Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/cloudfoundry/capi-workspace
Machine setup for CAPI developers
https://github.com/cloudfoundry/capi-workspace
brewfile capi dotfiles setup team workspace workstation
Last synced: 3 months ago
JSON representation
Machine setup for CAPI developers
- Host: GitHub
- URL: https://github.com/cloudfoundry/capi-workspace
- Owner: cloudfoundry
- License: apache-2.0
- Created: 2017-12-20T16:45:03.000Z (about 7 years ago)
- Default Branch: main
- Last Pushed: 2024-06-25T15:26:19.000Z (7 months ago)
- Last Synced: 2024-06-26T04:29:51.695Z (7 months ago)
- Topics: brewfile, capi, dotfiles, setup, team, workspace, workstation
- Language: Shell
- Homepage:
- Size: 537 KB
- Stars: 7
- Watchers: 36
- Forks: 9
- Open Issues: 11
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
## CAPI Workstation Helpers 🐋
Hello there! This repo is intended to help with development work on CAPI
projects.
## Directory Overview
Description of the folders in this project:
Folder | Description
---------------- | -----------
assets | random static files
bash-it | all the [bash-it](https://github.com/Bash-it/bash-it) stuff, only bash-it stuff
helpers | random helper files
lib | scripts that could be sourced into your shell
bin | scripts that can be added to your `$PATH`
install-scripts | executable scripts that install a thing or two
Why is `lib` separate from `bash-it`? Not everyone uses bash-it, so `lib` is
similar to `scripts`, giving people a way to manually load the things they care
about in.
## Minimal Install
One may manually (and minimally) load in capi-workspace content by including
the following in their `~/.zshrc` (or equivalent):
```bash
source <(cat $HOME/workspace/capi-workspace/lib/*) > /dev/null
path+=("$HOME/workspace/capi-workspace/bin")
```
## CAPI Helper Commands
This table is generated from `assets/capidoc.yml`. It is dynamically available
in your terminal using the `capi` command.
```
backup-restore
compare_db_rows Compares counts of two db's rows to validate backup & restore
bosh-deploy
create_and_deploy Create a capi release, upload to current bosh target, and deploy with default ops files (interactive)
create_and_upload Create a capi release and upload to current bosh target
deploy Create a capi release, upload to current bosh target, and deploy with default ops files (non interactive)
deploy_only_new_capi Deploy just the current CAPI to current bosh target
quick_deploy Quickly upload changes in cloud_controller_ng to relevant vms and restart"
bosh-lite-pool
attach_debugger Attaches the rubymine debugger to your bosh lite
bootstrap_cf Target a CF, creates an org, space, and targets them
cf_admin_password Get the CF admin password for a current bosh target (requires credhub)
cfu Alias for seed_users
claim_bosh_lite Claim an available bosh lite CF
mysql_bosh_lite Connect to current bosh target's MySQL DB
print_env_info List all the claimed bosh-lites in the pool
psql_bosh_lite Connect to current bosh target's MySQL DB
rebootstrap_cf Delete the 'org' org then call bootstrap_cf. For cleaning the slate.
seed_users Add users of every role to the current org and space
target_bosh Target a bosh environment from the pool
target_cf Change cf cli target to the cloud foundry deployed on currently targeted bosh (hint: use "target_bosh" first).
target_uaa Change uaac target to currently targeted bosh and login as admin
unclaim_bosh_lite Return a bosh lite CF back to the pool
which_bosh Print which bosh environment is currently targeted
capi-release-author
sync_package_specs Add go submodule dependencies to bosh package specs
sync_submodule_config Sync git submodules with go dependencies using gosub
unused_blobs Detects unused blobs from the blobs.yml file
file-system
v Open matching location in vim (fasd -e vim)
z Change current directory to directory matching provided pattern (ex: z ng)
git
fixcommitter Cleans up committer/author after a rebase
g git status
gd git diff
gdc git diff --cached
git-open Open a git repo in your browser, courtesy of Paul Irish.
pullify Pull down all PRs for current git repo as branches
staged_shortlog Get commit shortlogs and authors from submodule changes
update Update all the git submodules
which_capi Given a cloud_controller_ng commit, print which version of capi-release it first appears in
ruby-dev
b bundle exec
bake bundle exec rake
gi gem install
mybake DB=mysql bundle exec rake
pgbake DB=postgres bundle exec rake
testing
baras Runs baras with local integration_config
cats Runs cats with local integration_config
cats_cleanup Deletes builpacks, orgs, quotes, and service brokers in a targeted CF
check_certificate_expirations Reads YAML files, checks for expired (or premature) certs. Useful for pipeline troubleshooting
delete_orgs Delete all orgs for current cf target
generate_integration_config Create an integration_config.json for running CATS against current bosh target.
run_bridge_tests Run Bridge unit tests
sits Runs sits against a bosh lite
workspaces
capi-workspaces List and connect to existing K8s CAPI workspaces
create-capi-workspace Create a CAPI workspace in K8s cluster
```
## Using the Bosh Lite Pool
As a perk, [CAPI
approvers](https://github.com/cloudfoundry/community/blob/main/toc/working-groups/app-runtime-interfaces.md)
get access to a pool of cf-deployment environments, deployed on bosh lites in a
CFF-managed IaaS account. These environments are disposable, and are only
intended for development use.
Though it is technically possible to use the pool without them, there are a
number of convenience scripts provided by capi-workspace that make the pool
usable by humans.
### Prerequisites
- git
- Clone capi-workspace (this repo)
- Add capi-workspace scripts/functions to your shell (see installation instructions above (or below))
- Clone https://github.com/cloudfoundry/capi-env-pool
### Example Pool Dev Workflow
Claim an environment from the pool:
```
❯ claim_bosh_lite
Claiming 'distaffs'...
Writing out .envrc...
Pushing reservation to capi-env-pool...
Use 'target_bosh distaffs' to set bosh environment variables.
```
Observe that you successfully claimed the environment:
```
❯ print_env_info
Rounding up claimed environments...
* ENV * * CLAIMED BY * * CLAIMED ON * * CLAIMED SINCE *
apologia Tim Downey tdowney.hostname 3 weeks ago
distaffs Greg Cobb gcobb.hostname 2 minutes ago
riskier Greg Cobb gcobb.hostname 13 days ago
zoned Seth Boyles capi-ws-cc-203 3 weeks ago
```
Set bosh environment variables in your shell to "target" that environment:
```
❯ target_bosh distaffs
Refreshing bosh lite pool state...
Already up to date.
Sourcing /.../capi-env-pool/bosh-lites/claimed/distaffs to set bosh environment variables...
Writing a capi-specific integration_config.json...
Generated /.../capi-env-pool/distaffs/integration_config.json. Enjoy!
Writing a bosh gateway ssh key...
Generated /.../capi-env-pool/distaffs/bosh.pem. Enjoy!
Setting BOSH_GW_PRIVATE_KEY, BOSH_ALL_PROXY, and CONFIG environment variables...
Success!
```
Log in to the environment with the `cf` CLI:
```
❯ target_cf
Getting CF admin password from credhub...
Setting API endpoint to https://api.distaffs.app-runtime-interfaces.ci.cloudfoundry.org...
OK
API endpoint: https://api.distaffs.app-runtime-interfaces.ci.cloudfoundry.org
API version: 3.166.0
Not logged in. Use 'cf login' or 'cf login --sso' to log in.
API endpoint: https://api.distaffs.app-runtime-interfaces.ci.cloudfoundry.org
Authenticating...
OK
Use 'cf target' to view or set your target org and space.
```
Create a default Organization and Space:
```
❯ bootstrap_cf
Getting CF admin password from credhub...
Setting API endpoint to https://api.distaffs.app-runtime-interfaces.ci.cloudfoundry.org...
OK
API endpoint: https://api.distaffs.app-runtime-interfaces.ci.cloudfoundry.org
API version: 3.166.0
Not logged in. Use 'cf login' or 'cf login --sso' to log in.
API endpoint: https://api.distaffs.app-runtime-interfaces.ci.cloudfoundry.org
Authenticating...
OK
Use 'cf target' to view or set your target org and space.
Creating org org as admin...
OK
TIP: Use 'cf target -o "org"' to target new org
API endpoint: https://api.distaffs.app-runtime-interfaces.ci.cloudfoundry.org
API version: 3.166.0
user: admin
org: org
No space targeted, use 'cf target -s SPACE'
Creating space space in org org as admin...
OK
Assigning role SpaceManager to user admin in org org / space space as admin...
OK
Assigning role SpaceDeveloper to user admin in org org / space space as admin...
OK
TIP: Use 'cf target -o "org" -s "space"' to target new space
API endpoint: https://api.distaffs.app-runtime-interfaces.ci.cloudfoundry.org
API version: 3.166.0
user: admin
org: org
space: space
```
Deploy the cf-deployment and capi-release versions from your dev machine:
```
❯ deploy
Uploading new release to 35.185.252.196.
Syncing bosh blobs...
bosh sync-blobs
...
Creating bosh release...
create-release --force --name capi
...
Uploading release to bosh director...
bosh upload-release --rebase
...
Deploying uploaded release...
bosh deploy cf-deployment.yml -o...
...
Succeeded
```
Make local changes and deploy a new CAPI version:
```
❯ deploy_only_new_capi
Building a new CAPI from local filesystem and deploying to 35.185.252.196.
bosh deploy <(bosh manifest) -o ...use-created-capi.yml ...
...
Succeeded
```
Release the environment for deletion, once you are done with it:
```
❯ unclaim_bosh_lite riskier
Refreshing bosh lite pool state...
Hit enter to release 'riskier'
rm 'riskier/.envrc'
Pushing the release commit to capi-env-pool...
Done!
```
## Automated Workstation Setup
If you want, there are also some opinionated scripts you can run to set up a
new computer for CAPI development. These may be useful if you are dynamically
provisioning remote workstations, or something like that.
### Dependencies
* MacOS
* pip
### Running the Installer
```
mkdir -p ~/workspace && cd ~/workspace
git clone [email protected]:cloudfoundry/capi-workspace.git && cd capi-workspace
```
You can now either install via `./install.sh` or `./install-core.sh`.
Or curl the bootstrap script, which will create the `workspace` directory and clone this repo for you.
This is primarily intended for quickly setting up virtual workstations.
```
bash <(curl -s https://raw.githubusercontent.com/cloudfoundry/capi-workspace/main/bootstrap.sh)
```
## Manual steps (to be automated later)
* Open System Preferences / Users & Groups / / Login Items
* Add `flycut` from Applications
* Add `spectacle` from Applications
* Log out/Log in
* This will cause Flycut and Spectacle to run and ask for permissions
* Open up Rubymine manually and enter a `License Server`
* Install the `mine` cli shortcut (RubyMine -> Tools -> Create Command-line Launcher...)
* If you are using Goland do the same two previous steps for Goland
* If you have access to the private CAPI backlog, add a secure note called `tracker_api_token` to LastPass with your Pivotal Tracker API token. This token will be used in scripts such as `claim_bosh_lite` to pull story titles currently in flight.
## Contributing to this repo
* kindly consider when a change is worth making to `install-core.sh` versus `install.sh`.
* [bash-it](https://github.com/Bash-it/bash-it) We use bash-it to organize and streamline our bash settings. This includes stuff like color schemes, aliases, shell settings, and the shell prompt formatting. Adding "plugins" to `custom-bash-it-plugins` will cause them to be installed in every new shell.
## What this installation does
`install-core.sh`: intended to be more friendly for developers who don't want this repository to take over their machine. This is intended to be the minimal set of things required for working with a CAPI repository.
* Only installs core utilities, languages, services, and binaries.
* Is NOT responsible for installing & configuring any editors (nvim/vscode/intellij)
* Is NOT responsible for installing tools for improving developer workflows (jq/rg)
* Is NOT responsible for applying any git configuration
* Is NOT responsible for loading in scripts that need to be `source`'d.
* Is NOT responsible for modifying your `$PATH` to include `script`
`./install.sh`: runs `install-core.sh` and configures the machine with many more packages/preferences
## install.sh or install-core.sh ?
`install-core.sh` is a subset of `install.sh` and is intended to be less invasive to the machine, without compromising on provisioning a workstation suitable to working on CAPI projects.
For fresh workstations && full-time CAPI developers, install.sh might make more sense.
For folks working on multiple projects, `install-core.sh` might make more sense.
For folks with existing workstations & configuration, `install-core.sh` might make more sense.