNCTION_NAME


faas-cli store deploy figlet
journalctl -t openfaas-fn:figlet -f &
echo logs | faas-cli invoke figlet

Logs for the core services

Core services as defined in the docker-compose.yaml file are deployed as containers by faasd.

View the logs for a component by giving its NAME:

journalctl -t default:NAME

journalctl -t default:gateway

journalctl -t default:queue-worker

You can also use -f to follow the logs, or --lines to tail a number of lines, or --since to give a timeframe.

Exposing core services

The OpenFaaS stack is made up of several core services including NATS and Prometheus. You can expose these through the docker-compose.yaml file located at /var/lib/faasd.

Expose the gateway to all adapters:

  gateway:
    ports:
       - "8080:8080"

Expose Prometheus only to 127.0.0.1:

  prometheus:
    ports:
       - "127.0.0.1:9090:9090"

Upgrading faasd

To upgrade faasd either re-create your VM using Terraform, or simply replace the faasd binary with a newer one.

systemctl stop faasd-provider
systemctl stop faasd

# Replace /usr/local/bin/faasd with the desired release

# Replace /var/lib/faasd/docker-compose.yaml with the matching version for
# that release.
# Remember to keep any custom patches you make such as exposing additional 
# ports, or updating timeout values

systemctl start faasd
systemctl start faasd-provider

You could also perform this task over SSH, or use a configuration management tool.

Note: if you are using Caddy or Let's Encrypt for free SSL certificates, that you may hit rate-limits for generating new certificates if you do this too often within a given week.

Memory limits for functions

Memory limits for functions are supported. When the limit is exceeded the function will be killed.

Example:

functions:
  figlet:
    skip_build: true
    image: functions/figlet:latest
    limits:
      memory: 20Mi

What does faasd deploy?

You'll also need:

You can use the standard faas-cli along with pre-packaged functions from the Function Store, or build your own using any OpenFaaS template.

Manual / developer instructions

See here for manual / developer instructions

Getting help

Docs

The OpenFaaS docs provide a wealth of information and are kept up to date with new features.

Function and template store

For community functions see faas-cli store --help

For templates built by the community see: faas-cli template store list, you can also use the dockerfile template if you just want to migrate an existing service without the benefits of using a template.

Training and courses

LinuxFoundation training course

The founder of faasd and OpenFaaS has written a training course for the LinuxFoundation which also covers how to use OpenFaaS on Kubernetes. Much of the same concepts can be applied to faasd, and the course is free:

Community workshop

The OpenFaaS workshop is a set of 12 self-paced labs and provides a great starting point for learning the features of openfaas. Not all features will be available or usable with faasd.

Community support

An active community of almost 3000 users awaits you on Slack. Over 250 of those users are also contributors and help maintain the code.

Roadmap

Supported operations

  • faas login
  • faas up
  • faas list
  • faas describe
  • faas deploy --update=true --replace=false
  • faas invoke --async
  • faas invoke
  • faas rm
  • faas store list/deploy/inspect
  • faas version
  • faas namespace
  • faas secret
  • faas logs

Scale from and to zero is also supported. On a Dell XPS with a small, pre-pulled image unpausing an existing task took 0.19s and starting a task for a killed function took 0.39s. There may be further optimizations to be gained.

Other operations are pending development in the provider such as:

  • faas auth - supported for Basic Authentication, but SSO, OAuth2 & OIDC may require a patch

Backlog

Should have:

  • Resolve core services from functions by populating/sharing /etc/hosts between faasd and faasd-provider
  • Docs or examples on how to use the various connectors and connector-sdk
  • Monitor and restart any of the core components at runtime if the container stops
  • Asynchronous deletion instead of synchronous

Nice to Have:

  • Total memory limits - if a node has 1GB of RAM, don't allow more than 1000MB of RAM to be reserved via limits
  • Offer live rolling-updates, with zero downtime - requires moving to IDs vs. names for function containers
  • Multiple replicas per function

Known-issues

Completed

  • Provide a cloud-init configuration for faasd bootstrap
  • Configure core services from a docker-compose.yaml file
  • Store and fetch logs from the journal
  • Add support for using container images in third-party public registries
  • Add support for using container images in private third-party registries
  • Provide a cloud-config.txt file for automated deployments of faasd
  • Inject / manage IPs between core components for service to service communication - i.e. so Prometheus can scrape the OpenFaaS gateway - done via /etc/hosts mount
  • Add queue-worker and NATS
  • Create faasd.service and faasd-provider.service
  • Self-install / create systemd service via faasd install
  • Restart containers upon restart of faasd
  • Clear / remove containers and tasks with SIGTERM / SIGINT
  • Determine armhf/arm64 containers to run for gateway
  • Configure basic_auth to protect the OpenFaaS gateway and faasd-provider HTTP API
  • Setup custom working directory for faasd /var/lib/faasd/
  • Use CNI to create network namespaces and adapters
  • Optionally expose core services from the docker-compose.yaml file, locally or to all adapters.
  • containerd can't pull image from Github Docker Package Registry ghcr.io support
  • Provide simple Caddyfile example in the README showing how to expose the faasd proxy on port 80/443 with TLS
  • Annotation support
  • Hard memory limits for functions
  • Terraform for DigitalOcean
  • Store and retrieve annotations in function spec - in progress
  • An installer for faasd and dependencies - runc, containerd

WIP:

  • Terraform for AWS

faasd's People

Contributors

akihirosuda avatar akosveres avatar alexellis avatar atomic77 avatar carlosedp avatar devries avatar gabeduke avatar jsiebens avatar kadern0 avatar lucasroesler avatar martindekov avatar mehyedes avatar mrsimonemms avatar utsavanand2 avatar waterdrips avatar xinydev avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.