Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/hyperledgendary/hlfsupport-in-a-box
https://github.com/hyperledgendary/hlfsupport-in-a-box
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/hyperledgendary/hlfsupport-in-a-box
- Owner: hyperledgendary
- License: apache-2.0
- Created: 2021-09-10T16:14:26.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2022-03-17T10:03:39.000Z (almost 3 years ago)
- Last Synced: 2024-04-17T19:39:30.134Z (10 months ago)
- Language: Shell
- Size: 149 KB
- Stars: 1
- Watchers: 4
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# HLF-Support-In-A-Box
A single docker image with the Ansible and configuration scripts required to setup the HLFSupport images in an Open Shift or K8S Cluster
along with a test Fabric Network for 2 organizations.
Everything is contained within a single docker image; this needs a configuration file to point to the cluster, and give the entitlement
key to the IBM Container Library.
A local directory is also mapped that will contain connection information for applications etc.
---
*For IBM Software Library Images, your entitlement keys can be obtained from [this site](https://myibm.ibm.com/products-services/containerlibrary)*
---
**Uber-quick-start**
- Get a fresh provisioned Linux machine. As this is going to run a cluster locally it needs to have about 8CPUs, 16Gb memory.
- This will install and setup the Docker, Kind, and basic config for you
```bash
curl -sSL https://raw.githubusercontent.com/hyperledgendary/hlfsupport-in-a-box/main/prereqs.sh -o prereqs.sh && chmod +x prereqs.sh
./prereqs.sh
```
- Add you email and entitlement key to the `cfg.env` file
- Install the HLF support into the local Kind cluster
```bash
curl -sSL https://raw.githubusercontent.com/hyperledgendary/hlfsupport-in-a-box/main/unbox.sh -o unbox.sh && chmod +x unbox.sh
./unbox.sh
```
## Using IBM Cloud Transient Virtual Servers
The IBM Cloud Classic Infrastructure has a class of virtual services defined as "Transient". These are cheap, and fine for exploration usage.
When creating these, it's suggested to use the 8 vCPUs / 16 GB memory. It's worth getting the 100Gb storage just in case.
Select the Ubuntu 20.04 LTS miniaml as the operating system.
## Detailed Requirements
If you want to use an existing system these are the tools that you will need
- Either provisioned OpenShift cluster or a locally running k8s cluster
- Entitlement credentials for the HLFSupport images
- Docker installed locally
- Suggested that you have some tooling-of-choice to work with your k8s cluster.
- Recommended that you create a directory to work in
- If you want to provision a Linux VM then 8 vCPU, 16Gb Memory are worth getting.
## Configuration for a Cloud-hosted Cluster
A `cfg.env` file needs to be created that contains the details of the provisioned OpenShift Cluster, along with
credentials to the docker registry with the images.
This is the example `cfg.env`
```bash
# iks & ocp are all accepted values
#
# iks will assumet the current kubectl context is the one to use
# ocp will use the `oc` cli to login to the cluster
CLUSTER_TYPE=iks
# For OCP token authentication use these settings
OCP_TOKEN_SERVER=api.gabbros.cp.fyre.ibm.com:6443
OCP_TOKEN=sha256~XX
# username/password authentication is possible, comment out the OCP_TOKEN and use these settings
# OCP_URL=uuuu
# OCP_PASSWORD=pppp
# OCP_USERNAME=xxxx
# The domain where the console is to be hosted
CONSOLE_DOMAIN=apps.gabbros.cp.fyre.ibm.com
# Depending on the cluster you might need to alter the storage class being used
# CONSOLE_STORAGE_CLASS=standard
# Credentials to access the IBM Container Library
[email protected]
DOCKER_PW=xxxxxxxxx
```
Create the cfg.env file in the new empty directory
```bash
mkdir hlf && cd hlf
touch cfg.env
# and edit
```
## Configuration for a locally-hosted cluster
This can be run entirely locally on one machine, this is very good for demos and exploration.
The suggestion is to setup a KIND based k8s cluster. The simplest way to get a good KIND configuration is to use the `test-network-k8s` from the `hyperledger/fabric-samples` github repository
```bash
git clone https://github.com/hyperledger/fabric-samples.git
cd fabric-samples/test-network-k8s
./network kind
```
This setups your local kubcetl context; this means that the `cfg.env` file you need to created is similar to this.
Note that the `console-domain` has an IP address that will needs to resolve to the host that you are running KIND on. A quick way to get this is to issues `hostname -I` and use the first IP address returned.
```bash
CLUSTER_TYPE=iks
CONSOLE_DOMAIN=172-22-140-212.nip.io
CONSOLE_STORAGE_CLASS=standard
[email protected]
DOCKER_PW=
```
## Storage Classes
Depending on creation, the storage classes may might need to be configured. The `scripts/setup_storage_classes.sh` can be used in this.
## Running
- Get the 'unbox.sh' script... this has the docker run commands in it to save typing
```
curl -sSL https://raw.githubusercontent.com/hyperledgendary/hlfsupport-in-a-box/main/unbox.sh -o unbox.sh && chmod +x unbox.sh
```
- Run the unbox script
```bash
./unbox.sh
```
The process will take about 10 minutes depending on network speeds etc.
When complete, a message will be printed with where the Console is running, and the local `_cfg` directory will contain configuration files to help connect applications.