Compare commits

..

63 Commits

Author SHA1 Message Date
7ec89cdac5 fix: cargo fmt
All checks were successful
Run Check Script / check (pull_request) Successful in 1m58s
2025-06-26 11:26:07 -04:00
17ad92402d feat: added webhook receiver to alertchannels
Some checks failed
Run Check Script / check (pull_request) Failing after 47s
2025-06-26 10:12:18 -04:00
29e74a2712 Merge pull request 'feat: added alert rule and impl for prometheus as well as a few preconfigured bmc alerts for dell server that are used in the monitoring example' (#67) from feat/alert_rules into master
All checks were successful
Run Check Script / check (push) Successful in -1s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 13m10s
Reviewed-on: #67
2025-06-26 13:16:38 +00:00
e16f8fa82e fix: modified directory names to be in line with alert functions and deployment environments
All checks were successful
Run Check Script / check (pull_request) Successful in 1m43s
2025-06-25 16:10:45 -04:00
c21f3084dc feat: added alert rule and impl for prometheus as well as a few preconfigured bmc alerts for dell server that are used in the monitoring example
All checks were successful
Run Check Script / check (pull_request) Successful in 1m35s
2025-06-25 15:10:16 -04:00
2c706225a1 feat: Publishing a release of harmony composer binary as latest-snapshot (#65)
All checks were successful
Run Check Script / check (push) Successful in 1m42s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 3m44s
Co-authored-by: tahahawa <tahahawa@gmail.com>
Reviewed-on: #65
Reviewed-by: taha <taha@noreply.git.nationtech.io>
Co-authored-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
Co-committed-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
2025-06-25 15:14:45 +00:00
f437c40428 impl_monitoring_alerting_kube_prometheus (#64)
All checks were successful
Run Check Script / check (push) Successful in 1m29s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 2m59s
Co-authored-by: tahahawa <tahahawa@gmail.com>
Reviewed-on: #64
Co-authored-by: Willem <wrolleman@nationtech.io>
Co-committed-by: Willem <wrolleman@nationtech.io>
2025-06-24 18:54:15 +00:00
e06548ac44 feat: Alerting module architecture to make it easy to use and extensible by external crates
All checks were successful
Run Check Script / check (push) Successful in 1m34s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 3m26s
Co-authored-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
Reviewed-on: #61
Reviewed-by: johnride <jg@nationtech.io>
Co-authored-by: Willem <wrolleman@nationtech.io>
Co-committed-by: Willem <wrolleman@nationtech.io>
2025-06-19 14:37:16 +00:00
155e9bac28 feat: create harmony_composer initial version + rework CI (#58)
All checks were successful
Run Check Script / check (push) Successful in 1m38s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 4m15s
Co-authored-by: tahahawa <tahahawa@gmail.com>
Reviewed-on: #58
Reviewed-by: johnride <jg@nationtech.io>
Co-authored-by: Taha Hawa <taha@taha.dev>
Co-committed-by: Taha Hawa <taha@taha.dev>
2025-06-18 19:52:37 +00:00
7bebc58615 feat: add tenant credential management (#63)
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Adds the foundation for managing tenant credentials, including:

- `TenantCredentialScore` for scoring credential-related operations.
- `TenantCredentialManager` trait for creating users.
- `CredentialMetadata` struct to store credential information.
- `CredentialData` enum to hold credential content.
- `TenantCredentialBundle` struct to encapsulate metadata and content.

This provides a starting point for implementing credential creation, storage, and retrieval within the harmony system.

Reviewed-on: #63
Co-authored-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
Co-committed-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
2025-06-17 18:28:04 +00:00
246d6718c3 docs: Introduce project delivery automation ADR. This is still WIP (#51)
All checks were successful
Run Check Script / check (push) Successful in 1m52s
Reviewed-on: #51
Co-authored-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
Co-committed-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
2025-06-12 20:00:22 +00:00
d776042e20 docs: Improve README formatting
All checks were successful
Run Check Script / check (push) Successful in 1m49s
Signed-off-by: johnride <jg@nationtech.io>
2025-06-12 18:23:17 +00:00
86c681be70 docs: New README, two options to choose from right now (#59)
All checks were successful
Run Check Script / check (push) Successful in 1m52s
Reviewed-on: #59
Co-authored-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
Co-committed-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
2025-06-12 18:16:43 +00:00
b94dd1e595 feat: add support for custom CIDR ingress/egress rules (#60)
All checks were successful
Run Check Script / check (push) Successful in 1m53s
- Added `additional_allowed_cidr_ingress` and `additional_allowed_cidr_egress` fields to `TenantNetworkPolicy` to allow specifying custom CIDR blocks for network access.
- Updated K8sTenantManager to parse and apply these CIDR rules to NetworkPolicy ingress and egress rules.
- Added `cidr` dependency to `harmony_macros` and a custom proc macro `cidrv4` to easily parse CIDR strings.
- Updated TenantConfig to default inter tenant and internet egress to deny all and added default empty vectors for CIDR ingress and egress.
- Updated ResourceLimits to implement default.

Reviewed-on: #60
Co-authored-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
Co-committed-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
2025-06-12 15:24:03 +00:00
ef5ec4a131 Merge pull request 'feat: Pass configuration when initializing K8sAnywhereTopology' (#57) from feat/configK8sAnywhere into master
All checks were successful
Run Check Script / check (push) Successful in 1m47s
Reviewed-on: #57
2025-06-10 13:01:50 +00:00
a8eb06f686 feat: Pass configuration when initializing K8sAnywhereTopology
All checks were successful
Run Check Script / check (push) Successful in 1m47s
Run Check Script / check (pull_request) Successful in 1m47s
2025-06-10 09:00:38 -04:00
d1678b529e Merge pull request 'feat: K8s Tenant looks good, basic isolation working now' (#56) from feat/k8sTenant into master
All checks were successful
Run Check Script / check (push) Successful in 1m57s
Reviewed-on: #56
2025-06-10 12:59:13 +00:00
1451260d4d feat: K8s Tenant looks good, basic isolation working now
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Run Check Script / check (pull_request) Successful in 1m48s
2025-06-09 20:39:15 -04:00
415488ba39 feat: K8s apply function now correctly emulates kubectl apply behavior by either creating or updating resources (#55)
Some checks failed
Run Check Script / check (push) Has been cancelled
Reviewed-on: #55
Co-authored-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
Co-committed-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
2025-06-09 20:19:54 +00:00
bf7a6d590c Merge pull request 'TenantManager_impl_k8s_anywhere' (#47) from TenantManager_impl_k8s_anywhere into master
All checks were successful
Run Check Script / check (push) Successful in 1m56s
Reviewed-on: #47
2025-06-09 18:07:32 +00:00
8d8120bbfd fix: K8s ingress module was completely broken, fixed resource definition structure and types
All checks were successful
Run Check Script / check (push) Successful in 1m47s
Run Check Script / check (pull_request) Successful in 1m48s
2025-06-09 14:02:06 -04:00
6cf61ae67c feat: Tenant manager k8s implementation progress : ResourceQuota, NetworkPolicy and Namespace look good. Still WIP 2025-06-09 13:59:49 -04:00
8c65aef127 feat: Can now apply any k8s resource type, both namespaced or cluster scoped 2025-06-09 13:58:40 -04:00
00e71b97f6 chore: Move ADR helper files into folders with their corresponding ADR number
All checks were successful
Run Check Script / check (push) Successful in 1m49s
Run Check Script / check (pull_request) Successful in 1m48s
2025-06-09 13:54:23 -04:00
ee2bba5623 Merge pull request 'feat: Add Default implementation for Harmony Id along with documentation.' (#53) from feat/id_default into TenantManager_impl_k8s_anywhere
All checks were successful
Run Check Script / check (push) Successful in 1m47s
Run Check Script / check (pull_request) Successful in 1m47s
Reviewed-on: #53
2025-06-09 17:47:33 +00:00
118d34db55 Merge pull request 'feat: Initialize k8s tenant properly' (#54) from feat/init_k8s_tenant into feat/id_default
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Run Check Script / check (pull_request) Successful in 1m47s
Reviewed-on: #54
2025-06-09 17:42:10 +00:00
24e466fadd fix: formatting
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Run Check Script / check (pull_request) Successful in 1m46s
2025-06-08 23:51:11 -04:00
14fc4345c1 feat: Initialize k8s tenant properly
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Run Check Script / check (pull_request) Successful in 1m49s
2025-06-08 23:49:08 -04:00
8e472e4c65 feat: Add Default implementation for Harmony Id along with documentation.
Some checks failed
Run Check Script / check (push) Failing after 47s
Run Check Script / check (pull_request) Failing after 45s
This Id implementation is optimized for ease of use. Ids are prefixed with the unix epoch and suffixed with 7 alphanumeric characters. But Ids can also contain any String the user wants to pass it
2025-06-08 21:23:29 -04:00
ec17ccc246 feat: Add example-tenant (WIP)
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Run Check Script / check (pull_request) Successful in 1m53s
2025-06-06 13:59:48 -04:00
5127f44ab3 docs: Add note about pod privilege escalation in ADR 011 Tenant
All checks were successful
Run Check Script / check (push) Successful in 1m47s
Run Check Script / check (pull_request) Successful in 1m46s
2025-06-06 13:56:40 -04:00
2ff70db0b1 wip: Tenant example project
All checks were successful
Run Check Script / check (push) Successful in 1m49s
Run Check Script / check (pull_request) Successful in 1m48s
2025-06-06 13:52:40 -04:00
e17ac1af83 Merge remote-tracking branch 'origin/master' into TenantManager_impl_k8s_anywhere
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Run Check Script / check (pull_request) Successful in 1m47s
2025-06-04 16:14:21 -04:00
31e59937dc Merge pull request 'feat: Initial setup for monitoring and alerting' (#48) from feat/monitor into master
All checks were successful
Run Check Script / check (push) Successful in 1m50s
Reviewed-on: #48
Reviewed-by: johnride <jg@nationtech.io>
2025-06-03 18:17:13 +00:00
12eb4ae31f fix: cargo fmt
All checks were successful
Run Check Script / check (push) Successful in 1m47s
Run Check Script / check (pull_request) Successful in 1m47s
2025-06-02 16:20:49 -04:00
a2be9457b9 wip: removed AlertReceiverConfig
Some checks failed
Run Check Script / check (push) Failing after 44s
Run Check Script / check (pull_request) Failing after 44s
2025-06-02 16:11:36 -04:00
0d56fbc09d wip: applied comments in pr, changed naming of AlertChannel to AlertReceiver and added rust doc to Monitor for clarity
All checks were successful
Run Check Script / check (push) Successful in 1m49s
Run Check Script / check (pull_request) Successful in 1m47s
2025-06-02 14:44:43 -04:00
56dc1e93c1 fix: modified files in mod
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Run Check Script / check (pull_request) Successful in 1m46s
2025-06-02 11:47:21 -04:00
691540fe64 wip: modified initial monitoring architecture based on pr review
Some checks failed
Run Check Script / check (push) Failing after 46s
Run Check Script / check (pull_request) Failing after 43s
2025-06-02 11:42:37 -04:00
7e3f1b1830 fix:cargo fmt
All checks were successful
Run Check Script / check (push) Successful in 1m45s
Run Check Script / check (pull_request) Successful in 1m45s
2025-05-30 13:59:29 -04:00
b631e8ccbb feat: Initial setup for monitoring and alerting
Some checks failed
Run Check Script / check (push) Failing after 43s
Run Check Script / check (pull_request) Failing after 45s
2025-05-30 13:21:38 -04:00
60f2f31d6c feat: Add TenantScore and TenantInterpret (#45)
All checks were successful
Run Check Script / check (push) Successful in 1m47s
Reviewed-on: #45
Co-authored-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
Co-committed-by: Jean-Gabriel Gill-Couture <jg@nationtech.io>
2025-05-30 13:13:43 +00:00
045954f8d3 start network policy
All checks were successful
Run Check Script / check (push) Successful in 1m50s
Run Check Script / check (pull_request) Successful in 1m46s
2025-05-29 18:06:16 -04:00
27f1a9dbdd feat: add more to the tenantmanager k8s impl (#46)
All checks were successful
Run Check Script / check (push) Successful in 1m55s
Co-authored-by: Willem <wrolleman@nationtech.io>
Reviewed-on: #46
Co-authored-by: Taha Hawa <taha@taha.dev>
Co-committed-by: Taha Hawa <taha@taha.dev>
2025-05-29 20:15:38 +00:00
7c809bf18a Make k8stenantmanager a oncecell
All checks were successful
Run Check Script / check (push) Successful in 1m49s
Run Check Script / check (pull_request) Successful in 1m46s
2025-05-29 16:03:58 -04:00
6490e5e82a Hardcode some limits to protect the overall cluster
Some checks failed
Run Check Script / check (push) Failing after 42s
Run Check Script / check (pull_request) Failing after 47s
2025-05-29 15:49:46 -04:00
5e51f7490c Update request quota
Some checks failed
Run Check Script / check (push) Failing after 46s
2025-05-29 15:41:57 -04:00
97fba07f4e feat: adding kubernetes implentation of tenant manager
Some checks failed
Run Check Script / check (push) Failing after 43s
2025-05-29 14:35:58 -04:00
624e4330bb boilerplate
All checks were successful
Run Check Script / check (push) Successful in 1m47s
2025-05-29 13:36:30 -04:00
e7917843bc Merge pull request 'feat: Add initial Tenant traits and data structures' (#43) from feat/tenant into master
Some checks failed
Run Check Script / check (push) Has been cancelled
Reviewed-on: #43
2025-05-29 15:51:33 +00:00
7cd541bdd8 chore: Fix pr comments, remove many YAGNI things
All checks were successful
Run Check Script / check (push) Successful in 1m46s
Run Check Script / check (pull_request) Successful in 1m46s
2025-05-29 11:47:25 -04:00
270dd49567 Merge pull request 'docs: Add CONTRIBUTING.md guide' (#44) from doc/contributor into master
All checks were successful
Run Check Script / check (push) Successful in 1m46s
Reviewed-on: #44
2025-05-29 14:48:18 +00:00
0187300473 docs: Add CONTRIBUTING.md guide
All checks were successful
Run Check Script / check (push) Successful in 1m46s
Run Check Script / check (pull_request) Successful in 1m47s
2025-05-29 10:47:38 -04:00
bf16566b4e wip: Clean up some unnecessary bits in the Tenant module and move manager to its own file
All checks were successful
Run Check Script / check (push) Successful in 1m48s
Run Check Script / check (pull_request) Successful in 1m46s
2025-05-29 07:25:45 -04:00
895fb02f4e feat: Add initial Tenant traits and data structures
All checks were successful
Run Check Script / check (push) Successful in 1m46s
Run Check Script / check (pull_request) Successful in 1m45s
2025-05-28 22:33:46 -04:00
88d6af9815 Merge pull request 'feat/basicCI' (#42) from feat/basicCI into master
All checks were successful
Run Check Script / check (push) Successful in 1m50s
Reviewed-on: #42
Reviewed-by: taha <taha@noreply.git.nationtech.io>
2025-05-28 19:42:19 +00:00
5aa9dc701f fix: Removed forgotten refactoring bits and formatting
All checks were successful
Run Check Script / check (push) Successful in 1m46s
Run Check Script / check (pull_request) Successful in 1m48s
2025-05-28 15:19:39 -04:00
f4ef895d2e feat: Add basic CI configuration
Some checks failed
Run Check Script / check (push) Failing after 51s
2025-05-28 14:40:19 -04:00
6e7148a945 Merge pull request 'adr: Add ADR on multi tenancy using namespace based customer isolation' (#41) from adr/multi-tenancy into master
Reviewed-on: #41
2025-05-26 20:26:36 +00:00
83453273c6 adr: Add ADR on multi tenancy using namespace based customer isolation 2025-05-26 11:56:45 -04:00
76ae5eb747 fix: make HelmRepository public (#39)
Co-authored-by: tahahawa <tahahawa@gmail.com>
Reviewed-on: #39
Reviewed-by: johnride <jg@nationtech.io>
2025-05-22 20:07:42 +00:00
9c51040f3b Merge pull request 'feat:added Slack notifications support' (#38) from feat/slack-notifs into master
Reviewed-on: #38
Reviewed-by: johnride <jg@nationtech.io>
2025-05-22 20:04:51 +00:00
19bd47a545 Merge pull request 'monitoringalerting' (#37) from monitoringalerting into master
Reviewed-on: #37
Reviewed-by: johnride <jg@nationtech.io>
2025-05-21 17:32:26 +00:00
92 changed files with 3674 additions and 5994 deletions

2
.dockerignore Normal file
View File

@@ -0,0 +1,2 @@
target/
Dockerfile

View File

@@ -0,0 +1,18 @@
name: Run Check Script
on:
push:
branches:
- master
pull_request:
jobs:
check:
runs-on: docker
container:
image: hub.nationtech.io/harmony/harmony_composer:latest@sha256:eb0406fcb95c63df9b7c4b19bc50ad7914dd8232ce98e9c9abef628e07c69386
steps:
- name: Checkout code
uses: actions/checkout@v4
- name: Run check script
run: bash check.sh

View File

@@ -0,0 +1,95 @@
name: Compile and package harmony_composer
on:
push:
branches:
- master
jobs:
package_harmony_composer:
container:
image: hub.nationtech.io/harmony/harmony_composer:latest@sha256:eb0406fcb95c63df9b7c4b19bc50ad7914dd8232ce98e9c9abef628e07c69386
runs-on: dind
steps:
- name: Checkout code
uses: actions/checkout@v4
- name: Build for Linux x86_64
run: cargo build --release --bin harmony_composer --target x86_64-unknown-linux-gnu
- name: Build for Windows x86_64 GNU
run: cargo build --release --bin harmony_composer --target x86_64-pc-windows-gnu
- name: Setup log into hub.nationtech.io
uses: docker/login-action@v3
with:
registry: hub.nationtech.io
username: ${{ secrets.HUB_BOT_USER }}
password: ${{ secrets.HUB_BOT_PASSWORD }}
# TODO: build ARM images and MacOS binaries (or other targets) too
- name: Update snapshot-latest tag
run: |
git config user.name "Gitea CI"
git config user.email "ci@nationtech.io"
git tag -f snapshot-latest
git push origin snapshot-latest --force
- name: Install jq
run: apt install -y jq # The current image includes apt lists so we don't have to apt update and rm /var/lib/apt... every time. But if the image is optimized it won't work anymore
- name: Create or update release
run: |
# First, check if release exists and delete it if it does
RELEASE_ID=$(curl -s -X GET \
-H "Authorization: token ${{ secrets.GITEATOKEN }}" \
"https://git.nationtech.io/api/v1/repos/nationtech/harmony/releases/tags/snapshot-latest" \
| jq -r '.id // empty')
if [ -n "$RELEASE_ID" ]; then
# Delete existing release
curl -X DELETE \
-H "Authorization: token ${{ secrets.GITEATOKEN }}" \
"https://git.nationtech.io/api/v1/repos/nationtech/harmony/releases/$RELEASE_ID"
fi
# Create new release
RESPONSE=$(curl -X POST \
-H "Authorization: token ${{ secrets.GITEATOKEN }}" \
-H "Content-Type: application/json" \
-d '{
"tag_name": "snapshot-latest",
"name": "Latest Snapshot",
"body": "Automated snapshot build from master branch",
"draft": false,
"prerelease": true
}' \
"https://git.nationtech.io/api/v1/repos/nationtech/harmony/releases")
echo "RELEASE_ID=$(echo $RESPONSE | jq -r '.id')" >> $GITHUB_ENV
- name: Upload Linux binary
run: |
curl -X POST \
-H "Authorization: token ${{ secrets.GITEATOKEN }}" \
-H "Content-Type: application/octet-stream" \
--data-binary "@target/x86_64-unknown-linux-gnu/release/harmony_composer" \
"https://git.nationtech.io/api/v1/repos/nationtech/harmony/releases/${{ env.RELEASE_ID }}/assets?name=harmony_composer"
- name: Upload Windows binary
run: |
curl -X POST \
-H "Authorization: token ${{ secrets.GITEATOKEN }}" \
-H "Content-Type: application/octet-stream" \
--data-binary "@target/x86_64-pc-windows-gnu/release/harmony_composer.exe" \
"https://git.nationtech.io/api/v1/repos/nationtech/harmony/releases/${{ env.RELEASE_ID }}/assets?name=harmony_composer.exe"
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
- name: Build and push
uses: docker/build-push-action@v6
with:
context: .
push: true
tags: hub.nationtech.io/harmony/harmony_composer:latest

36
CONTRIBUTING.md Normal file
View File

@@ -0,0 +1,36 @@
# Contributing to the Harmony project
## Write small P-R
Aim for the smallest piece of work that is mergeable.
Mergeable means that :
- it does not break the build
- it moves the codebase one step forward
P-Rs can be many things, they do not have to be complete features.
### What a P-R **should** be
- Introduce a new trait : This will be the place to discuss the new trait addition, its design and implementation
- A new implementation of a trait : a new concrete implementation of the LoadBalancer trait
- A new CI check : something that improves quality, robustness, ci performance
- Documentation improvements
- Refactoring
- Bugfix
### What a P-R **should not** be
- Large. Anything over 200 lines (excluding generated lines) should have a very good reason to be this large.
- A mix of refactoring, bug fixes and new features.
- Introducing multiple new features or ideas at once.
- Multiple new implementations of a trait/functionnality at once
The general idea is to keep P-Rs small and single purpose.
## Commit message formatting
We follow conventional commits guidelines.
https://www.conventionalcommits.org/en/v1.0.0/

336
Cargo.lock generated
View File

@@ -295,6 +295,51 @@ dependencies = [
"cipher",
]
[[package]]
name = "bollard"
version = "0.19.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "af706e9dc793491dd382c99c22fde6e9934433d4cc0d6a4b34eb2cdc57a5c917"
dependencies = [
"base64 0.22.1",
"bollard-stubs",
"bytes",
"futures-core",
"futures-util",
"hex",
"http 1.3.1",
"http-body-util",
"hyper 1.6.0",
"hyper-named-pipe",
"hyper-util",
"hyperlocal",
"log",
"pin-project-lite",
"serde",
"serde_derive",
"serde_json",
"serde_repr",
"serde_urlencoded",
"thiserror 2.0.12",
"tokio",
"tokio-util",
"tower-service",
"url",
"winapi",
]
[[package]]
name = "bollard-stubs"
version = "1.48.2-rc.28.0.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "79cdf0fccd5341b38ae0be74b74410bdd5eceeea8876dc149a13edfe57e3b259"
dependencies = [
"serde",
"serde_json",
"serde_repr",
"serde_with",
]
[[package]]
name = "bstr"
version = "1.12.0"
@@ -324,6 +369,55 @@ version = "1.10.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d71b6127be86fdcfddb610f7182ac57211d4b18a3e9c82eb2d17662f2227ad6a"
[[package]]
name = "camino"
version = "1.1.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0da45bc31171d8d6960122e222a67740df867c1dd53b4d51caa297084c185cab"
dependencies = [
"serde",
]
[[package]]
name = "cargo-platform"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "84982c6c0ae343635a3a4ee6dedef965513735c8b183caa7289fa6e27399ebd4"
dependencies = [
"serde",
]
[[package]]
name = "cargo-util-schemas"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e63d2780ac94487eb9f1fea7b0d56300abc9eb488800854ca217f102f5caccca"
dependencies = [
"semver",
"serde",
"serde-untagged",
"serde-value",
"thiserror 1.0.69",
"toml",
"unicode-xid",
"url",
]
[[package]]
name = "cargo_metadata"
version = "0.20.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4f7835cfc6135093070e95eb2b53e5d9b5c403dc3a6be6040ee026270aa82502"
dependencies = [
"camino",
"cargo-platform",
"cargo-util-schemas",
"semver",
"serde",
"serde_json",
"thiserror 2.0.12",
]
[[package]]
name = "cassowary"
version = "0.3.0"
@@ -394,6 +488,9 @@ name = "cidr"
version = "0.2.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6bdf600c45bd958cf2945c445264471cca8b6c8e67bc87b71affd6d7e5682621"
dependencies = [
"serde",
]
[[package]]
name = "cipher"
@@ -666,6 +763,12 @@ dependencies = [
"cipher",
]
[[package]]
name = "current_platform"
version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a74858bcfe44b22016cb49337d7b6f04618c58e5dbfdef61b06b8c434324a0bc"
[[package]]
name = "curve25519-dalek"
version = "4.1.3"
@@ -752,6 +855,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "9c9e6a11ca8224451684bc0d7d5a7adbf8f2fd6887261a1cfc3c0432f9d4068e"
dependencies = [
"powerfmt",
"serde",
]
[[package]]
@@ -977,6 +1081,16 @@ version = "1.0.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "877a4ace8713b0bcf2a4e7eec82529c029f1d0619886d18145fea96c3ffe5c0f"
[[package]]
name = "erased-serde"
version = "0.4.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e004d887f51fcb9fef17317a2f3525c887d8aa3f4f50fed920816a688284a5b7"
dependencies = [
"serde",
"typeid",
]
[[package]]
name = "errno"
version = "0.3.11"
@@ -1040,6 +1154,16 @@ dependencies = [
"url",
]
[[package]]
name = "example-monitoring"
version = "0.1.0"
dependencies = [
"harmony",
"harmony_cli",
"tokio",
"url",
]
[[package]]
name = "example-nanodc"
version = "0.1.0"
@@ -1070,6 +1194,21 @@ dependencies = [
"url",
]
[[package]]
name = "example-tenant"
version = "0.1.0"
dependencies = [
"cidr",
"env_logger",
"harmony",
"harmony_cli",
"harmony_macros",
"harmony_types",
"log",
"tokio",
"url",
]
[[package]]
name = "example-tui"
version = "0.1.0"
@@ -1373,7 +1512,7 @@ dependencies = [
"futures-sink",
"futures-util",
"http 0.2.12",
"indexmap",
"indexmap 2.9.0",
"slab",
"tokio",
"tokio-util",
@@ -1392,7 +1531,7 @@ dependencies = [
"futures-core",
"futures-sink",
"http 1.3.1",
"indexmap",
"indexmap 2.9.0",
"slab",
"tokio",
"tokio-util",
@@ -1404,17 +1543,20 @@ name = "harmony"
version = "0.1.0"
dependencies = [
"async-trait",
"chrono",
"cidr",
"convert_case",
"derive-new",
"directories",
"dockerfile_builder",
"dyn-clone",
"email_address",
"env_logger",
"fqdn",
"harmony_macros",
"harmony_types",
"helm-wrapper-rs",
"hex",
"http 1.3.1",
"inquire",
"k3d-rs",
@@ -1426,6 +1568,7 @@ dependencies = [
"non-blank-string-rs",
"opnsense-config",
"opnsense-config-xml",
"rand 0.9.1",
"reqwest 0.11.27",
"russh",
"rust-ipmi",
@@ -1454,10 +1597,26 @@ dependencies = [
"tokio",
]
[[package]]
name = "harmony_composer"
version = "0.1.0"
dependencies = [
"bollard",
"cargo_metadata",
"clap",
"current_platform",
"env_logger",
"futures-util",
"log",
"serde_json",
"tokio",
]
[[package]]
name = "harmony_macros"
version = "0.1.0"
dependencies = [
"cidr",
"harmony_types",
"quote",
"serde",
@@ -1490,6 +1649,12 @@ dependencies = [
"serde",
]
[[package]]
name = "hashbrown"
version = "0.12.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8a9ee70c43aaf417c914396645a0fa852624801b24ebb7ae78fe8272889ac888"
[[package]]
name = "hashbrown"
version = "0.15.3"
@@ -1550,6 +1715,12 @@ version = "0.3.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d231dfb89cfffdbc30e7fc41579ed6066ad03abda9e567ccafae602b97ec5024"
[[package]]
name = "hex"
version = "0.4.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "7f24254aa9a54b5c858eaee2f5bccdb46aaf0e486a595ed5fd8f86ba55232a70"
[[package]]
name = "hex-literal"
version = "0.4.1"
@@ -1740,6 +1911,21 @@ dependencies = [
"tower-service",
]
[[package]]
name = "hyper-named-pipe"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "73b7d8abf35697b81a825e386fc151e0d503e8cb5fcb93cc8669c376dfd6f278"
dependencies = [
"hex",
"hyper 1.6.0",
"hyper-util",
"pin-project-lite",
"tokio",
"tower-service",
"winapi",
]
[[package]]
name = "hyper-rustls"
version = "0.27.5"
@@ -1821,6 +2007,21 @@ dependencies = [
"tracing",
]
[[package]]
name = "hyperlocal"
version = "0.9.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "986c5ce3b994526b3cd75578e62554abd09f0899d6206de48b3e96ab34ccc8c7"
dependencies = [
"hex",
"http-body-util",
"hyper 1.6.0",
"hyper-util",
"pin-project-lite",
"tokio",
"tower-service",
]
[[package]]
name = "iana-time-zone"
version = "0.1.63"
@@ -1964,6 +2165,17 @@ version = "0.3.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ce23b50ad8242c51a442f3ff322d56b02f08852c77e4c0b4d3fd684abc89c683"
[[package]]
name = "indexmap"
version = "1.9.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bd070e393353796e801d209ad339e89596eb4c8d430d18ede6a1cced8fafbd99"
dependencies = [
"autocfg",
"hashbrown 0.12.3",
"serde",
]
[[package]]
name = "indexmap"
version = "2.9.0"
@@ -1971,7 +2183,8 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cea70ddb795996207ad57735b50c5982d8844f38ba9ee5f1aedcfb708a2aa11e"
dependencies = [
"equivalent",
"hashbrown",
"hashbrown 0.15.3",
"serde",
]
[[package]]
@@ -2309,7 +2522,7 @@ version = "0.12.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "234cf4f4a04dc1f57e24b96cc0cd600cf2af460d4161ac5ecdd0af8e1f3b2a38"
dependencies = [
"hashbrown",
"hashbrown 0.15.3",
]
[[package]]
@@ -3659,6 +3872,9 @@ name = "semver"
version = "1.0.26"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "56e6fa9c48d24d85fb3de5ad847117517440f6beceb7798af16b4a87d616b8d0"
dependencies = [
"serde",
]
[[package]]
name = "serde"
@@ -3669,6 +3885,17 @@ dependencies = [
"serde_derive",
]
[[package]]
name = "serde-untagged"
version = "0.1.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "299d9c19d7d466db4ab10addd5703e4c615dec2a5a16dbbafe191045e87ee66e"
dependencies = [
"erased-serde",
"serde",
"typeid",
]
[[package]]
name = "serde-value"
version = "0.7.0"
@@ -3712,6 +3939,26 @@ dependencies = [
"serde",
]
[[package]]
name = "serde_repr"
version = "0.1.20"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "175ee3e80ae9982737ca543e96133087cbd9a485eecc3bc4de9c1a37b47ea59c"
dependencies = [
"proc-macro2",
"quote",
"syn",
]
[[package]]
name = "serde_spanned"
version = "0.6.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bf41e0cfaf7226dca15e8197172c295a782857fcb97fad1808a166870dee75a3"
dependencies = [
"serde",
]
[[package]]
name = "serde_tokenstream"
version = "0.2.2"
@@ -3736,13 +3983,30 @@ dependencies = [
"serde",
]
[[package]]
name = "serde_with"
version = "3.12.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d6b6f7f2fcb69f747921f79f3926bd1e203fce4fef62c268dd3abfb6d86029aa"
dependencies = [
"base64 0.22.1",
"chrono",
"hex",
"indexmap 1.9.3",
"indexmap 2.9.0",
"serde",
"serde_derive",
"serde_json",
"time",
]
[[package]]
name = "serde_yaml"
version = "0.9.34+deprecated"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6a8b1a1a2ebf674015cc02edccce75287f1a0130d394307b36743c2f5d504b47"
dependencies = [
"indexmap",
"indexmap 2.9.0",
"itoa",
"ryu",
"serde",
@@ -4286,6 +4550,47 @@ dependencies = [
"tokio",
]
[[package]]
name = "toml"
version = "0.8.23"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "dc1beb996b9d83529a9e75c17a1686767d148d70663143c7854d8b4a09ced362"
dependencies = [
"serde",
"serde_spanned",
"toml_datetime",
"toml_edit",
]
[[package]]
name = "toml_datetime"
version = "0.6.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "22cddaf88f4fbc13c51aebbf5f8eceb5c7c5a9da2ac40a13519eb5b0a0e8f11c"
dependencies = [
"serde",
]
[[package]]
name = "toml_edit"
version = "0.22.27"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "41fe8c660ae4257887cf66394862d21dbca4a6ddd26f04a3560410406a2f819a"
dependencies = [
"indexmap 2.9.0",
"serde",
"serde_spanned",
"toml_datetime",
"toml_write",
"winnow",
]
[[package]]
name = "toml_write"
version = "0.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5d99f8c9a7727884afe522e9bd5edbfc91a3312b36a77b5fb8926e4c31a41801"
[[package]]
name = "tower"
version = "0.5.2"
@@ -4411,6 +4716,12 @@ dependencies = [
"unicode-segmentation",
]
[[package]]
name = "typeid"
version = "1.0.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bc7d623258602320d5c55d1bc22793b57daff0ec7efc270ea7d55ce1d5f5471c"
[[package]]
name = "typenum"
version = "1.18.0"
@@ -4458,6 +4769,12 @@ version = "0.2.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1fc81956842c57dac11422a97c3b8195a1ff727f06e85c84ed2e8aa277c9a0fd"
[[package]]
name = "unicode-xid"
version = "0.2.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ebc1c04c71510c7f702b52b7c350734c9ff1295c464a03335b00bb84fc54f853"
[[package]]
name = "universal-hash"
version = "0.5.1"
@@ -4995,6 +5312,15 @@ version = "0.53.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "271414315aff87387382ec3d271b52d7ae78726f5d44ac98b4f4030c91880486"
[[package]]
name = "winnow"
version = "0.7.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "74c7b26e3480b707944fc872477815d29a8e429d2f93a1ce000f5fa84a15cbcd"
dependencies = [
"memchr",
]
[[package]]
name = "winreg"
version = "0.50.0"

View File

@@ -11,6 +11,7 @@ members = [
"opnsense-config-xml",
"harmony_cli",
"k3d",
"harmony_composer",
]
[workspace.package]
@@ -23,8 +24,13 @@ log = "0.4.22"
env_logger = "0.11.5"
derive-new = "0.7.0"
async-trait = "0.1.82"
tokio = { version = "1.40.0", features = ["io-std", "fs", "macros", "rt-multi-thread"] }
cidr = "0.2.3"
tokio = { version = "1.40.0", features = [
"io-std",
"fs",
"macros",
"rt-multi-thread",
] }
cidr = { features = ["serde"], version = "0.2" }
russh = "0.45.0"
russh-keys = "0.45.0"
rand = "0.8.5"
@@ -35,7 +41,8 @@ serde_yaml = "0.9.34"
serde-value = "0.7.0"
http = "1.2.0"
inquire = "0.7.5"
convert_case = "0.8.0"
convert_case = "0.8.0"
chrono = "0.4"
[workspace.dependencies.uuid]
version = "1.11.0"

25
Dockerfile Normal file
View File

@@ -0,0 +1,25 @@
FROM docker.io/rust:1.87.0 AS build
WORKDIR /app
COPY . .
RUN cargo build --release --bin harmony_composer
FROM docker.io/rust:1.87.0
WORKDIR /app
RUN rustup target add x86_64-pc-windows-gnu
RUN rustup target add x86_64-unknown-linux-gnu
RUN rustup component add rustfmt
RUN apt update
# TODO: Consider adding more supported targets
# nodejs for checkout action, docker for building containers, mingw for cross-compiling for windows
RUN apt install -y nodejs docker.io mingw-w64
COPY --from=build /app/target/release/harmony_composer .
ENTRYPOINT ["/app/harmony_composer"]

252
README.md
View File

@@ -1,171 +1,151 @@
# Harmony : Open Infrastructure Orchestration
# Harmony : Open-source infrastructure orchestration that treats your platform like first-class code.
*By [NationTech](https://nationtech.io)*
## Quick demo
[![Build](https://git.nationtech.io/NationTech/harmony/actions/workflows/check.yml/badge.svg)](https://git.nationtech.io/nationtech/harmony)
[![License](https://img.shields.io/badge/license-AGPLv3-blue?style=flat-square)](LICENSE)
`cargo run -p example-tui`
### Unify
This will launch Harmony's minimalist terminal ui which embeds a few demo scores.
- **Project Scaffolding**
- **Infrastructure Provisioning**
- **Application Deployment**
- **Day-2 operations**
Usage instructions will be displayed at the bottom of the TUI.
All in **one strongly-typed Rust codebase**.
`cargo run --bin example-cli -- --help`
### Deploy anywhere
This is the harmony CLI, a minimal implementation
From a **developer laptop** to a **global production cluster**, a single **source of truth** drives the **full software lifecycle.**
The current help text:
---
````
Usage: example-cli [OPTIONS]
## 1 · The Harmony Philosophy
Options:
-y, --yes Run score(s) or not
-f, --filter <FILTER> Filter query
-i, --interactive Run interactive TUI or not
-a, --all Run all or nth, defaults to all
-n, --number <NUMBER> Run nth matching, zero indexed [default: 0]
-l, --list list scores, will also be affected by run filter
-h, --help Print help
-V, --version Print version```
Infrastructure is essential, but it shouldnt be your core business. Harmony is built on three guiding principles that make modern platforms reliable, repeatable, and easy to reason about.
## Core architecture
| Principle | What it means for you |
|-----------|-----------------------|
| **Infrastructure as Resilient Code** | Replace sprawling YAML and bash scripts with type-safe Rust. Test, refactor, and version your platform just like application code. |
| **Prove It Works — Before You Deploy** | Harmony uses the compiler to verify that your applications needs match the target environments capabilities at **compile-time**, eliminating an entire class of runtime outages. |
| **One Unified Model** | Software and infrastructure are a single system. Harmony models them together, enabling deep automation—from bare-metal servers to Kubernetes workloads—with zero context switching. |
![Harmony Core Architecture](docs/diagrams/Harmony_Core_Architecture.drawio.svg)
````
## Supporting a new field in OPNSense `config.xml`
These principles surface as simple, ergonomic Rust APIs that let teams focus on their product while trusting the platform underneath.
Two steps:
- Supporting the field in `opnsense-config-xml`
- Enabling Harmony to control the field
---
We'll use the `filename` field in the `dhcpcd` section of the file as an example.
## 2 · Quick Start
### Supporting the field
The snippet below spins up a complete **production-grade LAMP stack** with monitoring. Swap it for your own scores to deploy anything from microservices to machine-learning pipelines.
As type checking if enforced, every field from `config.xml` must be known by the code. Each subsection of `config.xml` has its `.rs` file. For the `dhcpcd` section, we'll modify `opnsense-config-xml/src/data/dhcpd.rs`.
```rust
use harmony::{
data::Version,
inventory::Inventory,
maestro::Maestro,
modules::{
lamp::{LAMPConfig, LAMPScore},
monitoring::monitoring_alerting::MonitoringAlertingStackScore,
},
topology::{K8sAnywhereTopology, Url},
};
When a new field appears in the xml file, an error like this will be thrown and Harmony will panic :
```
Running `/home/stremblay/nt/dir/harmony/target/debug/example-nanodc`
Found unauthorized element filename
thread 'main' panicked at opnsense-config-xml/src/data/opnsense.rs:54:14:
OPNSense received invalid string, should be full XML: ()
#[tokio::main]
async fn main() {
// 1. Describe what you want
let lamp_stack = LAMPScore {
name: "harmony-lamp-demo".into(),
domain: Url::Url(url::Url::parse("https://lampdemo.example.com").unwrap()),
php_version: Version::from("8.3.0").unwrap(),
config: LAMPConfig {
project_root: "./php".into(),
database_size: "4Gi".into(),
..Default::default()
},
};
// 2. Pick where it should run
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
Inventory::autoload(), // auto-detect hardware / kube-config
K8sAnywhereTopology::from_env(), // local k3d, CI, staging, prod…
)
.await
.unwrap();
// 3. Enhance with extra scores (monitoring, CI/CD, …)
let mut monitoring = MonitoringAlertingStackScore::new();
monitoring.namespace = Some(lamp_stack.config.namespace.clone());
maestro.register_all(vec![Box::new(lamp_stack), Box::new(monitoring)]);
// 4. Launch an interactive CLI / TUI
harmony_cli::init(maestro, None).await.unwrap();
}
```
Define the missing field (`filename`) in the `DhcpInterface` struct of `opnsense-config-xml/src/data/dhcpd.rs`:
```
pub struct DhcpInterface {
...
pub filename: Option<String>,
Run it:
```bash
cargo run
```
Harmony should now be fixed, build and run.
Harmony analyses the code, shows an execution plan in a TUI, and applies it once you confirm. Same code, same binary—every environment.
### Controlling the field
---
Define the `xml field setter` in `opnsense-config/src/modules/dhcpd.rs`.
```
impl<'a> DhcpConfig<'a> {
...
pub fn set_filename(&mut self, filename: &str) {
self.enable_netboot();
self.get_lan_dhcpd().filename = Some(filename.to_string());
}
...
## 3 · Core Concepts
| Term | One-liner |
|------|-----------|
| **Score<T>** | Declarative description of the desired state (e.g., `LAMPScore`). |
| **Interpret<T>** | Imperative logic that realises a `Score` on a specific environment. |
| **Topology** | An environment (local k3d, AWS, bare-metal) exposing verified *Capabilities* (Kubernetes, DNS, …). |
| **Maestro** | Orchestrator that compiles Scores + Topology, ensuring all capabilities line up **at compile-time**. |
| **Inventory** | Optional catalogue of physical assets for bare-metal and edge deployments. |
A visual overview is in the diagram below.
[Harmony Core Architecture](docs/diagrams/Harmony_Core_Architecture.drawio.svg)
---
## 4 · Install
Prerequisites:
* Rust
* Docker (if you deploy locally)
* `kubectl` / `helm` for Kubernetes-based topologies
```bash
git clone https://git.nationtech.io/nationtech/harmony
cd harmony
cargo build --release # builds the CLI, TUI and libraries
```
Define the `value setter` in the `DhcpServer trait` in `domain/topology/network.rs`
```
#[async_trait]
pub trait DhcpServer: Send + Sync {
...
async fn set_filename(&self, filename: &str) -> Result<(), ExecutorError>;
...
```
---
Implement the `value setter` in each `DhcpServer` implementation.
`infra/opnsense/dhcp.rs`:
```
#[async_trait]
impl DhcpServer for OPNSenseFirewall {
...
async fn set_filename(&self, filename: &str) -> Result<(), ExecutorError> {
{
let mut writable_opnsense = self.opnsense_config.write().await;
writable_opnsense.dhcp().set_filename(filename);
debug!("OPNsense dhcp server set filename {filename}");
}
## 5 · Learning More
Ok(())
}
...
```
* **Architectural Decision Records** dive into the rationale
- [ADR-001 · Why Rust](adr/001-rust.md)
- [ADR-003 · Infrastructure Abstractions](adr/003-infrastructure-abstractions.md)
- [ADR-006 · Secret Management](adr/006-secret-management.md)
- [ADR-011 · Multi-Tenant Cluster](adr/011-multi-tenant-cluster.md)
`domain/topology/ha_cluster.rs`
```
#[async_trait]
impl DhcpServer for DummyInfra {
...
async fn set_filename(&self, _filename: &str) -> Result<(), ExecutorError> {
unimplemented!("{}", UNIMPLEMENTED_DUMMY_INFRA)
}
...
```
* **Extending Harmony** write new Scores / Interprets, add hardware like OPNsense firewalls, or embed Harmony in your own tooling (`/docs`).
Add the new field to the DhcpScore in `modules/dhcp.rs`
```
pub struct DhcpScore {
...
pub filename: Option<String>,
```
* **Community** discussions and roadmap live in [GitLab issues](https://git.nationtech.io/nationtech/harmony/-/issues). PRs, ideas, and feedback are welcome!
Define it in its implementation in `modules/okd/dhcp.rs`
```
impl OKDDhcpScore {
...
Self {
dhcp_score: DhcpScore {
...
filename: Some("undionly.kpxe".to_string()),
```
---
Define it in its implementation in `modules/okd/bootstrap_dhcp.rs`
```
impl OKDDhcpScore {
...
Self {
dhcp_score: DhcpScore::new(
...
Some("undionly.kpxe".to_string()),
```
## 6 · License
Update the interpret (function called by the `execute` fn of the interpret) so it now updates the `filename` field value in `modules/dhcp.rs`
```
impl DhcpInterpret {
...
let filename_outcome = match &self.score.filename {
Some(filename) => {
let dhcp_server = Arc::new(topology.dhcp_server.clone());
dhcp_server.set_filename(&filename).await?;
Outcome::new(
InterpretStatus::SUCCESS,
format!("Dhcp Interpret Set filename to {filename}"),
)
}
None => Outcome::noop(),
};
Harmony is released under the **GNU AGPL v3**.
if next_server_outcome.status == InterpretStatus::NOOP
&& boot_filename_outcome.status == InterpretStatus::NOOP
&& filename_outcome.status == InterpretStatus::NOOP
> We choose a strong copyleft license to ensure the project—and every improvement to it—remains open and benefits the entire community. Fork it, enhance it, even out-innovate us; just keep it open.
...
See [LICENSE](LICENSE) for the full text.
Ok(Outcome::new(
InterpretStatus::SUCCESS,
format!(
"Dhcp Interpret Set next boot to [{:?}], boot_filename to [{:?}], filename to [{:?}]",
self.score.boot_filename, self.score.boot_filename, self.score.filename
)
...
```
---
*Made with ❤️ & 🦀 by the NationTech and the Harmony community*

View File

@@ -1,6 +1,6 @@
# Architecture Decision Record: \<Title\>
Name: \<Name\>
Initial Author: \<Name\>
Initial Date: \<Date\>

View File

@@ -1,6 +1,6 @@
# Architecture Decision Record: Helm and Kustomize Handling
Name: Taha Hawa
Initial Author: Taha Hawa
Initial Date: 2025-04-15

View File

@@ -0,0 +1,73 @@
pub trait MonitoringSystem {}
// 1. Modified AlertReceiver trait:
// - Removed the problematic `clone` method.
// - Added `box_clone` which returns a Box<dyn AlertReceiver>.
pub trait AlertReceiver {
type M: MonitoringSystem;
fn install(&self, sender: &Self::M) -> Result<(), String>;
// This method allows concrete types to clone themselves into a Box<dyn AlertReceiver>
fn box_clone(&self) -> Box<dyn AlertReceiver<M = Self::M>>;
}
#[derive(Clone)]
struct Prometheus{}
impl MonitoringSystem for Prometheus {}
#[derive(Clone)] // Keep derive(Clone) for DiscordWebhook itself
struct DiscordWebhook{}
impl AlertReceiver for DiscordWebhook {
type M = Prometheus;
fn install(&self, sender: &Self::M) -> Result<(), String> {
// Placeholder for actual installation logic
println!("DiscordWebhook installed for Prometheus monitoring.");
Ok(())
}
// 2. Implement `box_clone` for DiscordWebhook:
// This uses the derived `Clone` for DiscordWebhook to create a new boxed instance.
fn box_clone(&self) -> Box<dyn AlertReceiver<M = Self::M>> {
Box::new(self.clone())
}
}
// 3. Implement `std::clone::Clone` for `Box<dyn AlertReceiver<M= M>>`:
// This allows `Box<dyn AlertReceiver>` to be cloned.
// The `+ 'static` lifetime bound is often necessary for trait objects stored in collections,
// ensuring they live long enough.
impl<M: MonitoringSystem + 'static> Clone for Box<dyn AlertReceiver<M= M>> {
fn clone(&self) -> Self {
self.box_clone() // Call the custom `box_clone` method
}
}
// MonitoringConfig can now derive Clone because its `receivers` field
// (Vec<Box<dyn AlertReceiver<M = M>>>) is now cloneable.
#[derive(Clone)]
struct MonitoringConfig <M: MonitoringSystem + 'static>{
receivers: Vec<Box<dyn AlertReceiver<M = M>>>
}
// Example usage to demonstrate compilation and functionality
fn main() {
let prometheus_instance = Prometheus{};
let discord_webhook_instance = DiscordWebhook{};
let mut config = MonitoringConfig {
receivers: Vec::new()
};
// Create a boxed alert receiver
let boxed_receiver: Box<dyn AlertReceiver<M = Prometheus>> = Box::new(discord_webhook_instance);
config.receivers.push(boxed_receiver);
// Clone the config, which will now correctly clone the boxed receiver
let cloned_config = config.clone();
println!("Original config has {} receivers.", config.receivers.len());
println!("Cloned config has {} receivers.", cloned_config.receivers.len());
// Example of using the installed receiver
if let Some(receiver) = config.receivers.get(0) {
let _ = receiver.install(&prometheus_instance);
}
}

View File

@@ -1,7 +1,7 @@
# Architecture Decision Record: Monitoring and Alerting
Proposed by: Willem Rolleman
Date: April 28 2025
Initial Author : Willem Rolleman
Date : April 28 2025
## Status

View File

@@ -0,0 +1,161 @@
# Architecture Decision Record: Multi-Tenancy Strategy for Harmony Managed Clusters
Initial Author: Jean-Gabriel Gill-Couture
Initial Date: 2025-05-26
## Status
Proposed
## Context
Harmony manages production OKD/Kubernetes clusters that serve multiple clients with varying trust levels and operational requirements. We need a multi-tenancy strategy that provides:
1. **Strong isolation** between client workloads while maintaining operational simplicity
2. **Controlled API access** allowing clients self-service capabilities within defined boundaries
3. **Security-first approach** protecting both the cluster infrastructure and tenant data
4. **Harmony-native implementation** using our Score/Interpret pattern for automated tenant provisioning
5. **Scalable management** supporting both small trusted clients and larger enterprise customers
The official Kubernetes multi-tenancy documentation identifies two primary models: namespace-based isolation and virtual control planes per tenant. Given Harmony's focus on operational simplicity, provider-agnostic abstractions (ADR-003), and hexagonal architecture (ADR-002), we must choose an approach that balances security, usability, and maintainability.
Our clients represent a hybrid tenancy model:
- **Customer multi-tenancy**: Each client operates independently with no cross-tenant trust
- **Team multi-tenancy**: Individual clients may have multiple team members requiring coordinated access
- **API access requirement**: Unlike pure SaaS scenarios, clients need controlled Kubernetes API access for self-service operations
The official kubernetes documentation on multi tenancy heavily inspired this ADR : https://kubernetes.io/docs/concepts/security/multi-tenancy/
## Decision
Implement **namespace-based multi-tenancy** with the following architecture:
### 1. Network Security Model
- **Private cluster access**: Kubernetes API and OpenShift console accessible only via WireGuard VPN
- **No public exposure**: Control plane endpoints remain internal to prevent unauthorized access attempts
- **VPN-based authentication**: Initial access control through WireGuard client certificates
### 2. Tenant Isolation Strategy
- **Dedicated namespace per tenant**: Each client receives an isolated namespace with access limited only to the required resources and operations
- **Complete network isolation**: NetworkPolicies prevent cross-namespace communication while allowing full egress to public internet
- **Resource governance**: ResourceQuotas and LimitRanges enforce CPU, memory, and storage consumption limits
- **Storage access control**: Clients can create PersistentVolumeClaims but cannot directly manipulate PersistentVolumes or access other tenants' storage
### 3. Access Control Framework
- **Principle of Least Privilege**: RBAC grants only necessary permissions within tenant namespace scope
- **Namespace-scoped**: Clients can create/modify/delete resources within their namespace
- **Cluster-level restrictions**: No access to cluster-wide resources, other namespaces, or sensitive cluster operations
- **Whitelisted operations**: Controlled self-service capabilities for ingress, secrets, configmaps, and workload management
### 4. Identity Management Evolution
- **Phase 1**: Manual provisioning of VPN access and Kubernetes ServiceAccounts/Users
- **Phase 2**: Migration to Keycloak-based identity management (aligning with ADR-006) for centralized authentication and lifecycle management
### 5. Harmony Integration
- **TenantScore implementation**: Declarative tenant provisioning using Harmony's Score/Interpret pattern
- **Topology abstraction**: Tenant configuration abstracted from underlying Kubernetes implementation details
- **Automated deployment**: Complete tenant setup automated through Harmony's orchestration capabilities
## Rationale
### Network Security Through VPN Access
- **Defense in depth**: VPN requirement adds critical security layer preventing unauthorized cluster access
- **Simplified firewall rules**: No need for complex public endpoint protections or rate limiting
- **Audit capability**: VPN access provides clear audit trail of cluster connections
- **Aligns with enterprise practices**: Most enterprise customers already use VPN infrastructure
### Namespace Isolation vs Virtual Control Planes
Following Kubernetes official guidance, namespace isolation provides:
- **Lower resource overhead**: Virtual control planes require dedicated etcd, API server, and controller manager per tenant
- **Operational simplicity**: Single control plane to maintain, upgrade, and monitor
- **Cross-tenant service integration**: Enables future controlled cross-tenant communication if required
- **Proven stability**: Namespace-based isolation is well-tested and widely deployed
- **Cost efficiency**: Significantly lower infrastructure costs compared to dedicated control planes
### Hybrid Tenancy Model Suitability
Our approach addresses both customer and team multi-tenancy requirements:
- **Customer isolation**: Strong network and RBAC boundaries prevent cross-tenant interference
- **Team collaboration**: Multiple team members can share namespace access through group-based RBAC
- **Self-service balance**: Controlled API access enables client autonomy without compromising security
### Harmony Architecture Alignment
- **Provider agnostic**: TenantScore abstracts multi-tenancy concepts, enabling future support for other Kubernetes distributions
- **Hexagonal architecture**: Tenant management becomes an infrastructure capability accessed through well-defined ports
- **Declarative automation**: Tenant lifecycle fully managed through Harmony's Score execution model
## Consequences
### Positive Consequences
- **Strong security posture**: VPN + namespace isolation provides robust tenant separation
- **Operational efficiency**: Single cluster management with automated tenant provisioning
- **Client autonomy**: Self-service capabilities reduce operational support burden
- **Scalable architecture**: Can support hundreds of tenants per cluster without architectural changes
- **Future flexibility**: Foundation supports evolution to more sophisticated multi-tenancy models
- **Cost optimization**: Shared infrastructure maximizes resource utilization
### Negative Consequences
- **VPN operational overhead**: Requires VPN infrastructure management
- **Manual provisioning complexity**: Phase 1 manual user management creates administrative burden
- **Network policy dependency**: Requires CNI with NetworkPolicy support (OVN-Kubernetes provides this and is the OKD/Openshift default)
- **Cluster-wide resource limitations**: Some advanced Kubernetes features require cluster-wide access
- **Single point of failure**: Cluster outage affects all tenants simultaneously
### Migration Challenges
- **Legacy client integration**: Existing clients may need VPN client setup and credential migration
- **Monitoring complexity**: Per-tenant observability requires careful metric and log segmentation
- **Backup considerations**: Tenant data backup must respect isolation boundaries
## Alternatives Considered
### Alternative 1: Virtual Control Plane Per Tenant
**Pros**: Complete control plane isolation, full Kubernetes API access per tenant
**Cons**: 3-5x higher resource usage, complex cross-tenant networking, operational complexity scales linearly with tenants
**Rejected**: Resource overhead incompatible with cost-effective multi-tenancy goals
### Alternative 2: Dedicated Clusters Per Tenant
**Pros**: Maximum isolation, independent upgrade cycles, simplified security model
**Cons**: Exponential operational complexity, prohibitive costs, resource waste
**Rejected**: Operational overhead makes this approach unsustainable for multiple clients
### Alternative 3: Public API with Advanced Authentication
**Pros**: No VPN requirement, potentially simpler client access
**Cons**: Larger attack surface, complex rate limiting and DDoS protection, increased security monitoring requirements
**Rejected**: Risk/benefit analysis favors VPN-based access control
### Alternative 4: Service Mesh Based Isolation
**Pros**: Fine-grained traffic control, encryption, advanced observability
**Cons**: Significant operational complexity, performance overhead, steep learning curve
**Rejected**: Complexity overhead outweighs benefits for current requirements; remains option for future enhancement
## Additional Notes
### Implementation Roadmap
1. **Phase 1**: Implement VPN access and manual tenant provisioning
2. **Phase 2**: Deploy TenantScore automation for namespace, RBAC, and NetworkPolicy management
4. **Phase 3**: Work on privilege escalation from pods, audit for weaknesses, enforce security policies on pod runtimes
3. **Phase 4**: Integrate Keycloak for centralized identity management
4. **Phase 5**: Add advanced monitoring and per-tenant observability
### TenantScore Structure Preview
```rust
pub struct TenantScore {
pub tenant_config: TenantConfig,
pub resource_quotas: ResourceQuotaConfig,
pub network_isolation: NetworkIsolationPolicy,
pub storage_access: StorageAccessConfig,
pub rbac_config: RBACConfig,
}
```
### Future Enhancements
- **Cross-tenant service mesh**: For approved inter-tenant communication
- **Advanced monitoring**: Per-tenant Prometheus/Grafana instances
- **Backup automation**: Tenant-scoped backup policies
- **Cost allocation**: Detailed per-tenant resource usage tracking
This ADR establishes the foundation for secure, scalable multi-tenancy in Harmony-managed clusters while maintaining operational simplicity and cost effectiveness. A follow-up ADR will detail the Tenant abstraction and user management mechanisms within the Harmony framework.

View File

@@ -0,0 +1,41 @@
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
name: tenant-isolation-policy
namespace: testtenant
spec:
podSelector: {} # Selects all pods in the namespace
policyTypes:
- Ingress
- Egress
ingress:
- from:
- podSelector: {} # Allow from all pods in the same namespace
egress:
- to:
- podSelector: {} # Allow to all pods in the same namespace
- to:
- podSelector: {}
namespaceSelector:
matchLabels:
kubernetes.io/metadata.name: openshift-dns # Target the openshift-dns namespace
# Note, only opening port 53 is not enough, will have to dig deeper into this one eventually
# ports:
# - protocol: UDP
# port: 53
# - protocol: TCP
# port: 53
# Allow egress to public internet only
- to:
- ipBlock:
cidr: 0.0.0.0/0
except:
- 10.0.0.0/8 # RFC1918
- 172.16.0.0/12 # RFC1918
- 192.168.0.0/16 # RFC1918
- 169.254.0.0/16 # Link-local
- 127.0.0.0/8 # Loopback
- 224.0.0.0/4 # Multicast
- 240.0.0.0/4 # Reserved
- 100.64.0.0/10 # Carrier-grade NAT
- 0.0.0.0/8 # Reserved

View File

@@ -0,0 +1,95 @@
apiVersion: v1
kind: Namespace
metadata:
name: testtenant
---
apiVersion: v1
kind: Namespace
metadata:
name: testtenant2
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: test-web
namespace: testtenant
spec:
replicas: 1
selector:
matchLabels:
app: test-web
template:
metadata:
labels:
app: test-web
spec:
containers:
- name: nginx
image: nginxinc/nginx-unprivileged
ports:
- containerPort: 80
---
apiVersion: v1
kind: Service
metadata:
name: test-web
namespace: testtenant
spec:
selector:
app: test-web
ports:
- port: 80
targetPort: 8080
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: test-client
namespace: testtenant
spec:
replicas: 1
selector:
matchLabels:
app: test-client
template:
metadata:
labels:
app: test-client
spec:
containers:
- name: curl
image: curlimages/curl:latest
command: ["/bin/sh", "-c", "sleep 3600"]
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: test-web
namespace: testtenant2
spec:
replicas: 1
selector:
matchLabels:
app: test-web
template:
metadata:
labels:
app: test-web
spec:
containers:
- name: nginx
image: nginxinc/nginx-unprivileged
ports:
- containerPort: 80
---
apiVersion: v1
kind: Service
metadata:
name: test-web
namespace: testtenant2
spec:
selector:
app: test-web
ports:
- port: 80
targetPort: 8080

View File

@@ -0,0 +1,63 @@
# Architecture Decision Record: \<Title\>
Initial Author: Jean-Gabriel Gill-Couture
Initial Date: 2025-06-04
Last Updated Date: 2025-06-04
## Status
Proposed
## Context
As Harmony's goal is to make software delivery easier, we must provide an easy way for developers to express their app's semantics and dependencies with great abstractions, in a similar fashion to what the score.dev project is doing.
Thus, we started working on ways to package common types of applications such as LAMP, which we started working on with `LAMPScore`.
Now is time for the next step : we want to pave the way towards complete lifecycle automation. To do this, we will start with a way to execute Harmony's modules easily from anywhere, starting with locally and in CI environments.
## Decision
To achieve easy, portable execution of Harmony, we will follow this architecture :
- Host a basic harmony release that is compiled with the CLI by our gitea/github server
- This binary will do the following : check if there is a `harmony` folder in the current path
- If yes
- Check if cargo is available locally and compile the harmony binary, or compile the harmony binary using a rust docker container, if neither cargo or a container runtime is available, output a message explaining the situation
- Run the newly compiled binary. (Ideally using pid handoff like exec does but some research around this should be done. I think handing off the process is to help with OS interaction such as terminal apps, signals, exit codes, process handling, etc but there might be some side effects)
- If not
- Suggest initializing a project by auto detecting what the project looks like
- When the project type cannot be auto detected, provide links to Harmony's documentation on how to set up a project, a link to the examples folder, and a ask the user if he wants to initialize an empty Harmony project in the current folder
- harmony/Cargo.toml with dependencies set
- harmony/src/main.rs with an example LAMPScore setup and ready to run
- This same binary can be used in a CI environment to run the target project's Harmony module. By default, we provide these opinionated steps :
1. **An empty check step.** The purpose of this step is to run all tests and checks against the codebase. For complex projects this could involve a very complex pipeline of test environments setup and execution but this is out of scope for now. This is not handled by harmony. For projects with automatic setup, we can fill this step with something like `cargo fmt --check; cargo test; cargo build` but Harmony is not directly involved in the execution of this step.
2. **Package and publish.** Once all checks have passed, the production ready container is built and pushed to a registry. This is done by Harmony.
3. **Deploy to staging automatically.**
4. **Run a sanity check on staging.** As Harmony is responsible for deploying, Harmony should have all the knowledge of how to perform a sanity check on the staging environment. This will, most of the time, be a simple verification of the kubernetes health of all deployed components, and a poke on the public endpoint when there is one.
5. **Deploy to production automatically.** Many projects will require manual approval here, this can be easily set up in the CI afterwards, but our opinion is that
6. **Run a sanity check on production.** Same check as staging, but on production.
*Note on providing a base pipeline :* Having a complete pipeline set up automatically will encourage development teams to build upon these by adding tests where they belong. The goal here is to provide an opiniated solution that works for most small and large projects. Of course, many orgnizations will need to add steps such as deploying to sandbox environments, requiring more advanced approvals, more complex publication and coordination with other projects. But this here encompasses the basics required to build and deploy software reliably at any scale.
### Environment setup
TBD : For now, environments (tenants) will be set up and configured manually. Harmony will rely on the kubeconfig provided in the environment where it is running to deploy in the namespace.
For the CD tool such as Argo or Flux they will be activated by default by Harmony when using application level Scores such as LAMPScore in a similar way that the container is automatically built. Then, CI deployment steps will be notifying the CD tool using its API of the new release to deploy.
## Rationale
Reasoning behind the decision
## Consequences
Pros/Cons of chosen solution
## Alternatives considered
Pros/Cons of various proposed solutions considered
## Additional Notes

1
docs/README.md Normal file
View File

@@ -0,0 +1 @@
Not much here yet, see the `adr` folder for now. More to come in time!

13
docs/cyborg-metaphor.md Normal file
View File

@@ -0,0 +1,13 @@
## Conceptual metaphor : The Cyborg and the Central Nervous System
At the heart of Harmony lies a core belief: in modern, decentralized systems, **software and infrastructure are not separate entities.** They are a single, symbiotic organism—a cyborg.
The software is the electronics, the "mind"; the infrastructure is the biological host, the "body". They live or die, thrive or sink together.
Traditional approaches attempt to manage this complex organism with fragmented tools: static YAML for configuration, brittle scripts for automation, and separate Infrastructure as Code (IaC) for provisioning. This creates a disjointed system that struggles to scale or heal itself, making it inadequate for the demands of fully automated, enterprise-grade clusters.
Harmony's goal is to provide the **central nervous system for this cyborg**. We aim to achieve the full automation of complex, decentralized clouds by managing this integrated entity holistically.
To achieve this, a tool must be both robust and powerful. It must manage the entire lifecycle—deployment, upgrades, failure recovery, and decommissioning—with precision. This requires full control over application packaging and a deep, intrinsic integration between the software and the infrastructure it inhabits.
This is why Harmony uses a powerful, living language like Rust. It replaces static, lifeless configuration files with a dynamic, breathing codebase. It allows us to express the complex relationships and behaviors of a modern distributed system, enabling the creation of truly automated, resilient, and powerful platforms that can thrive.

View File

@@ -16,5 +16,3 @@ harmony_macros = { path = "../../harmony_macros" }
log = { workspace = true }
env_logger = { workspace = true }
url = { workspace = true }
typetag = "0.2.20"
serde = "1.0.219"

View File

@@ -2,10 +2,7 @@ use harmony::{
data::Version,
inventory::Inventory,
maestro::Maestro,
modules::{
lamp::{LAMPConfig, LAMPScore},
monitoring::{kube_prometheus::prometheus_alert_channel::{DiscordChannel, SlackChannel}, monitoring_alerting::MonitoringAlertingScore},
},
modules::lamp::{LAMPConfig, LAMPScore},
topology::{K8sAnywhereTopology, Url},
};
@@ -32,42 +29,28 @@ async fn main() {
},
};
//let monitoring = MonitoringAlertingScore {
// alert_receivers: vec![Box::new(DiscordWebhook {
// url: Url::Url(url::Url::parse("https://discord.idonotexist.com").unwrap()),
// // TODO write url macro
// // url: url!("https://discord.idonotexist.com"),
// })],
// alert_rules: vec![],
// scrape_targets: vec![],
//};
// You can choose the type of Topology you want, we suggest starting with the
// K8sAnywhereTopology as it is the most automatic one that enables you to easily deploy
// locally, to development environment from a CI, to staging, and to production with settings
// that automatically adapt to each environment grade.
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize (
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
Inventory::autoload(),
K8sAnywhereTopology::new(),
K8sAnywhereTopology::from_env(),
)
.await
.unwrap();
let url = url::Url::parse(
"https://hooks.slack.com/services/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",
)
.expect("invalid URL");
let mut monitoring_stack_score = MonitoringAlertingScore::new();
monitoring_stack_score.namespace = Some(lamp_stack.config.namespace.clone());
monitoring_stack_score.alert_channels = vec![(Box::new(SlackChannel {
name: "alert-test".to_string(),
webhook_url: url,})),
(Box::new(DiscordChannel {
name: "discord".to_string(),
webhook_url: url::Url::parse("https://discord.com/api/webhooks/1372994201746276462/YRn4TA9pj8ve3lfmyj1j0Yx97i92gv4U_uavt4CV4_SSIVArYUqfDzMOmzSTic2d8XSL").expect("invalid URL"),}))];
//TODO in process of testing
//webhook depricated in MSTeams August 2025
//(AlertChannel::MSTeams {
// connector: "alert-test".to_string(),
// webhook_url: url::Url::parse("").expect("invalid URL"),
//}),
maestro.register_all(vec![Box::new(monitoring_stack_score)]);
maestro.register_all(vec![Box::new(lamp_stack)]);
// Here we bootstrap the CLI, this gives some nice features if you need them
harmony_cli::init(maestro, None).await.unwrap();
}

View File

@@ -1,5 +1,5 @@
[package]
name = "ms_teams_alert_channel"
name = "example-monitoring"
edition = "2024"
version.workspace = true
readme.workspace = true
@@ -8,7 +8,5 @@ license.workspace = true
[dependencies]
harmony = { version = "0.1.0", path = "../../harmony" }
harmony_cli = { version = "0.1.0", path = "../../harmony_cli" }
serde = "1.0.219"
tokio.workspace = true
typetag = "0.2.20"
url.workspace = true

View File

@@ -0,0 +1,57 @@
use harmony::{
inventory::Inventory,
maestro::Maestro,
modules::{
monitoring::{
alert_channel::discord_alert_channel::DiscordWebhook,
alert_rule::prometheus_alert_rule::AlertManagerRuleGroup,
kube_prometheus::helm_prometheus_alert_score::HelmPrometheusAlertingScore,
},
prometheus::alerts::{
infra::dell_server::{
alert_global_storage_status_critical, alert_global_storage_status_non_recoverable,
global_storage_status_degraded_non_critical,
},
k8s::pvc::high_pvc_fill_rate_over_two_days,
},
},
topology::{K8sAnywhereTopology, Url},
};
#[tokio::main]
async fn main() {
let discord_receiver = DiscordWebhook {
name: "test-discord".to_string(),
url: Url::Url(url::Url::parse("https://discord.doesnt.exist.com").unwrap()),
};
let high_pvc_fill_rate_over_two_days_alert = high_pvc_fill_rate_over_two_days();
let dell_system_storage_degraded = global_storage_status_degraded_non_critical();
let alert_global_storage_status_critical = alert_global_storage_status_critical();
let alert_global_storage_status_non_recoverable = alert_global_storage_status_non_recoverable();
let additional_rules =
AlertManagerRuleGroup::new("pvc-alerts", vec![high_pvc_fill_rate_over_two_days_alert]);
let additional_rules2 = AlertManagerRuleGroup::new(
"dell-server-alerts",
vec![
dell_system_storage_degraded,
alert_global_storage_status_critical,
alert_global_storage_status_non_recoverable,
],
);
let alerting_score = HelmPrometheusAlertingScore {
receivers: vec![Box::new(discord_receiver)],
rules: vec![Box::new(additional_rules), Box::new(additional_rules2)],
};
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
)
.await
.unwrap();
maestro.register_all(vec![Box::new(alerting_score)]);
harmony_cli::init(maestro, None).await.unwrap();
}

View File

@@ -1,65 +0,0 @@
mod prometheus_msteams;
use harmony::{
interpret::InterpretError, inventory::Inventory, maestro::Maestro, modules::{helm::chart::HelmChartScore, monitoring::{kube_prometheus::{prometheus_alert_channel::PrometheusAlertChannel, types::{AlertChannelConfig, AlertChannelReceiver, AlertChannelRoute, WebhookConfig}}, monitoring_alerting::MonitoringAlertingScore}}, topology::K8sAnywhereTopology
};
use prometheus_msteams::prometheus_msteams_score;
use url::Url;
use serde::{Serialize, Deserialize};
#[tokio::main]
async fn main() {
let alert_channels: Vec<Box<dyn PrometheusAlertChannel>> = vec![Box::new(MSTeamsChannel {
connector: "teams-test".to_string(),
webhook_url: url::Url::parse(
"https://msteams.com/services/dummy/dummy/dummy",
)
.expect("invalid URL"),
})];
let monitoring_score = MonitoringAlertingScore {
alert_channels,
namespace: None,
};
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
Inventory::autoload(),
K8sAnywhereTopology::new(),
)
.await
.unwrap();
maestro.register_all(vec![Box::new(monitoring_score)]);
harmony_cli::init(maestro, None).await.unwrap();
}
#[derive(Debug, Clone, Serialize, Deserialize)]
struct MSTeamsChannel {
connector: String,
webhook_url: Url,
}
#[typetag::serde]
impl PrometheusAlertChannel for MSTeamsChannel {
fn get_alert_manager_config_contribution(&self) -> Result<AlertChannelConfig, InterpretError> {
Ok(AlertChannelConfig{
receiver: AlertChannelReceiver{
name: format!("MSTeams-{}",self.connector),
slack_configs: None,
webhook_configs: Some(vec![WebhookConfig{
url: url::Url::parse("http://prometheus-msteams-prometheus-msteams.monitoring.svc.cluster.local:2000/alertmanager").expect("invalid url"),
send_resolved: true,}])
},
route: AlertChannelRoute{
receiver: format!("MSTeams-{}", self.connector),
matchers: vec!["alertname!=Watchdog".to_string()],
r#continue: true,
},
global_config: None, })
}
fn get_dependency_score(&self, ns: String) -> Option<HelmChartScore> {
Some(prometheus_msteams_score(self.connector.clone(), self.webhook_url.clone(), ns.clone()))
}
}

View File

@@ -1,30 +0,0 @@
use std::str::FromStr;
use harmony::modules::helm::chart::{HelmChartScore, NonBlankString};
use url::Url;
pub fn prometheus_msteams_score(
name: String,
webhook_url: Url,
namespace: String,
) -> HelmChartScore {
let values = format!(
r#"
connectors:
- default: "{webhook_url}"
"#,
);
HelmChartScore {
namespace: Some(NonBlankString::from_str(&namespace).unwrap()),
release_name: NonBlankString::from_str(&name).unwrap(),
chart_name: NonBlankString::from_str("oci://hub.nationtech.io/library/prometheus-msteams")
.unwrap(),
chart_version: None,
values_overrides: None,
values_yaml: Some(values.to_string()),
create_namespace: true,
install_only: true,
repository: None,
}
}

View File

@@ -0,0 +1,18 @@
[package]
name = "example-tenant"
edition = "2024"
version.workspace = true
readme.workspace = true
license.workspace = true
publish = false
[dependencies]
harmony = { path = "../../harmony" }
harmony_cli = { path = "../../harmony_cli" }
harmony_types = { path = "../../harmony_types" }
cidr = { workspace = true }
tokio = { workspace = true }
harmony_macros = { path = "../../harmony_macros" }
log = { workspace = true }
env_logger = { workspace = true }
url = { workspace = true }

View File

@@ -0,0 +1,41 @@
use harmony::{
data::Id,
inventory::Inventory,
maestro::Maestro,
modules::tenant::TenantScore,
topology::{K8sAnywhereTopology, tenant::TenantConfig},
};
#[tokio::main]
async fn main() {
let tenant = TenantScore {
config: TenantConfig {
id: Id::from_str("test-tenant-id"),
name: "testtenant".to_string(),
..Default::default()
},
};
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
)
.await
.unwrap();
maestro.register_all(vec![Box::new(tenant)]);
harmony_cli::init(maestro, None).await.unwrap();
}
// TODO write tests
// - Create Tenant with default config mostly, make sure namespace is created
// - deploy sample client/server app with nginx unprivileged and a service
// - exec in the client pod and validate the following
// - can reach internet
// - can reach server pod
// - can resolve dns queries to internet
// - can resolve dns queries to services
// - cannot reach services and pods in other namespaces
// - Create Tenant with specific cpu/ram/storage requests / limits and make sure they are enforced by trying to
// deploy a pod with lower requests/limits (accepted) and higher requests/limits (rejected)
// - Create TenantCredentials and make sure they give only access to the correct tenant

View File

@@ -6,6 +6,8 @@ readme.workspace = true
license.workspace = true
[dependencies]
rand = "0.9"
hex = "0.4"
libredfish = "0.1.1"
reqwest = { version = "0.11", features = ["blocking", "json"] }
russh = "0.45.0"
@@ -40,6 +42,7 @@ dockerfile_builder = "0.1.5"
temp-file = "0.1.9"
convert_case.workspace = true
email_address = "0.2.9"
chrono.workspace = true
fqdn = { version = "0.4.6", features = [
"domain-label-cannot-start-or-end-with-hyphen",
"domain-label-length-limited-to-63",
@@ -49,5 +52,4 @@ fqdn = { version = "0.4.6", features = [
"serde",
] }
temp-dir = "0.1.14"
typetag = "0.2.20"
dyn-clone = "1.0.19"

View File

@@ -1,6 +1,24 @@
use rand::distr::Alphanumeric;
use rand::distr::SampleString;
use std::time::SystemTime;
use std::time::UNIX_EPOCH;
use serde::{Deserialize, Serialize};
#[derive(Debug, Clone, Serialize, Deserialize)]
/// A unique identifier designed for ease of use.
///
/// You can pass it any String to use and Id, or you can use the default format with `Id::default()`
///
/// The default format looks like this
///
/// `462d4c_g2COgai`
///
/// The first part is the unix timesamp in hexadecimal which makes Id easily sorted by creation time.
/// Second part is a serie of 7 random characters.
///
/// **It is not meant to be very secure or unique**, it is suitable to generate up to 10 000 items per
/// second with a reasonable collision rate of 0,000014 % as calculated by this calculator : https://kevingal.com/apps/collision.html
#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
pub struct Id {
value: String,
}
@@ -9,4 +27,31 @@ impl Id {
pub fn from_string(value: String) -> Self {
Self { value }
}
pub fn from_str(value: &str) -> Self {
Self::from_string(value.to_string())
}
}
impl std::fmt::Display for Id {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.write_str(&self.value)
}
}
impl Default for Id {
fn default() -> Self {
let start = SystemTime::now();
let since_the_epoch = start
.duration_since(UNIX_EPOCH)
.expect("Time went backwards");
let timestamp = since_the_epoch.as_secs();
let hex_timestamp = format!("{:x}", timestamp & 0xffffff);
let random_part: String = Alphanumeric.sample_string(&mut rand::rng(), 7);
let value = format!("{}_{}", hex_timestamp, random_part);
Self { value }
}
}

View File

@@ -20,6 +20,7 @@ pub enum InterpretName {
Panic,
OPNSense,
K3dInstallation,
TenantInterpret,
}
impl std::fmt::Display for InterpretName {
@@ -35,6 +36,7 @@ impl std::fmt::Display for InterpretName {
InterpretName::Panic => f.write_str("Panic"),
InterpretName::OPNSense => f.write_str("OPNSense"),
InterpretName::K3dInstallation => f.write_str("K3dInstallation"),
InterpretName::TenantInterpret => f.write_str("Tenant"),
}
}
}

View File

@@ -0,0 +1,12 @@
use async_trait::async_trait;
use crate::{interpret::InterpretError, inventory::Inventory};
#[async_trait]
pub trait Installable<T>: Send + Sync {
async fn ensure_installed(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<(), InterpretError>;
}

View File

@@ -1,11 +1,11 @@
use derive_new::new;
use k8s_openapi::NamespaceResourceScope;
use k8s_openapi::{ClusterResourceScope, NamespaceResourceScope};
use kube::{
Api, Client, Config, Error, Resource,
api::PostParams,
api::{Patch, PatchParams},
config::{KubeConfigOptions, Kubeconfig},
};
use log::error;
use log::{debug, error, trace};
use serde::de::DeserializeOwned;
#[derive(new)]
@@ -20,52 +20,50 @@ impl K8sClient {
})
}
pub async fn apply_all<
K: Resource<Scope = NamespaceResourceScope>
+ std::fmt::Debug
+ Sync
+ DeserializeOwned
+ Default
+ serde::Serialize
+ Clone,
>(
&self,
resource: &Vec<K>,
) -> Result<Vec<K>, kube::Error>
/// Apply a resource in namespace
///
/// See `kubectl apply` for more information on the expected behavior of this function
pub async fn apply<K>(&self, resource: &K, namespace: Option<&str>) -> Result<K, Error>
where
K: Resource + Clone + std::fmt::Debug + DeserializeOwned + serde::Serialize,
<K as Resource>::Scope: ApplyStrategy<K>,
<K as kube::Resource>::DynamicType: Default,
{
let mut result = vec![];
for r in resource.iter() {
let api: Api<K> = Api::all(self.client.clone());
result.push(api.create(&PostParams::default(), &r).await?);
}
Ok(result)
debug!(
"Applying resource {:?} with ns {:?}",
resource.meta().name,
namespace
);
trace!(
"{:#}",
serde_json::to_value(resource).unwrap_or(serde_json::Value::Null)
);
let api: Api<K> =
<<K as Resource>::Scope as ApplyStrategy<K>>::get_api(&self.client, namespace);
// api.create(&PostParams::default(), &resource).await
let patch_params = PatchParams::apply("harmony");
let name = resource
.meta()
.name
.as_ref()
.expect("K8s Resource should have a name");
api.patch(name, &patch_params, &Patch::Apply(resource))
.await
}
pub async fn apply_namespaced<K>(
&self,
resource: &Vec<K>,
ns: Option<&str>,
) -> Result<Vec<K>, Error>
pub async fn apply_many<K>(&self, resource: &Vec<K>, ns: Option<&str>) -> Result<Vec<K>, Error>
where
K: Resource<Scope = NamespaceResourceScope>
+ Clone
+ std::fmt::Debug
+ DeserializeOwned
+ serde::Serialize
+ Default,
K: Resource + Clone + std::fmt::Debug + DeserializeOwned + serde::Serialize,
<K as Resource>::Scope: ApplyStrategy<K>,
<K as kube::Resource>::DynamicType: Default,
{
let mut resources = Vec::new();
let mut result = Vec::new();
for r in resource.iter() {
let api: Api<K> = match ns {
Some(ns) => Api::namespaced(self.client.clone(), ns),
None => Api::default_namespaced(self.client.clone()),
};
resources.push(api.create(&PostParams::default(), &r).await?);
result.push(self.apply(r, ns).await?);
}
Ok(resources)
Ok(result)
}
pub(crate) async fn from_kubeconfig(path: &str) -> Option<K8sClient> {
@@ -86,3 +84,35 @@ impl K8sClient {
))
}
}
pub trait ApplyStrategy<K: Resource> {
fn get_api(client: &Client, ns: Option<&str>) -> Api<K>;
}
/// Implementation for all resources that are cluster-scoped.
/// It will always use `Api::all` and ignore the namespace parameter.
impl<K> ApplyStrategy<K> for ClusterResourceScope
where
K: Resource<Scope = ClusterResourceScope>,
<K as kube::Resource>::DynamicType: Default,
{
fn get_api(client: &Client, _ns: Option<&str>) -> Api<K> {
Api::all(client.clone())
}
}
/// Implementation for all resources that are namespace-scoped.
/// It will use `Api::namespaced` if a namespace is provided, otherwise
/// it falls back to the default namespace configured in your kubeconfig.
impl<K> ApplyStrategy<K> for NamespaceResourceScope
where
K: Resource<Scope = NamespaceResourceScope>,
<K as kube::Resource>::DynamicType: Default,
{
fn get_api(client: &Client, ns: Option<&str>) -> Api<K> {
match ns {
Some(ns) => Api::namespaced(client.clone(), ns),
None => Api::default_namespaced(client.clone()),
}
}
}

View File

@@ -2,10 +2,11 @@ use std::{process::Command, sync::Arc};
use async_trait::async_trait;
use inquire::Confirm;
use log::{info, warn};
use log::{debug, info, warn};
use tokio::sync::OnceCell;
use crate::{
executors::ExecutorError,
interpret::{InterpretError, Outcome},
inventory::Inventory,
maestro::Maestro,
@@ -13,14 +14,19 @@ use crate::{
topology::LocalhostTopology,
};
use super::{HelmCommand, K8sclient, Topology, k8s::K8sClient};
use super::{
HelmCommand, K8sclient, Topology,
k8s::K8sClient,
tenant::{TenantConfig, TenantManager, k8s::K8sTenantManager},
};
struct K8sState {
client: Arc<K8sClient>,
source: K8sSource,
_source: K8sSource,
message: String,
}
#[derive(Debug)]
enum K8sSource {
LocalK3d,
Kubeconfig,
@@ -28,6 +34,8 @@ enum K8sSource {
pub struct K8sAnywhereTopology {
k8s_state: OnceCell<Option<K8sState>>,
tenant_manager: OnceCell<K8sTenantManager>,
config: K8sAnywhereConfig,
}
#[async_trait]
@@ -48,9 +56,19 @@ impl K8sclient for K8sAnywhereTopology {
}
impl K8sAnywhereTopology {
pub fn new() -> Self {
pub fn from_env() -> Self {
Self {
k8s_state: OnceCell::new(),
tenant_manager: OnceCell::new(),
config: K8sAnywhereConfig::from_env(),
}
}
pub fn with_config(config: K8sAnywhereConfig) -> Self {
Self {
k8s_state: OnceCell::new(),
tenant_manager: OnceCell::new(),
config,
}
}
@@ -91,27 +109,15 @@ impl K8sAnywhereTopology {
}
async fn try_get_or_install_k8s_client(&self) -> Result<Option<K8sState>, InterpretError> {
let k8s_anywhere_config = K8sAnywhereConfig {
kubeconfig: std::env::var("KUBECONFIG").ok().map(|v| v.to_string()),
use_system_kubeconfig: std::env::var("HARMONY_USE_SYSTEM_KUBECONFIG")
.map_or_else(|_| false, |v| v.parse().ok().unwrap_or(false)),
autoinstall: std::env::var("HARMONY_AUTOINSTALL")
.map_or_else(|_| false, |v| v.parse().ok().unwrap_or(false)),
};
let k8s_anywhere_config = &self.config;
if k8s_anywhere_config.use_system_kubeconfig {
match self.try_load_system_kubeconfig().await {
Some(_client) => todo!(),
None => todo!(),
}
}
if let Some(kubeconfig) = k8s_anywhere_config.kubeconfig {
if let Some(kubeconfig) = &k8s_anywhere_config.kubeconfig {
debug!("Loading kubeconfig {kubeconfig}");
match self.try_load_kubeconfig(&kubeconfig).await {
Some(client) => {
return Ok(Some(K8sState {
client: Arc::new(client),
source: K8sSource::Kubeconfig,
_source: K8sSource::Kubeconfig,
message: format!("Loaded k8s client from kubeconfig {kubeconfig}"),
}));
}
@@ -123,6 +129,14 @@ impl K8sAnywhereTopology {
}
}
if k8s_anywhere_config.use_system_kubeconfig {
debug!("Loading system kubeconfig");
match self.try_load_system_kubeconfig().await {
Some(_client) => todo!(),
None => todo!(),
}
}
info!("No kubernetes configuration found");
if !k8s_anywhere_config.autoinstall {
@@ -151,7 +165,7 @@ impl K8sAnywhereTopology {
let state = match k3d.get_client().await {
Ok(client) => K8sState {
client: Arc::new(K8sClient::new(client)),
source: K8sSource::LocalK3d,
_source: K8sSource::LocalK3d,
message: "Successfully installed K3D cluster and acquired client".to_string(),
},
Err(_) => todo!(),
@@ -159,27 +173,64 @@ impl K8sAnywhereTopology {
Ok(Some(state))
}
async fn ensure_k8s_tenant_manager(&self) -> Result<(), String> {
if let Some(_) = self.tenant_manager.get() {
return Ok(());
}
self.tenant_manager
.get_or_try_init(async || -> Result<K8sTenantManager, String> {
let k8s_client = self.k8s_client().await?;
Ok(K8sTenantManager::new(k8s_client))
})
.await
.unwrap();
Ok(())
}
fn get_k8s_tenant_manager(&self) -> Result<&K8sTenantManager, ExecutorError> {
match self.tenant_manager.get() {
Some(t) => Ok(t),
None => Err(ExecutorError::UnexpectedError(
"K8sTenantManager not available".to_string(),
)),
}
}
}
struct K8sAnywhereConfig {
pub struct K8sAnywhereConfig {
/// The path of the KUBECONFIG file that Harmony should use to interact with the Kubernetes
/// cluster
///
/// Default : None
kubeconfig: Option<String>,
pub kubeconfig: Option<String>,
/// Whether to use the system KUBECONFIG, either the environment variable or the file in the
/// default or configured location
///
/// Default : false
use_system_kubeconfig: bool,
pub use_system_kubeconfig: bool,
/// Whether to install automatically a kubernetes cluster
///
/// When enabled, autoinstall will setup a K3D cluster on the localhost. https://k3d.io/stable/
///
/// Default: true
autoinstall: bool,
pub autoinstall: bool,
}
impl K8sAnywhereConfig {
fn from_env() -> Self {
Self {
kubeconfig: std::env::var("KUBECONFIG").ok().map(|v| v.to_string()),
use_system_kubeconfig: std::env::var("HARMONY_USE_SYSTEM_KUBECONFIG")
.map_or_else(|_| false, |v| v.parse().ok().unwrap_or(false)),
autoinstall: std::env::var("HARMONY_AUTOINSTALL")
.map_or_else(|_| false, |v| v.parse().ok().unwrap_or(false)),
}
}
}
#[async_trait]
@@ -198,6 +249,10 @@ impl Topology for K8sAnywhereTopology {
"No K8s client could be found or installed".to_string(),
))?;
self.ensure_k8s_tenant_manager()
.await
.map_err(|e| InterpretError::new(e))?;
match self.is_helm_available() {
Ok(()) => Ok(Outcome::success(format!(
"{} + helm available",
@@ -209,3 +264,12 @@ impl Topology for K8sAnywhereTopology {
}
impl HelmCommand for K8sAnywhereTopology {}
#[async_trait]
impl TenantManager for K8sAnywhereTopology {
async fn provision_tenant(&self, config: &TenantConfig) -> Result<(), ExecutorError> {
self.get_k8s_tenant_manager()?
.provision_tenant(config)
.await
}
}

View File

@@ -7,6 +7,12 @@ use serde::Serialize;
use super::{IpAddress, LogicalHost};
use crate::executors::ExecutorError;
impl std::fmt::Debug for dyn LoadBalancer {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.write_fmt(format_args!("LoadBalancer {}", self.get_ip()))
}
}
#[async_trait]
pub trait LoadBalancer: Send + Sync {
fn get_ip(&self) -> IpAddress;
@@ -32,11 +38,6 @@ pub trait LoadBalancer: Send + Sync {
}
}
impl std::fmt::Debug for dyn LoadBalancer {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.write_fmt(format_args!("LoadBalancer {}", self.get_ip()))
}
}
#[derive(Debug, PartialEq, Clone, Serialize)]
pub struct LoadBalancerService {
pub backend_servers: Vec<BackendServer>,

View File

@@ -1,8 +1,11 @@
mod ha_cluster;
mod host_binding;
mod http;
pub mod installable;
mod k8s_anywhere;
mod localhost;
pub mod oberservability;
pub mod tenant;
pub use k8s_anywhere::*;
pub use localhost::*;
pub mod k8s;

View File

@@ -0,0 +1 @@
pub mod monitoring;

View File

@@ -0,0 +1,76 @@
use async_trait::async_trait;
use log::debug;
use crate::{
data::{Id, Version},
interpret::{Interpret, InterpretError, InterpretName, InterpretStatus, Outcome},
inventory::Inventory,
topology::{Topology, installable::Installable},
};
#[async_trait]
pub trait AlertSender: Send + Sync + std::fmt::Debug {
fn name(&self) -> String;
}
#[derive(Debug)]
pub struct AlertingInterpret<S: AlertSender> {
pub sender: S,
pub receivers: Vec<Box<dyn AlertReceiver<S>>>,
pub rules: Vec<Box<dyn AlertRule<S>>>,
}
#[async_trait]
impl<S: AlertSender + Installable<T>, T: Topology> Interpret<T> for AlertingInterpret<S> {
async fn execute(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
for receiver in self.receivers.iter() {
receiver.install(&self.sender).await?;
}
for rule in self.rules.iter() {
debug!("installing rule: {:#?}", rule);
rule.install(&self.sender).await?;
}
self.sender.ensure_installed(inventory, topology).await?;
Ok(Outcome::success(format!(
"successfully installed alert sender {}",
self.sender.name()
)))
}
fn get_name(&self) -> InterpretName {
todo!()
}
fn get_version(&self) -> Version {
todo!()
}
fn get_status(&self) -> InterpretStatus {
todo!()
}
fn get_children(&self) -> Vec<Id> {
todo!()
}
}
#[async_trait]
pub trait AlertReceiver<S: AlertSender>: std::fmt::Debug + Send + Sync {
async fn install(&self, sender: &S) -> Result<Outcome, InterpretError>;
fn clone_box(&self) -> Box<dyn AlertReceiver<S>>;
}
#[async_trait]
pub trait AlertRule<S: AlertSender>: std::fmt::Debug + Send + Sync {
async fn install(&self, sender: &S) -> Result<Outcome, InterpretError>;
fn clone_box(&self) -> Box<dyn AlertRule<S>>;
}
#[async_trait]
pub trait ScrapeTarger<S: AlertSender> {
async fn install(&self, sender: &S) -> Result<(), InterpretError>;
}

View File

@@ -0,0 +1,327 @@
use std::sync::Arc;
use crate::{
executors::ExecutorError,
topology::k8s::{ApplyStrategy, K8sClient},
};
use async_trait::async_trait;
use derive_new::new;
use k8s_openapi::{
api::{
core::v1::{Namespace, ResourceQuota},
networking::v1::{
NetworkPolicy, NetworkPolicyEgressRule, NetworkPolicyIngressRule, NetworkPolicyPort,
},
},
apimachinery::pkg::util::intstr::IntOrString,
};
use kube::Resource;
use log::{debug, info, warn};
use serde::de::DeserializeOwned;
use serde_json::json;
use super::{TenantConfig, TenantManager};
#[derive(new)]
pub struct K8sTenantManager {
k8s_client: Arc<K8sClient>,
}
impl K8sTenantManager {
fn get_namespace_name(&self, config: &TenantConfig) -> String {
config.name.clone()
}
fn ensure_constraints(&self, _namespace: &Namespace) -> Result<(), ExecutorError> {
warn!("Validate that when tenant already exists (by id) that name has not changed");
warn!("Make sure other Tenant constraints are respected by this k8s implementation");
Ok(())
}
async fn apply_resource<
K: Resource + std::fmt::Debug + Sync + DeserializeOwned + Default + serde::Serialize + Clone,
>(
&self,
mut resource: K,
config: &TenantConfig,
) -> Result<K, ExecutorError>
where
<K as kube::Resource>::DynamicType: Default,
<K as kube::Resource>::Scope: ApplyStrategy<K>,
{
self.apply_labels(&mut resource, config);
self.k8s_client
.apply(&resource, Some(&self.get_namespace_name(config)))
.await
.map_err(|e| {
ExecutorError::UnexpectedError(format!("Could not create Tenant resource : {e}"))
})
}
fn apply_labels<K: Resource>(&self, resource: &mut K, config: &TenantConfig) {
let labels = resource.meta_mut().labels.get_or_insert_default();
labels.insert(
"app.kubernetes.io/managed-by".to_string(),
"harmony".to_string(),
);
labels.insert("harmony/tenant-id".to_string(), config.id.to_string());
labels.insert("harmony/tenant-name".to_string(), config.name.clone());
}
fn build_namespace(&self, config: &TenantConfig) -> Result<Namespace, ExecutorError> {
let namespace = json!(
{
"apiVersion": "v1",
"kind": "Namespace",
"metadata": {
"labels": {
"harmony.nationtech.io/tenant.id": config.id.to_string(),
"harmony.nationtech.io/tenant.name": config.name,
},
"name": self.get_namespace_name(config),
},
}
);
serde_json::from_value(namespace).map_err(|e| {
ExecutorError::ConfigurationError(format!(
"Could not build TenantManager Namespace. {}",
e
))
})
}
fn build_resource_quota(&self, config: &TenantConfig) -> Result<ResourceQuota, ExecutorError> {
let resource_quota = json!(
{
"apiVersion": "v1",
"kind": "ResourceQuota",
"metadata": {
"name": format!("{}-quota", config.name),
"labels": {
"harmony.nationtech.io/tenant.id": config.id.to_string(),
"harmony.nationtech.io/tenant.name": config.name,
},
"namespace": self.get_namespace_name(config),
},
"spec": {
"hard": {
"limits.cpu": format!("{:.0}",config.resource_limits.cpu_limit_cores),
"limits.memory": format!("{:.3}Gi", config.resource_limits.memory_limit_gb),
"requests.cpu": format!("{:.0}",config.resource_limits.cpu_request_cores),
"requests.memory": format!("{:.3}Gi", config.resource_limits.memory_request_gb),
"requests.storage": format!("{:.3}Gi", config.resource_limits.storage_total_gb),
"pods": "20",
"services": "10",
"configmaps": "30",
"secrets": "30",
"persistentvolumeclaims": "15",
"services.loadbalancers": "2",
"services.nodeports": "5",
"limits.ephemeral-storage": "10Gi",
}
}
}
);
serde_json::from_value(resource_quota).map_err(|e| {
ExecutorError::ConfigurationError(format!(
"Could not build TenantManager ResourceQuota. {}",
e
))
})
}
fn build_network_policy(&self, config: &TenantConfig) -> Result<NetworkPolicy, ExecutorError> {
let network_policy = json!({
"apiVersion": "networking.k8s.io/v1",
"kind": "NetworkPolicy",
"metadata": {
"name": format!("{}-network-policy", config.name),
},
"spec": {
"podSelector": {},
"egress": [
{ "to": [ {"podSelector": {}}]},
{ "to":
[
{
"podSelector": {},
"namespaceSelector": {
"matchLabels": {
"kubernetes.io/metadata.name":"openshift-dns"
}
}
},
]
},
{ "to": [
{
"ipBlock": {
"cidr": "0.0.0.0/0",
// See https://en.wikipedia.org/wiki/Reserved_IP_addresses
"except": [
"10.0.0.0/8",
"172.16.0.0/12",
"192.168.0.0/16",
"192.0.0.0/24",
"192.0.2.0/24",
"192.88.99.0/24",
"192.18.0.0/15",
"198.51.100.0/24",
"169.254.0.0/16",
"203.0.113.0/24",
"127.0.0.0/8",
// Not sure we should block this one as it is
// used for multicast. But better block more than less.
"224.0.0.0/4",
"240.0.0.0/4",
"100.64.0.0/10",
"233.252.0.0/24",
"0.0.0.0/8",
],
}
}
]
},
],
"ingress": [
{ "from": [ {"podSelector": {}}]}
],
"policyTypes": [
"Ingress", "Egress",
]
}
});
let mut network_policy: NetworkPolicy =
serde_json::from_value(network_policy).map_err(|e| {
ExecutorError::ConfigurationError(format!(
"Could not build TenantManager NetworkPolicy. {}",
e
))
})?;
config
.network_policy
.additional_allowed_cidr_ingress
.iter()
.try_for_each(|c| -> Result<(), ExecutorError> {
let cidr_list: Vec<serde_json::Value> =
c.0.iter()
.map(|ci| {
json!({
"ipBlock": {
"cidr": ci.to_string(),
}
})
})
.collect();
let rule = serde_json::from_value::<NetworkPolicyIngressRule>(json!({
"from": cidr_list
}))
.map_err(|e| {
ExecutorError::ConfigurationError(format!(
"Could not build TenantManager NetworkPolicyIngressRule. {}",
e
))
})?;
network_policy
.spec
.as_mut()
.unwrap()
.ingress
.as_mut()
.unwrap()
.push(rule);
Ok(())
})?;
config
.network_policy
.additional_allowed_cidr_egress
.iter()
.try_for_each(|c| -> Result<(), ExecutorError> {
let cidr_list: Vec<serde_json::Value> =
c.0.iter()
.map(|ci| {
json!({
"ipBlock": {
"cidr": ci.to_string(),
}
})
})
.collect();
let ports: Option<Vec<NetworkPolicyPort>> =
c.1.as_ref().map(|spec| match &spec.data {
super::PortSpecData::SinglePort(port) => vec![NetworkPolicyPort {
port: Some(IntOrString::Int(port.clone().into())),
..Default::default()
}],
super::PortSpecData::PortRange(start, end) => vec![NetworkPolicyPort {
port: Some(IntOrString::Int(start.clone().into())),
end_port: Some(end.clone().into()),
protocol: None, // Not currently supported by Harmony
}],
super::PortSpecData::ListOfPorts(items) => items
.iter()
.map(|i| NetworkPolicyPort {
port: Some(IntOrString::Int(i.clone().into())),
..Default::default()
})
.collect(),
});
let rule = serde_json::from_value::<NetworkPolicyEgressRule>(json!({
"to": cidr_list,
"ports": ports,
}))
.map_err(|e| {
ExecutorError::ConfigurationError(format!(
"Could not build TenantManager NetworkPolicyEgressRule. {}",
e
))
})?;
network_policy
.spec
.as_mut()
.unwrap()
.egress
.as_mut()
.unwrap()
.push(rule);
Ok(())
})?;
Ok(network_policy)
}
}
#[async_trait]
impl TenantManager for K8sTenantManager {
async fn provision_tenant(&self, config: &TenantConfig) -> Result<(), ExecutorError> {
let namespace = self.build_namespace(config)?;
let resource_quota = self.build_resource_quota(config)?;
let network_policy = self.build_network_policy(config)?;
self.ensure_constraints(&namespace)?;
debug!("Creating namespace for tenant {}", config.name);
self.apply_resource(namespace, config).await?;
debug!("Creating resource_quota for tenant {}", config.name);
self.apply_resource(resource_quota, config).await?;
debug!("Creating network_policy for tenant {}", config.name);
self.apply_resource(network_policy, config).await?;
info!(
"Success provisionning K8s tenant id {} name {}",
config.id, config.name
);
Ok(())
}
}

View File

@@ -0,0 +1,18 @@
use super::*;
use async_trait::async_trait;
use crate::executors::ExecutorError;
#[async_trait]
pub trait TenantManager {
/// Creates or update tenant based on the provided configuration.
/// This operation should be idempotent; if a tenant with the same `config.id`
/// already exists and matches the config, it will succeed without changes.
///
/// If it exists but differs, it will be updated, or return an error if the update
/// action is not supported
///
/// # Arguments
/// * `config`: The desired configuration for the new tenant.
async fn provision_tenant(&self, config: &TenantConfig) -> Result<(), ExecutorError>;
}

View File

@@ -0,0 +1,224 @@
pub mod k8s;
mod manager;
use std::str::FromStr;
pub use manager::*;
use serde::{Deserialize, Serialize};
use crate::data::Id;
#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)] // Assuming serde for Scores
pub struct TenantConfig {
/// This will be used as the primary unique identifier for management operations and will never
/// change for the entire lifetime of the tenant
pub id: Id,
/// A human-readable name for the tenant (e.g., "client-alpha", "project-phoenix").
pub name: String,
/// Desired resource allocations and limits for the tenant.
pub resource_limits: ResourceLimits,
/// High-level network isolation policies for the tenant.
pub network_policy: TenantNetworkPolicy,
}
impl Default for TenantConfig {
fn default() -> Self {
let id = Id::default();
Self {
name: format!("tenant_{id}"),
id,
resource_limits: ResourceLimits::default(),
network_policy: TenantNetworkPolicy {
default_inter_tenant_ingress: InterTenantIngressPolicy::DenyAll,
default_internet_egress: InternetEgressPolicy::AllowAll,
additional_allowed_cidr_ingress: vec![],
additional_allowed_cidr_egress: vec![],
},
}
}
}
#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
pub struct ResourceLimits {
/// Requested/guaranteed CPU cores (e.g., 2.0).
pub cpu_request_cores: f32,
/// Maximum CPU cores the tenant can burst to (e.g., 4.0).
pub cpu_limit_cores: f32,
/// Requested/guaranteed memory in Gigabytes (e.g., 8.0).
pub memory_request_gb: f32,
/// Maximum memory in Gigabytes tenant can burst to (e.g., 16.0).
pub memory_limit_gb: f32,
/// Total persistent storage allocation in Gigabytes across all volumes.
pub storage_total_gb: f32,
}
impl Default for ResourceLimits {
fn default() -> Self {
Self {
cpu_request_cores: 4.0,
cpu_limit_cores: 4.0,
memory_request_gb: 4.0,
memory_limit_gb: 4.0,
storage_total_gb: 20.0,
}
}
}
#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
pub struct TenantNetworkPolicy {
/// Policy for ingress traffic originating from other tenants within the same Harmony-managed environment.
pub default_inter_tenant_ingress: InterTenantIngressPolicy,
/// Policy for egress traffic destined for the public internet.
pub default_internet_egress: InternetEgressPolicy,
pub additional_allowed_cidr_ingress: Vec<(Vec<cidr::Ipv4Cidr>, Option<PortSpec>)>,
pub additional_allowed_cidr_egress: Vec<(Vec<cidr::Ipv4Cidr>, Option<PortSpec>)>,
}
impl Default for TenantNetworkPolicy {
fn default() -> Self {
TenantNetworkPolicy {
default_inter_tenant_ingress: InterTenantIngressPolicy::DenyAll,
default_internet_egress: InternetEgressPolicy::DenyAll,
additional_allowed_cidr_ingress: vec![],
additional_allowed_cidr_egress: vec![],
}
}
}
#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
pub enum InterTenantIngressPolicy {
/// Deny all traffic from other tenants by default.
DenyAll,
}
#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
pub enum InternetEgressPolicy {
/// Allow all outbound traffic to the internet.
AllowAll,
/// Deny all outbound traffic to the internet by default.
DenyAll,
}
/// Represents a port specification that can be either a single port, a comma-separated list of ports,
/// or a range separated by a dash.
#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
pub struct PortSpec {
/// The actual representation of the ports as strings for serialization/deserialization purposes.
pub data: PortSpecData,
}
impl PortSpec {
/// TODO write short rust doc that shows what types of input are supported
fn parse_from_str(spec: &str) -> Result<PortSpec, String> {
// Check for single port
if let Ok(port) = spec.parse::<u16>() {
let spec = PortSpecData::SinglePort(port);
return Ok(Self { data: spec });
}
if let Some(range) = spec.find('-') {
let start_str = &spec[..range];
let end_str = &spec[(range + 1)..];
if let (Ok(start), Ok(end)) = (start_str.parse::<u16>(), end_str.parse::<u16>()) {
let spec = PortSpecData::PortRange(start, end);
return Ok(Self { data: spec });
}
}
let ports: Vec<&str> = spec.split(',').collect();
if !ports.is_empty() && ports.iter().all(|p| p.parse::<u16>().is_ok()) {
let maybe_ports = ports.iter().try_fold(vec![], |mut list, &p| {
if let Ok(p) = p.parse::<u16>() {
list.push(p);
return Ok(list);
}
Err(())
});
if let Ok(ports) = maybe_ports {
let spec = PortSpecData::ListOfPorts(ports);
return Ok(Self { data: spec });
}
}
Err(format!("Invalid port spec format {spec}"))
}
}
impl FromStr for PortSpec {
type Err = String;
fn from_str(s: &str) -> Result<Self, Self::Err> {
Self::parse_from_str(s)
}
}
#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
pub enum PortSpecData {
SinglePort(u16),
PortRange(u16, u16),
ListOfPorts(Vec<u16>),
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_single_port() {
let port_spec = "2144".parse::<PortSpec>().unwrap();
match port_spec.data {
PortSpecData::SinglePort(port) => assert_eq!(port, 2144),
_ => panic!("Expected SinglePort"),
}
}
#[test]
fn test_port_range() {
let port_spec = "80-90".parse::<PortSpec>().unwrap();
match port_spec.data {
PortSpecData::PortRange(start, end) => {
assert_eq!(start, 80);
assert_eq!(end, 90);
}
_ => panic!("Expected PortRange"),
}
}
#[test]
fn test_list_of_ports() {
let port_spec = "2144,3424".parse::<PortSpec>().unwrap();
match port_spec.data {
PortSpecData::ListOfPorts(ports) => {
assert_eq!(ports[0], 2144);
assert_eq!(ports[1], 3424);
}
_ => panic!("Expected ListOfPorts"),
}
}
#[test]
fn test_invalid_port_spec() {
let result = "invalid".parse::<PortSpec>();
assert!(result.is_err());
}
#[test]
fn test_empty_input() {
let result = "".parse::<PortSpec>();
assert!(result.is_err());
}
#[test]
fn test_only_coma() {
let result = ",".parse::<PortSpec>();
assert!(result.is_err());
}
}

View File

@@ -23,7 +23,7 @@ pub struct HelmRepository {
force_update: bool,
}
impl HelmRepository {
pub(crate) fn new(name: String, url: Url, force_update: bool) -> Self {
pub fn new(name: String, url: Url, force_update: bool) -> Self {
Self {
name,
url,
@@ -104,7 +104,10 @@ impl HelmChartInterpret {
fn run_helm_command(args: &[&str]) -> Result<Output, InterpretError> {
let command_str = format!("helm {}", args.join(" "));
debug!("Got KUBECONFIG: `{}`", std::env::var("KUBECONFIG").unwrap());
debug!(
"Got KUBECONFIG: `{}`",
std::env::var("KUBECONFIG").unwrap_or("".to_string())
);
debug!("Running Helm command: `{}`", command_str);
let output = Command::new("helm")

View File

@@ -1,12 +1,9 @@
use async_trait::async_trait;
use log::debug;
use non_blank_string_rs::NonBlankString;
use serde::Serialize;
use std::collections::HashMap;
use std::env::temp_dir;
use std::ffi::OsStr;
use std::io::ErrorKind;
use std::path::{Path, PathBuf};
use std::path::PathBuf;
use std::process::{Command, Output};
use temp_dir::{self, TempDir};
use temp_file::TempFile;
@@ -314,7 +311,7 @@ impl<T: Topology + K8sclient + HelmCommand> Interpret<T> for HelmChartInterpretV
_inventory: &Inventory,
_topology: &T,
) -> Result<Outcome, InterpretError> {
let ns = self
let _ns = self
.score
.chart
.namespace
@@ -342,7 +339,7 @@ impl<T: Topology + K8sclient + HelmCommand> Interpret<T> for HelmChartInterpretV
let res = helm_executor.generate();
let output = match res {
let _output = match res {
Ok(output) => output,
Err(err) => return Err(InterpretError::new(err.to_string())),
};

View File

@@ -1,5 +1,6 @@
use harmony_macros::ingress_path;
use k8s_openapi::api::networking::v1::Ingress;
use log::{debug, trace};
use serde::Serialize;
use serde_json::json;
@@ -56,22 +57,24 @@ impl<T: Topology + K8sclient> Score<T> for K8sIngressScore {
let ingress = json!(
{
"metadata": {
"name": self.name
"name": self.name.to_string(),
},
"spec": {
"rules": [
{ "host": self.host,
{ "host": self.host.to_string(),
"http": {
"paths": [
{
"path": path,
"pathType": path_type.as_str(),
"backend": [
{
"service": self.backend_service,
"port": self.port
"backend": {
"service": {
"name": self.backend_service.to_string(),
"port": {
"number": self.port,
}
}
]
}
}
]
}
@@ -81,13 +84,16 @@ impl<T: Topology + K8sclient> Score<T> for K8sIngressScore {
}
);
trace!("Building ingresss object from Value {ingress:#}");
let ingress: Ingress = serde_json::from_value(ingress).unwrap();
debug!(
"Successfully built Ingress for host {:?}",
ingress.metadata.name
);
Box::new(K8sResourceInterpret {
score: K8sResourceScore::single(
ingress.clone(),
self.namespace
.clone()
.map(|f| f.as_c_str().to_str().unwrap().to_string()),
self.namespace.clone().map(|f| f.to_string()),
),
})
}

View File

@@ -1,6 +1,7 @@
use async_trait::async_trait;
use k8s_openapi::NamespaceResourceScope;
use kube::Resource;
use log::info;
use serde::{Serialize, de::DeserializeOwned};
use crate::{
@@ -75,11 +76,12 @@ where
_inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
info!("Applying {} resources", self.score.resource.len());
topology
.k8s_client()
.await
.expect("Environment should provide enough information to instanciate a client")
.apply_namespaced(&self.score.resource, self.score.namespace.as_deref())
.apply_many(&self.score.resource, self.score.namespace.as_deref())
.await?;
Ok(Outcome::success(

View File

@@ -12,4 +12,6 @@ pub mod load_balancer;
pub mod monitoring;
pub mod okd;
pub mod opnsense;
pub mod prometheus;
pub mod tenant;
pub mod tftp;

View File

@@ -0,0 +1,125 @@
use async_trait::async_trait;
use serde::Serialize;
use serde_yaml::{Mapping, Value};
use crate::{
interpret::{InterpretError, Outcome},
modules::monitoring::kube_prometheus::{
prometheus::{Prometheus, PrometheusReceiver},
types::{AlertChannelConfig, AlertManagerChannelConfig},
},
topology::{Url, oberservability::monitoring::AlertReceiver},
};
#[derive(Debug, Clone, Serialize)]
pub struct DiscordWebhook {
pub name: String,
pub url: Url,
}
#[async_trait]
impl AlertReceiver<Prometheus> for DiscordWebhook {
async fn install(&self, sender: &Prometheus) -> Result<Outcome, InterpretError> {
sender.install_receiver(self).await
}
fn clone_box(&self) -> Box<dyn AlertReceiver<Prometheus>> {
Box::new(self.clone())
}
}
#[async_trait]
impl PrometheusReceiver for DiscordWebhook {
fn name(&self) -> String {
self.name.clone()
}
async fn configure_receiver(&self) -> AlertManagerChannelConfig {
self.get_config().await
}
}
#[async_trait]
impl AlertChannelConfig for DiscordWebhook {
async fn get_config(&self) -> AlertManagerChannelConfig {
let channel_global_config = None;
let channel_receiver = self.alert_channel_receiver().await;
let channel_route = self.alert_channel_route().await;
AlertManagerChannelConfig {
channel_global_config,
channel_receiver,
channel_route,
}
}
}
impl DiscordWebhook {
async fn alert_channel_route(&self) -> serde_yaml::Value {
let mut route = Mapping::new();
route.insert(
Value::String("receiver".to_string()),
Value::String(self.name.clone()),
);
route.insert(
Value::String("matchers".to_string()),
Value::Sequence(vec![Value::String("alertname!=Watchdog".to_string())]),
);
route.insert(Value::String("continue".to_string()), Value::Bool(true));
Value::Mapping(route)
}
async fn alert_channel_receiver(&self) -> serde_yaml::Value {
let mut receiver = Mapping::new();
receiver.insert(
Value::String("name".to_string()),
Value::String(self.name.clone()),
);
let mut discord_config = Mapping::new();
discord_config.insert(
Value::String("webhook_url".to_string()),
Value::String(self.url.to_string()),
);
receiver.insert(
Value::String("discord_configs".to_string()),
Value::Sequence(vec![Value::Mapping(discord_config)]),
);
Value::Mapping(receiver)
}
}
#[cfg(test)]
mod tests {
use super::*;
#[tokio::test]
async fn discord_serialize_should_match() {
let discord_receiver = DiscordWebhook {
name: "test-discord".to_string(),
url: Url::Url(url::Url::parse("https://discord.i.dont.exist.com").unwrap()),
};
let discord_receiver_receiver =
serde_yaml::to_string(&discord_receiver.alert_channel_receiver().await).unwrap();
println!("receiver \n{:#}", discord_receiver_receiver);
let discord_receiver_receiver_yaml = r#"name: test-discord
discord_configs:
- webhook_url: https://discord.i.dont.exist.com/
"#
.to_string();
let discord_receiver_route =
serde_yaml::to_string(&discord_receiver.alert_channel_route().await).unwrap();
println!("route \n{:#}", discord_receiver_route);
let discord_receiver_route_yaml = r#"receiver: test-discord
matchers:
- alertname!=Watchdog
continue: true
"#
.to_string();
assert_eq!(discord_receiver_receiver, discord_receiver_receiver_yaml);
assert_eq!(discord_receiver_route, discord_receiver_route_yaml);
}
}

View File

@@ -0,0 +1,2 @@
pub mod discord_alert_channel;
pub mod webhook_receiver;

View File

@@ -0,0 +1,124 @@
use async_trait::async_trait;
use serde::Serialize;
use serde_yaml::{Mapping, Value};
use crate::{
interpret::{InterpretError, Outcome},
modules::monitoring::kube_prometheus::{
prometheus::{Prometheus, PrometheusReceiver},
types::{AlertChannelConfig, AlertManagerChannelConfig},
},
topology::{Url, oberservability::monitoring::AlertReceiver},
};
#[derive(Debug, Clone, Serialize)]
pub struct WebhookReceiver {
pub name: String,
pub url: Url,
}
#[async_trait]
impl AlertReceiver<Prometheus> for WebhookReceiver {
async fn install(&self, sender: &Prometheus) -> Result<Outcome, InterpretError> {
sender.install_receiver(self).await
}
fn clone_box(&self) -> Box<dyn AlertReceiver<Prometheus>> {
Box::new(self.clone())
}
}
#[async_trait]
impl PrometheusReceiver for WebhookReceiver {
fn name(&self) -> String {
self.name.clone()
}
async fn configure_receiver(&self) -> AlertManagerChannelConfig {
self.get_config().await
}
}
#[async_trait]
impl AlertChannelConfig for WebhookReceiver {
async fn get_config(&self) -> AlertManagerChannelConfig {
let channel_global_config = None;
let channel_receiver = self.alert_channel_receiver().await;
let channel_route = self.alert_channel_route().await;
AlertManagerChannelConfig {
channel_global_config,
channel_receiver,
channel_route,
}
}
}
impl WebhookReceiver {
async fn alert_channel_route(&self) -> serde_yaml::Value {
let mut route = Mapping::new();
route.insert(
Value::String("receiver".to_string()),
Value::String(self.name.clone()),
);
route.insert(
Value::String("matchers".to_string()),
Value::Sequence(vec![Value::String("alertname!=Watchdog".to_string())]),
);
route.insert(Value::String("continue".to_string()), Value::Bool(true));
Value::Mapping(route)
}
async fn alert_channel_receiver(&self) -> serde_yaml::Value {
let mut receiver = Mapping::new();
receiver.insert(
Value::String("name".to_string()),
Value::String(self.name.clone()),
);
let mut webhook_config = Mapping::new();
webhook_config.insert(
Value::String("url".to_string()),
Value::String(self.url.to_string()),
);
receiver.insert(
Value::String("webhook_configs".to_string()),
Value::Sequence(vec![Value::Mapping(webhook_config)]),
);
Value::Mapping(receiver)
}
}
#[cfg(test)]
mod tests {
use super::*;
#[tokio::test]
async fn webhook_serialize_should_match() {
let webhook_receiver = WebhookReceiver {
name: "test-webhook".to_string(),
url: Url::Url(url::Url::parse("https://webhook.i.dont.exist.com").unwrap()),
};
let webhook_receiver_receiver =
serde_yaml::to_string(&webhook_receiver.alert_channel_receiver().await).unwrap();
println!("receiver \n{:#}", webhook_receiver_receiver);
let webhook_receiver_receiver_yaml = r#"name: test-webhook
webhook_configs:
- url: https://webhook.i.dont.exist.com/
"#
.to_string();
let webhook_receiver_route =
serde_yaml::to_string(&webhook_receiver.alert_channel_route().await).unwrap();
println!("route \n{:#}", webhook_receiver_route);
let webhook_receiver_route_yaml = r#"receiver: test-webhook
matchers:
- alertname!=Watchdog
continue: true
"#
.to_string();
assert_eq!(webhook_receiver_receiver, webhook_receiver_receiver_yaml);
assert_eq!(webhook_receiver_route, webhook_receiver_route_yaml);
}
}

View File

@@ -0,0 +1 @@
pub mod prometheus_alert_rule;

View File

@@ -0,0 +1,99 @@
use std::collections::{BTreeMap, HashMap};
use async_trait::async_trait;
use serde::Serialize;
use crate::{
interpret::{InterpretError, Outcome},
modules::monitoring::kube_prometheus::{
prometheus::{Prometheus, PrometheusRule},
types::{AlertGroup, AlertManagerAdditionalPromRules},
},
topology::oberservability::monitoring::AlertRule,
};
#[async_trait]
impl AlertRule<Prometheus> for AlertManagerRuleGroup {
async fn install(&self, sender: &Prometheus) -> Result<Outcome, InterpretError> {
sender.install_rule(&self).await
}
fn clone_box(&self) -> Box<dyn AlertRule<Prometheus>> {
Box::new(self.clone())
}
}
#[async_trait]
impl PrometheusRule for AlertManagerRuleGroup {
fn name(&self) -> String {
self.name.clone()
}
async fn configure_rule(&self) -> AlertManagerAdditionalPromRules {
let mut additional_prom_rules = BTreeMap::new();
additional_prom_rules.insert(
self.name.clone(),
AlertGroup {
groups: vec![self.clone()],
},
);
AlertManagerAdditionalPromRules {
rules: additional_prom_rules,
}
}
}
impl AlertManagerRuleGroup {
pub fn new(name: &str, rules: Vec<PrometheusAlertRule>) -> AlertManagerRuleGroup {
AlertManagerRuleGroup {
name: name.to_string().to_lowercase(),
rules,
}
}
}
#[derive(Debug, Clone, Serialize)]
///logical group of alert rules
///evaluates to:
///name:
/// groups:
/// - name: name
/// rules: PrometheusAlertRule
pub struct AlertManagerRuleGroup {
pub name: String,
pub rules: Vec<PrometheusAlertRule>,
}
#[derive(Debug, Clone, Serialize)]
pub struct PrometheusAlertRule {
pub alert: String,
pub expr: String,
#[serde(skip_serializing_if = "Option::is_none")]
pub r#for: Option<String>,
pub labels: HashMap<String, String>,
pub annotations: HashMap<String, String>,
}
impl PrometheusAlertRule {
pub fn new(alert_name: &str, expr: &str) -> Self {
Self {
alert: alert_name.into(),
expr: expr.into(),
r#for: Some("1m".into()),
labels: HashMap::new(),
annotations: HashMap::new(),
}
}
pub fn for_duration(mut self, duration: &str) -> Self {
self.r#for = Some(duration.into());
self
}
pub fn label(mut self, key: &str, value: &str) -> Self {
self.labels.insert(key.into(), value.into());
self
}
pub fn annotation(mut self, key: &str, value: &str) -> Self {
self.annotations.insert(key.into(), value.into());
self
}
}

View File

@@ -1,32 +0,0 @@
use std::str::FromStr;
use non_blank_string_rs::NonBlankString;
use url::Url;
use crate::modules::helm::chart::HelmChartScore;
pub fn discord_alert_manager_score(name: String, webhook: Url, namespace: String) -> HelmChartScore {
let url = webhook;
let values = format!(
r#"
environment:
- name: "DISCORD_WEBHOOK"
value: "{url}"
"#,
);
HelmChartScore {
namespace: Some(NonBlankString::from_str(&namespace).unwrap()),
release_name: NonBlankString::from_str(&name).unwrap(),
chart_name: NonBlankString::from_str(
"oci://hub.nationtech.io/library/alertmanager-discord",
)
.unwrap(),
chart_version: None,
values_overrides: None,
values_yaml: Some(values.to_string()),
create_namespace: true,
install_only: true,
repository: None,
}
}

View File

@@ -1,16 +1,16 @@
use serde::Serialize;
use super::kube_prometheus::{prometheus_alert_channel::PrometheusAlertChannel, types::AlertManagerValues};
use crate::modules::monitoring::{
alert_rule::prometheus_alert_rule::AlertManagerRuleGroup,
kube_prometheus::types::{AlertManagerAdditionalPromRules, AlertManagerChannelConfig},
};
#[derive(Debug, Clone, Serialize)]
pub struct KubePrometheusChartConfig {
pub struct KubePrometheusConfig {
pub namespace: String,
pub default_rules: bool,
pub windows_monitoring: bool,
pub alert_manager: bool,
pub alert_manager_values: AlertManagerValues,
pub node_exporter: bool,
pub prometheus: bool,
pub grafana: bool,
@@ -24,17 +24,16 @@ pub struct KubePrometheusChartConfig {
pub kube_proxy: bool,
pub kube_state_metrics: bool,
pub prometheus_operator: bool,
pub alert_channels: Vec<Box<dyn PrometheusAlertChannel>>,
pub alert_receiver_configs: Vec<AlertManagerChannelConfig>,
pub alert_rules: Vec<AlertManagerAdditionalPromRules>,
}
impl KubePrometheusChartConfig {
impl KubePrometheusConfig {
pub fn new() -> Self {
Self {
namespace: "monitoring".into(),
default_rules: true,
windows_monitoring: false,
alert_manager: true,
alert_manager_values: AlertManagerValues::default(),
alert_channels: Vec::new(),
grafana: true,
node_exporter: false,
prometheus: true,
@@ -48,6 +47,8 @@ impl KubePrometheusChartConfig {
prometheus_operator: true,
core_dns: false,
kube_scheduler: false,
alert_receiver_configs: vec![],
alert_rules: vec![],
}
}
}

View File

@@ -0,0 +1,201 @@
use super::config::KubePrometheusConfig;
use log::debug;
use non_blank_string_rs::NonBlankString;
use serde_yaml::{Mapping, Value};
use std::{
collections::BTreeMap,
str::FromStr,
sync::{Arc, Mutex},
};
use crate::modules::{
helm::chart::HelmChartScore,
monitoring::kube_prometheus::types::{
AlertGroup, AlertManager, AlertManagerAdditionalPromRules, AlertManagerConfig,
AlertManagerRoute, AlertManagerValues,
},
};
pub fn kube_prometheus_helm_chart_score(
config: Arc<Mutex<KubePrometheusConfig>>,
) -> HelmChartScore {
let config = config.lock().unwrap();
let default_rules = config.default_rules.to_string();
let windows_monitoring = config.windows_monitoring.to_string();
let grafana = config.grafana.to_string();
let kubernetes_service_monitors = config.kubernetes_service_monitors.to_string();
let kubernetes_api_server = config.kubernetes_api_server.to_string();
let kubelet = config.kubelet.to_string();
let kube_controller_manager = config.kube_controller_manager.to_string();
let core_dns = config.core_dns.to_string();
let kube_etcd = config.kube_etcd.to_string();
let kube_scheduler = config.kube_scheduler.to_string();
let kube_proxy = config.kube_proxy.to_string();
let kube_state_metrics = config.kube_state_metrics.to_string();
let node_exporter = config.node_exporter.to_string();
let prometheus_operator = config.prometheus_operator.to_string();
let prometheus = config.prometheus.to_string();
let mut values = format!(
r#"
defaultRules:
create: {default_rules}
rules:
alertmanager: true
etcd: true
configReloaders: true
general: true
k8sContainerCpuUsageSecondsTotal: true
k8sContainerMemoryCache: true
k8sContainerMemoryRss: true
k8sContainerMemorySwap: true
k8sContainerResource: true
k8sContainerMemoryWorkingSetBytes: true
k8sPodOwner: true
kubeApiserverAvailability: true
kubeApiserverBurnrate: true
kubeApiserverHistogram: true
kubeApiserverSlos: true
kubeControllerManager: true
kubelet: true
kubeProxy: true
kubePrometheusGeneral: true
kubePrometheusNodeRecording: true
kubernetesApps: true
kubernetesResources: true
kubernetesStorage: true
kubernetesSystem: true
kubeSchedulerAlerting: true
kubeSchedulerRecording: true
kubeStateMetrics: true
network: true
node: true
nodeExporterAlerting: true
nodeExporterRecording: true
prometheus: true
prometheusOperator: true
windows: true
windowsMonitoring:
enabled: {windows_monitoring}
grafana:
enabled: {grafana}
kubernetesServiceMonitors:
enabled: {kubernetes_service_monitors}
kubeApiServer:
enabled: {kubernetes_api_server}
kubelet:
enabled: {kubelet}
kubeControllerManager:
enabled: {kube_controller_manager}
coreDns:
enabled: {core_dns}
kubeEtcd:
enabled: {kube_etcd}
kubeScheduler:
enabled: {kube_scheduler}
kubeProxy:
enabled: {kube_proxy}
kubeStateMetrics:
enabled: {kube_state_metrics}
nodeExporter:
enabled: {node_exporter}
prometheusOperator:
enabled: {prometheus_operator}
prometheus:
enabled: {prometheus}
"#,
);
// add required null receiver for prometheus alert manager
let mut null_receiver = Mapping::new();
null_receiver.insert(
Value::String("receiver".to_string()),
Value::String("null".to_string()),
);
null_receiver.insert(
Value::String("matchers".to_string()),
Value::Sequence(vec![Value::String("alertname!=Watchdog".to_string())]),
);
null_receiver.insert(Value::String("continue".to_string()), Value::Bool(true));
//add alert channels
let mut alert_manager_channel_config = AlertManagerConfig {
global: Mapping::new(),
route: AlertManagerRoute {
routes: vec![Value::Mapping(null_receiver)],
},
receivers: vec![serde_yaml::from_str("name: 'null'").unwrap()],
};
for receiver in config.alert_receiver_configs.iter() {
if let Some(global) = receiver.channel_global_config.clone() {
alert_manager_channel_config
.global
.insert(global.0, global.1);
}
alert_manager_channel_config
.route
.routes
.push(receiver.channel_route.clone());
alert_manager_channel_config
.receivers
.push(receiver.channel_receiver.clone());
}
let alert_manager_values = AlertManagerValues {
alertmanager: AlertManager {
enabled: config.alert_manager,
config: alert_manager_channel_config,
},
};
let alert_manager_yaml =
serde_yaml::to_string(&alert_manager_values).expect("Failed to serialize YAML");
debug!("serialized alert manager: \n {:#}", alert_manager_yaml);
values.push_str(&alert_manager_yaml);
//format alert manager additional rules for helm chart
let mut merged_rules: BTreeMap<String, AlertGroup> = BTreeMap::new();
for additional_rule in config.alert_rules.clone() {
for (key, group) in additional_rule.rules {
merged_rules.insert(key, group);
}
}
let merged_rules = AlertManagerAdditionalPromRules {
rules: merged_rules,
};
let mut alert_manager_additional_rules = serde_yaml::Mapping::new();
let rules_value = serde_yaml::to_value(merged_rules).unwrap();
alert_manager_additional_rules.insert(
serde_yaml::Value::String("additionalPrometheusRulesMap".to_string()),
rules_value,
);
let alert_manager_additional_rules_yaml =
serde_yaml::to_string(&alert_manager_additional_rules).expect("Failed to serialize YAML");
debug!(
"alert_rules_yaml:\n{:#}",
alert_manager_additional_rules_yaml
);
values.push_str(&alert_manager_additional_rules_yaml);
debug!("full values.yaml: \n {:#}", values);
HelmChartScore {
namespace: Some(NonBlankString::from_str(&config.namespace).unwrap()),
release_name: NonBlankString::from_str("kube-prometheus").unwrap(),
chart_name: NonBlankString::from_str(
"oci://ghcr.io/prometheus-community/charts/kube-prometheus-stack",
)
.unwrap(),
chart_version: None,
values_overrides: None,
values_yaml: Some(values.to_string()),
create_namespace: true,
install_only: true,
repository: None,
}
}

View File

@@ -0,0 +1,2 @@
pub mod config;
pub mod kube_prometheus_helm_chart;

View File

@@ -0,0 +1,33 @@
use std::sync::{Arc, Mutex};
use serde::Serialize;
use super::{helm::config::KubePrometheusConfig, prometheus::Prometheus};
use crate::{
score::Score,
topology::{
HelmCommand, Topology,
oberservability::monitoring::{AlertReceiver, AlertRule, AlertingInterpret},
},
};
#[derive(Clone, Debug, Serialize)]
pub struct HelmPrometheusAlertingScore {
pub receivers: Vec<Box<dyn AlertReceiver<Prometheus>>>,
pub rules: Vec<Box<dyn AlertRule<Prometheus>>>,
}
impl<T: Topology + HelmCommand> Score<T> for HelmPrometheusAlertingScore {
fn create_interpret(&self) -> Box<dyn crate::interpret::Interpret<T>> {
Box::new(AlertingInterpret {
sender: Prometheus {
config: Arc::new(Mutex::new(KubePrometheusConfig::new())),
},
receivers: self.receivers.clone(),
rules: self.rules.clone(),
})
}
fn name(&self) -> String {
"HelmPrometheusAlertingScore".to_string()
}
}

View File

@@ -1,168 +0,0 @@
use crate::modules::{helm::chart::HelmChartScore, monitoring::config::KubePrometheusChartConfig};
use log::info;
use non_blank_string_rs::NonBlankString;
use serde_yaml::{self};
use std::str::FromStr;
pub fn kube_prometheus_helm_chart_score(config: &KubePrometheusChartConfig) -> HelmChartScore {
//TODO this should be make into a rule with default formatting that can be easily passed as a vec
//to the overrides or something leaving the user to deal with formatting here seems bad
let default_rules = config.default_rules.to_string();
let windows_monitoring = config.windows_monitoring.to_string();
let grafana = config.grafana.to_string();
let kubernetes_service_monitors = config.kubernetes_service_monitors.to_string();
let kubernetes_api_server = config.kubernetes_api_server.to_string();
let kubelet = config.kubelet.to_string();
let kube_controller_manager = config.kube_controller_manager.to_string();
let core_dns = config.core_dns.to_string();
let kube_etcd = config.kube_etcd.to_string();
let kube_scheduler = config.kube_scheduler.to_string();
let kube_proxy = config.kube_proxy.to_string();
let kube_state_metrics = config.kube_state_metrics.to_string();
let node_exporter = config.node_exporter.to_string();
let prometheus_operator = config.prometheus_operator.to_string();
let prometheus = config.prometheus.to_string();
let alert_manager_values = config.alert_manager_values.clone();
let mut values = format!(
r#"
additionalPrometheusRulesMap:
pods-status-alerts:
groups:
- name: pods
rules:
- alert: "[CRIT] POD not healthy"
expr: min_over_time(sum by (namespace, pod) (kube_pod_status_phase{{phase=~"Pending|Unknown|Failed"}})[15m:1m]) > 0
for: 0m
labels:
severity: critical
annotations:
title: "[CRIT] POD not healthy : {{{{ $labels.pod }}}}"
description: |
A POD is in a non-ready state!
- **Pod**: {{{{ $labels.pod }}}}
- **Namespace**: {{{{ $labels.namespace }}}}
- alert: "[CRIT] POD crash looping"
expr: increase(kube_pod_container_status_restarts_total[5m]) > 3
for: 0m
labels:
severity: critical
annotations:
title: "[CRIT] POD crash looping : {{{{ $labels.pod }}}}"
description: |
A POD is drowning in a crash loop!
- **Pod**: {{{{ $labels.pod }}}}
- **Namespace**: {{{{ $labels.namespace }}}}
- **Instance**: {{{{ $labels.instance }}}}
pvc-alerts:
groups:
- name: pvc-alerts
rules:
- alert: 'PVC Fill Over 95 Percent In 2 Days'
expr: |
(
kubelet_volume_stats_used_bytes
/
kubelet_volume_stats_capacity_bytes
) > 0.95
AND
predict_linear(kubelet_volume_stats_used_bytes[2d], 2 * 24 * 60 * 60)
/
kubelet_volume_stats_capacity_bytes
> 0.95
for: 1m
labels:
severity: warning
annotations:
description: The PVC {{{{ $labels.persistentvolumeclaim }}}} in namespace {{{{ $labels.namespace }}}} is predicted to fill over 95% in less than 2 days.
title: PVC {{{{ $labels.persistentvolumeclaim }}}} in namespace {{{{ $labels.namespace }}}} will fill over 95% in less than 2 days
defaultRules:
create: {default_rules}
rules:
alertmanager: true
etcd: true
configReloaders: true
general: true
k8sContainerCpuUsageSecondsTotal: true
k8sContainerMemoryCache: true
k8sContainerMemoryRss: true
k8sContainerMemorySwap: true
k8sContainerResource: true
k8sContainerMemoryWorkingSetBytes: true
k8sPodOwner: true
kubeApiserverAvailability: true
kubeApiserverBurnrate: true
kubeApiserverHistogram: true
kubeApiserverSlos: true
kubeControllerManager: true
kubelet: true
kubeProxy: true
kubePrometheusGeneral: true
kubePrometheusNodeRecording: true
kubernetesApps: true
kubernetesResources: true
kubernetesStorage: true
kubernetesSystem: true
kubeSchedulerAlerting: true
kubeSchedulerRecording: true
kubeStateMetrics: true
network: true
node: true
nodeExporterAlerting: true
nodeExporterRecording: true
prometheus: true
prometheusOperator: true
windows: true
windowsMonitoring:
enabled: {windows_monitoring}
grafana:
enabled: {grafana}
kubernetesServiceMonitors:
enabled: {kubernetes_service_monitors}
kubeApiServer:
enabled: {kubernetes_api_server}
kubelet:
enabled: {kubelet}
kubeControllerManager:
enabled: {kube_controller_manager}
coreDns:
enabled: {core_dns}
kubeEtcd:
enabled: {kube_etcd}
kubeScheduler:
enabled: {kube_scheduler}
kubeProxy:
enabled: {kube_proxy}
kubeStateMetrics:
enabled: {kube_state_metrics}
nodeExporter:
enabled: {node_exporter}
prometheusOperator:
enabled: {prometheus_operator}
prometheus:
enabled: {prometheus}
prometheusSpec:
maximumStartupDurationSeconds: 240
"#,
);
let alert_manager_yaml = serde_yaml::to_string(&alert_manager_values).expect("Failed to serialize YAML");
values.push_str(&alert_manager_yaml);
info!("{}", values);
HelmChartScore {
namespace: Some(NonBlankString::from_str(&config.namespace).unwrap()),
release_name: NonBlankString::from_str("kube-prometheus").unwrap(),
chart_name: NonBlankString::from_str(
"oci://ghcr.io/prometheus-community/charts/kube-prometheus-stack",
)
.unwrap(),
chart_version: None,
values_overrides: None,
values_yaml: Some(values.to_string()),
create_namespace: true,
install_only: true,
repository: None,
}
}

View File

@@ -1,3 +1,4 @@
pub mod kube_prometheus;
pub mod helm;
pub mod helm_prometheus_alert_score;
pub mod prometheus;
pub mod types;
pub mod prometheus_alert_channel;

View File

@@ -0,0 +1,139 @@
use std::sync::{Arc, Mutex};
use async_trait::async_trait;
use log::debug;
use serde::Serialize;
use crate::{
interpret::{InterpretError, Outcome},
inventory::Inventory,
modules::monitoring::alert_rule::prometheus_alert_rule::AlertManagerRuleGroup,
score,
topology::{
HelmCommand, Topology,
installable::Installable,
oberservability::monitoring::{AlertReceiver, AlertRule, AlertSender},
},
};
use score::Score;
use super::{
helm::{
config::KubePrometheusConfig, kube_prometheus_helm_chart::kube_prometheus_helm_chart_score,
},
types::{AlertManagerAdditionalPromRules, AlertManagerChannelConfig},
};
#[async_trait]
impl AlertSender for Prometheus {
fn name(&self) -> String {
"HelmKubePrometheus".to_string()
}
}
#[async_trait]
impl<T: Topology + HelmCommand> Installable<T> for Prometheus {
async fn ensure_installed(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<(), InterpretError> {
self.install_prometheus(inventory, topology).await?;
Ok(())
}
}
#[derive(Debug)]
pub struct Prometheus {
pub config: Arc<Mutex<KubePrometheusConfig>>,
}
impl Prometheus {
pub async fn install_receiver(
&self,
prometheus_receiver: &dyn PrometheusReceiver,
) -> Result<Outcome, InterpretError> {
let prom_receiver = prometheus_receiver.configure_receiver().await;
debug!(
"adding alert receiver to prometheus config: {:#?}",
&prom_receiver
);
let mut config = self.config.lock().unwrap();
config.alert_receiver_configs.push(prom_receiver);
let prom_receiver_name = prometheus_receiver.name();
debug!("installed alert receiver {}", &prom_receiver_name);
Ok(Outcome::success(format!(
"Sucessfully installed receiver {}",
prom_receiver_name
)))
}
pub async fn install_rule(
&self,
prometheus_rule: &AlertManagerRuleGroup,
) -> Result<Outcome, InterpretError> {
let prometheus_rule = prometheus_rule.configure_rule().await;
let mut config = self.config.lock().unwrap();
config.alert_rules.push(prometheus_rule.clone());
Ok(Outcome::success(format!(
"Successfully installed alert rule: {:#?},",
prometheus_rule
)))
}
pub async fn install_prometheus<T: Topology + HelmCommand + Send + Sync>(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
kube_prometheus_helm_chart_score(self.config.clone())
.create_interpret()
.execute(inventory, topology)
.await
}
}
#[async_trait]
pub trait PrometheusReceiver: Send + Sync + std::fmt::Debug {
fn name(&self) -> String;
async fn configure_receiver(&self) -> AlertManagerChannelConfig;
}
impl Serialize for Box<dyn AlertReceiver<Prometheus>> {
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{
todo!()
}
}
impl Clone for Box<dyn AlertReceiver<Prometheus>> {
fn clone(&self) -> Self {
self.clone_box()
}
}
#[async_trait]
pub trait PrometheusRule: Send + Sync + std::fmt::Debug {
fn name(&self) -> String;
async fn configure_rule(&self) -> AlertManagerAdditionalPromRules;
}
impl Serialize for Box<dyn AlertRule<Prometheus>> {
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{
todo!()
}
}
impl Clone for Box<dyn AlertRule<Prometheus>> {
fn clone(&self) -> Self {
self.clone_box()
}
}

View File

@@ -1,140 +0,0 @@
use crate::{
interpret::InterpretError,
modules::{
helm::chart::HelmChartScore,
monitoring::{
discord_alert_manager::discord_alert_manager_score,
kube_prometheus::types::{
AlertChannelConfig, AlertChannelGlobalConfig, AlertChannelReceiver,
AlertChannelRoute, SlackConfig, WebhookConfig,
},
},
},
};
use dyn_clone::DynClone;
use serde::{Deserialize, Serialize};
use std::fmt::Debug;
use url::Url;
#[typetag::serde(tag = "channel_type")]
#[async_trait::async_trait]
pub trait PrometheusAlertChannel: DynClone + Debug + Send + Sync {
fn get_alert_manager_config_contribution(&self) -> Result<AlertChannelConfig, InterpretError>;
fn get_dependency_score(&self, namespace:String) -> Option<HelmChartScore>;
}
dyn_clone::clone_trait_object!(PrometheusAlertChannel);
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct DiscordChannel {
pub name: String,
pub webhook_url: Url,
}
#[typetag::serde]
impl PrometheusAlertChannel for DiscordChannel {
fn get_alert_manager_config_contribution(&self) -> Result<AlertChannelConfig, InterpretError> {
let service_url = format!("http://{}-alertmanager-discord:9094", &self.name);
Ok(AlertChannelConfig {
receiver: AlertChannelReceiver {
name: format!("Discord-{}", self.name),
slack_configs: None,
webhook_configs: Some(vec![WebhookConfig {
url: url::Url::parse(&service_url)
.expect("invalid url"),
send_resolved: true,
}]),
},
route: AlertChannelRoute {
receiver: format!("Discord-{}", self.name),
matchers: vec!["alertname!=Watchdog".to_string()],
r#continue: true,
},
global_config: None,
})
}
fn get_dependency_score(&self, namespace: String) -> Option<HelmChartScore> {
Some(discord_alert_manager_score(self.name.clone(), self.webhook_url.clone(), namespace.clone()))
}
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct SlackChannel {
pub name: String,
pub webhook_url: Url,
}
#[typetag::serde]
impl PrometheusAlertChannel for SlackChannel {
fn get_alert_manager_config_contribution(&self) -> Result<AlertChannelConfig, InterpretError> {
Ok(AlertChannelConfig {
receiver: AlertChannelReceiver {
name: format!("Slack-{}", self.name),
slack_configs: Some(vec![SlackConfig {
channel: self.name.clone(),
send_resolved: true,
title: "{{ .CommonAnnotations.title }}".to_string(),
text: ">-
*Alert:* {{ .CommonLabels.alertname }}
*Severity:* {{ .CommonLabels.severity }}
*Namespace:* {{ .CommonLabels.namespace }}
*Pod:* {{ .CommonLabels.pod }}
*ExternalURL:* {{ .ExternalURL }}
{{ range .Alerts }}
*Instance:* {{ .Labels.instance }}
*Summary:* {{ .Annotations.summary }}
*Description:* {{ .Annotations.description }}
*Starts At:* {{ .StartsAt }}
*Status:* {{ .Status }}
{{ end }}"
.to_string(),
}]),
webhook_configs: None,
},
route: AlertChannelRoute {
receiver: format!("Slack-{}", self.name),
matchers: vec!["alertname!=Watchdog".to_string()],
r#continue: true,
},
global_config: Some(AlertChannelGlobalConfig {
slack_api_url: Some(self.webhook_url.clone()),
}),
})
}
fn get_dependency_score(&self, _namespace: String) -> Option<HelmChartScore> {
None
}
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct NullReceiver {}
impl NullReceiver {
pub fn new() -> Self {
Self {}
}
}
#[typetag::serde]
impl PrometheusAlertChannel for NullReceiver {
fn get_alert_manager_config_contribution(&self) -> Result<AlertChannelConfig, InterpretError> {
Ok(AlertChannelConfig {
receiver: AlertChannelReceiver {
name: "null".to_string(),
slack_configs: None,
webhook_configs: None,
},
route: AlertChannelRoute {
receiver: "null".to_string(),
matchers: vec!["alertname=Watchdog".to_string()],
r#continue: false,
},
global_config: None,
})
}
fn get_dependency_score(&self, _namespace: String) -> Option<HelmChartScore> {
None
}
}

View File

@@ -1,94 +1,55 @@
use serde::{Deserialize, Serialize};
use url::Url;
use std::collections::BTreeMap;
#[derive(Debug, Clone, Serialize, Deserialize)]
use async_trait::async_trait;
use serde::Serialize;
use serde_yaml::{Mapping, Sequence, Value};
use crate::modules::monitoring::alert_rule::prometheus_alert_rule::AlertManagerRuleGroup;
#[async_trait]
pub trait AlertChannelConfig {
async fn get_config(&self) -> AlertManagerChannelConfig;
}
#[derive(Debug, Clone, Serialize)]
pub struct AlertManagerValues {
pub alertmanager: AlertManager,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
#[derive(Debug, Clone, Serialize)]
pub struct AlertManager {
pub enabled: bool,
pub config: AlertManagerConfig,
}
#[derive(Clone, Debug, Serialize, Deserialize)]
pub struct AlertChannelConfig {
pub receiver: AlertChannelReceiver,
pub route: AlertChannelRoute,
pub global_config: Option<AlertChannelGlobalConfig>,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct AlertChannelReceiver {
pub name: String,
#[serde(skip_serializing_if = "Option::is_none")]
pub slack_configs: Option<Vec<SlackConfig>>,
#[serde(skip_serializing_if = "Option::is_none")]
pub webhook_configs: Option<Vec<WebhookConfig>>,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct AlertManagerRoute {
pub group_by: Vec<String>,
pub group_wait: String,
pub group_interval: String,
pub repeat_interval: String,
pub routes: Vec<AlertChannelRoute>,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct AlertChannelGlobalConfig {
#[serde(skip_serializing_if = "Option::is_none")]
pub slack_api_url: Option<Url>,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct SlackConfig {
pub channel: String,
pub send_resolved: bool,
pub title: String,
pub text: String,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct WebhookConfig {
pub url: Url,
pub send_resolved: bool,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct AlertChannelRoute {
pub receiver: String,
pub matchers: Vec<String>,
#[serde(default)]
pub r#continue: bool,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
#[derive(Debug, Clone, Serialize)]
pub struct AlertManagerConfig {
pub global: Option<AlertChannelGlobalConfig>,
pub global: Mapping,
pub route: AlertManagerRoute,
pub receivers: Vec<AlertChannelReceiver>,
pub receivers: Sequence,
}
impl AlertManagerValues {
pub fn default() -> Self {
Self {
alertmanager: AlertManager {
enabled: true,
config: AlertManagerConfig {
global: None,
route: AlertManagerRoute {
group_by: vec!["job".to_string()],
group_wait: "30s".to_string(),
group_interval: "5m".to_string(),
repeat_interval: "12h".to_string(),
routes: vec![AlertChannelRoute{ receiver: "null".to_string(), matchers: vec!["alertname=Watchdog".to_string()], r#continue: false }],
},
receivers: vec![AlertChannelReceiver{ name: "null".to_string(), slack_configs: None, webhook_configs: None }],
},
},
}
}
#[derive(Debug, Clone, Serialize)]
pub struct AlertManagerRoute {
pub routes: Sequence,
}
#[derive(Debug, Clone, Serialize)]
pub struct AlertManagerChannelConfig {
///expecting an option that contains two values
///if necessary for the alertchannel
///[ jira_api_url: <string> ]
pub channel_global_config: Option<(Value, Value)>,
pub channel_route: Value,
pub channel_receiver: Value,
}
#[derive(Debug, Clone, Serialize)]
pub struct AlertManagerAdditionalPromRules {
#[serde(flatten)]
pub rules: BTreeMap<String, AlertGroup>,
}
#[derive(Debug, Clone, Serialize)]
pub struct AlertGroup {
pub groups: Vec<AlertManagerRuleGroup>,
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,5 +1,3 @@
mod config;
mod discord_alert_manager;
pub mod alert_channel;
pub mod alert_rule;
pub mod kube_prometheus;
pub mod monitoring_alerting;

View File

@@ -1,188 +0,0 @@
use async_trait::async_trait;
use log::info;
use serde::Serialize;
use crate::{
data::{Id, Version},
interpret::{Interpret, InterpretError, InterpretName, InterpretStatus, Outcome},
inventory::Inventory,
modules::monitoring::kube_prometheus::types::{
AlertManager, AlertManagerConfig, AlertManagerRoute,
},
score::Score,
topology::{HelmCommand, Topology},
};
use super::{
config::KubePrometheusChartConfig,
kube_prometheus::{
kube_prometheus::kube_prometheus_helm_chart_score,
prometheus_alert_channel::{NullReceiver, PrometheusAlertChannel},
types::AlertManagerValues,
},
};
#[derive(Debug, Clone, Serialize)]
pub struct MonitoringAlertingScore {
pub alert_channels: Vec<Box<dyn PrometheusAlertChannel>>,
pub namespace: Option<String>,
}
impl MonitoringAlertingScore {
pub fn new() -> Self {
Self {
alert_channels: Vec::new(),
namespace: None,
}
}
}
impl<T: Topology + HelmCommand> Score<T> for MonitoringAlertingScore {
fn create_interpret(&self) -> Box<dyn Interpret<T>> {
Box::new(MonitoringAlertingInterpret {
score: self.clone(),
})
}
fn name(&self) -> String {
format!("MonitoringAlertingStackScore")
}
}
#[derive(Debug, Clone, Serialize)]
struct MonitoringAlertingInterpret {
score: MonitoringAlertingScore,
}
impl MonitoringAlertingInterpret {
async fn build_kube_prometheus_helm_chart_config(&self) -> KubePrometheusChartConfig {
let mut config = KubePrometheusChartConfig::new();
let mut receivers = Vec::new();
let mut routes = Vec::new();
let mut global_config = None;
if let Some(ns) = &self.score.namespace {
config.namespace = ns.clone();
};
let null_channel = NullReceiver::new();
let null_channel = null_channel
.get_alert_manager_config_contribution()
.unwrap();
receivers.push(null_channel.receiver);
routes.push(null_channel.route);
for channel in self.score.alert_channels.clone() {
let alert_manager_config_contribution =
channel.get_alert_manager_config_contribution().unwrap();
receivers.push(alert_manager_config_contribution.receiver);
routes.push(alert_manager_config_contribution.route);
if let Some(global) = alert_manager_config_contribution.global_config {
global_config = Some(global);
}
}
info!("after alert receiver: {:#?}", receivers);
info!("after alert routes: {:#?}", routes);
let alert_manager_config = AlertManagerConfig {
global: global_config,
route: AlertManagerRoute {
group_by: vec!["job".to_string()],
group_wait: "30s".to_string(),
group_interval: "5m".to_string(),
repeat_interval: "12h".to_string(),
routes,
},
receivers,
};
info!("alert manager config: {:?}", config);
config.alert_manager_values = AlertManagerValues {
alertmanager: AlertManager {
enabled: true,
config: alert_manager_config,
},
};
config
}
async fn deploy_kube_prometheus_helm_chart_score<T: Topology + HelmCommand>(
&self,
inventory: &Inventory,
topology: &T,
config: &KubePrometheusChartConfig,
) -> Result<Outcome, InterpretError> {
let helm_chart = kube_prometheus_helm_chart_score(config);
helm_chart
.create_interpret()
.execute(inventory, topology)
.await
}
async fn deploy_alert_channel_dependencies<T: Topology + HelmCommand>(
&self,
inventory: &Inventory,
topology: &T,
config: &KubePrometheusChartConfig,
) -> Result<Outcome, InterpretError> {
let mut outcomes = Vec::new();
for channel in &self.score.alert_channels {
let ns = config.namespace.clone();
if let Some(dependency_score) = channel.get_dependency_score(ns) {
match dependency_score
.create_interpret()
.execute(inventory, topology)
.await
{
Ok(outcome) => outcomes.push(outcome),
Err(e) => {
info!("failed to deploy dependency: {}", { &e });
return Err(e);
}
}
}
}
Ok(Outcome::success("All alert channels deployed".to_string()))
}
}
#[async_trait]
impl<T: Topology + HelmCommand> Interpret<T> for MonitoringAlertingInterpret {
async fn execute(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
let config = self.build_kube_prometheus_helm_chart_config().await;
info!("Built kube prometheus config{:?}", config);
info!("Installing kube prometheus chart");
self.deploy_kube_prometheus_helm_chart_score(inventory, topology, &config)
.await?;
info!("Installing alert channel service");
self.deploy_alert_channel_dependencies(inventory, topology, &config)
.await?;
Ok(Outcome::success(format!(
"succesfully deployed monitoring and alerting score"
)))
}
fn get_name(&self) -> InterpretName {
todo!()
}
fn get_version(&self) -> Version {
todo!()
}
fn get_status(&self) -> InterpretStatus {
todo!()
}
fn get_children(&self) -> Vec<Id> {
todo!()
}
}

View File

@@ -0,0 +1,40 @@
use crate::modules::monitoring::alert_rule::prometheus_alert_rule::PrometheusAlertRule;
pub fn global_storage_status_degraded_non_critical() -> PrometheusAlertRule {
PrometheusAlertRule::new("GlobalStorageStatusNonCritical", "globalStorageStatus == 4")
.for_duration("5m")
.label("severity", "warning")
.annotation(
"description",
"- **system**: {{ $labels.instance }}\n- **Status**: nonCritical\n- **Value**: {{ $value }}\n- **Job**: {{ $labels.job }}",
)
.annotation("title", " System storage status is in degraded state")
}
pub fn alert_global_storage_status_critical() -> PrometheusAlertRule {
PrometheusAlertRule::new(
"GlobalStorageStatus critical",
"globalStorageStatus == 5",
)
.for_duration("5m")
.label("severity", "warning")
.annotation("title", "System storage status is critical at {{ $labels.instance }}")
.annotation(
"description",
"- **System**: {{ $labels.instance }}\n- **Status**: Critical\n- **Value**: {{ $value }}\n- **Job**: {{ $labels.job }}",
)
}
pub fn alert_global_storage_status_non_recoverable() -> PrometheusAlertRule {
PrometheusAlertRule::new(
"GlobalStorageStatus nonRecoverable",
"globalStorageStatus == 6",
)
.for_duration("5m")
.label("severity", "warning")
.annotation("title", "System storage status is nonRecoverable at {{ $labels.instance }}")
.annotation(
"description",
"- **System**: {{ $labels.instance }}\n- **Status**: nonRecoverable\n- **Value**: {{ $value }}\n- **Job**: {{ $labels.job }}",
)
}

View File

@@ -0,0 +1 @@
pub mod dell_server;

View File

@@ -0,0 +1 @@
pub mod pvc;

View File

@@ -0,0 +1,11 @@
use crate::modules::monitoring::alert_rule::prometheus_alert_rule::PrometheusAlertRule;
pub fn high_pvc_fill_rate_over_two_days() -> PrometheusAlertRule {
PrometheusAlertRule::new(
"PVC Fill Over 95 Percent In 2 Days",
"(kubelet_volume_stats_used_bytes/kubelet_volume_stats_capacity_bytes) > 0.95 AND predict_linear(kubelet_volume_stats_used_bytes[2d], 2 * 24 * 60 * 60)/kubelet_volume_stats_capacity_bytes > 0.95",)
.for_duration("1m")
.label("severity", "warning")
.annotation("summary", "The PVC {{ $labels.persistentvolumeclaim }} in namespace {{ $labels.namespace }} is predicted to fill over 95% in less than 2 days.")
.annotation("description", "PVC {{ $labels.persistentvolumeclaim }} in namespace {{ $labels.namespace }} will fill over 95% in less than 2 days",)
}

View File

@@ -0,0 +1,2 @@
pub mod infra;
pub mod k8s;

View File

@@ -0,0 +1 @@
pub mod alerts;

View File

@@ -0,0 +1,50 @@
use async_trait::async_trait;
use chrono::{DateTime, Utc};
use serde::Serialize;
use crate::{interpret::InterpretError, score::Score, topology::Topology};
/// Create and manage Tenant Credentials.
///
/// This is meant to be used by cluster administrators who need to provide their tenant users and
/// services with credentials to access their resources.
#[derive(Debug, Clone, Serialize)]
pub struct TenantCredentialScore;
impl<T: Topology + TenantCredentialManager> Score<T> for TenantCredentialScore {
fn create_interpret(&self) -> Box<dyn crate::interpret::Interpret<T>> {
todo!()
}
fn name(&self) -> String {
todo!()
}
}
#[async_trait]
pub trait TenantCredentialManager {
async fn create_user(&self) -> Result<TenantCredentialBundle, InterpretError>;
}
#[derive(Debug, Clone)]
pub struct CredentialMetadata {
pub tenant_id: String,
pub credential_id: String,
pub description: String,
pub created_at: DateTime<Utc>,
pub expires_at: Option<DateTime<Utc>>,
}
#[derive(Debug, Clone)]
pub enum CredentialData {
/// Used to store login instructions destined to a human. Akin to AWS login instructions email
/// upon new console user creation.
PlainText(String),
}
pub struct TenantCredentialBundle {
_metadata: CredentialMetadata,
_content: CredentialData,
}
impl TenantCredentialBundle {}

View File

@@ -0,0 +1,70 @@
mod credentials;
pub use credentials::*;
use async_trait::async_trait;
use serde::Serialize;
use crate::{
data::{Id, Version},
interpret::{Interpret, InterpretError, InterpretName, InterpretStatus, Outcome},
inventory::Inventory,
score::Score,
topology::{
Topology,
tenant::{TenantConfig, TenantManager},
},
};
#[derive(Debug, Serialize, Clone)]
pub struct TenantScore {
pub config: TenantConfig,
}
impl<T: Topology + TenantManager> Score<T> for TenantScore {
fn create_interpret(&self) -> Box<dyn crate::interpret::Interpret<T>> {
Box::new(TenantInterpret {
tenant_config: self.config.clone(),
})
}
fn name(&self) -> String {
format!("{} TenantScore", self.config.name)
}
}
#[derive(Debug)]
pub struct TenantInterpret {
tenant_config: TenantConfig,
}
#[async_trait]
impl<T: Topology + TenantManager> Interpret<T> for TenantInterpret {
async fn execute(
&self,
_inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
topology.provision_tenant(&self.tenant_config).await?;
Ok(Outcome::success(format!(
"Successfully provisioned tenant {} with id {}",
self.tenant_config.name, self.tenant_config.id
)))
}
fn get_name(&self) -> InterpretName {
InterpretName::TenantInterpret
}
fn get_version(&self) -> Version {
todo!()
}
fn get_status(&self) -> InterpretStatus {
todo!()
}
fn get_children(&self) -> Vec<Id> {
todo!()
}
}

27
harmony_cli/README.md Normal file
View File

@@ -0,0 +1,27 @@
## Quick demo
`cargo run -p example-tui`
This will launch Harmony's minimalist terminal ui which embeds a few demo scores.
Usage instructions will be displayed at the bottom of the TUI.
`cargo run --bin example-cli -- --help`
This is the harmony CLI, a minimal implementation
The current help text:
```
Usage: example-cli [OPTIONS]
Options:
-y, --yes Run score(s) or not
-f, --filter <FILTER> Filter query
-i, --interactive Run interactive TUI or not
-a, --all Run all or nth, defaults to all
-n, --number <NUMBER> Run nth matching, zero indexed [default: 0]
-l, --list list scores, will also be affected by run filter
-h, --help Print help
-V, --version Print version```

View File

@@ -0,0 +1,17 @@
[package]
name = "harmony_composer"
edition = "2024"
version.workspace = true
readme.workspace = true
license.workspace = true
[dependencies]
clap = { version = "4.5.35", features = ["derive"] }
tokio.workspace = true
env_logger.workspace = true
log.workspace = true
bollard = "0.19.0"
current_platform = "0.2.0"
futures-util = "0.3.31"
serde_json = "1.0.140"
cargo_metadata = "0.20.0"

View File

@@ -0,0 +1,66 @@
# harmony_composer
`harmony_composer` is a command-line utility for compiling and bootstrapping deployments for the Harmony orchestration framework.
It's designed to simplify the build process by either compiling a Harmony project found in a local harmony directory or by bootstrapping a new deployment through auto-detection of the current project type.
## ⚡ Quick Install & Run (Linux x86-64)
You can download and run the latest snapshot build with a single command. This will place the binary in ~/.local/bin, which should be in your PATH on most modern Linux distributions.
```bash
curl -Ls https://git.nationtech.io/NationTech/harmony/releases/download/snapshot-latest/harmony_composer \
-o ~/.local/bin/harmony_composer && \
chmod +x ~/.local/bin/harmony_composer
```
> ⚠️ Warning: Unstable Builds
> The snapshot-latest tag points to the latest build from the master branch. It is unstable, unsupported, and intended only for early testing of new features. Please do not use it in production environments.
## ⚙️ How It Works
harmony_composer requires either cargo or docker to be available on your system to compile the Harmony project.
- If cargo is found: It will be used to compile the project locally.
- If cargo is not found: It will automatically download and run the harmony_composer Docker image. This image is a self-contained build environment with the required Cargo binary and build targets for both Linux and Windows.
- If both cargo and docker are unavailable, `harmony_composer` will fail. Please install one of them.
## 📖 Basic Usage
Here are some common commands:
```bash
# Compile the repo's Harmony module
harmony_composer compile
# Run check script on the project
harmony_composer check
# Run the repo's entire harmony deployment sequence
harmony_composer deploy
# Run the full check, compile, and deploy pipeline
harmony_composer all
```
For a full list of commands and their options, run:
```bash
harmony_composer --help
```
## 🏗️ Supported Architectures
The build system currently supports compiling for:
x86_64-unknown-linux-gnu
x86_64-pc-windows-gnu
More target architectures are planned. If your platform is not yet supported, please open a feature request in the main repository.
## 🔗 Main Project
This tool is a small part of the main Harmony project. For complete documentation, contribution guidelines, and license information, please refer to the main repository.

View File

@@ -0,0 +1,327 @@
use bollard::models::ContainerCreateBody;
use bollard::query_parameters::{
CreateContainerOptionsBuilder, ListContainersOptionsBuilder, LogsOptions,
RemoveContainerOptions, StartContainerOptions, WaitContainerOptions,
};
use bollard::secret::HostConfig;
use cargo_metadata::{Artifact, Message, MetadataCommand};
use clap::{Args, Parser, Subcommand};
use futures_util::StreamExt;
use log::info;
use std::collections::HashMap;
use std::path::{Path, PathBuf};
use std::process::{Command, Stdio};
use tokio::fs;
#[derive(Parser)]
#[command(version, about, long_about = None, flatten_help = true, propagate_version = true)]
struct GlobalArgs {
#[arg(long, default_value = "harmony")]
harmony_path: String,
#[arg(long)]
compile_method: Option<CompileMethod>,
#[arg(long)]
compile_platform: Option<String>,
#[command(subcommand)]
command: Option<Commands>,
}
#[derive(Subcommand, Clone, Debug)]
enum Commands {
Check(CheckArgs),
Compile,
Deploy(DeployArgs),
All(AllArgs),
}
#[derive(Args, Clone, Debug)]
struct CheckArgs {
#[arg(long, default_value = "check.sh")]
check_script_path: String,
}
#[derive(Args, Clone, Debug)]
struct DeployArgs {
#[arg(long, default_value_t = false)]
staging: bool,
#[arg(long, default_value_t = false)]
prod: bool,
#[arg(long, default_value_t = false)]
smoke_test: bool,
}
#[derive(Args, Clone, Debug)]
struct AllArgs {
#[command(flatten)]
check: CheckArgs,
#[command(flatten)]
deploy: DeployArgs,
}
#[tokio::main]
async fn main() {
env_logger::init();
let cli_args = GlobalArgs::parse();
let harmony_path = Path::new(&cli_args.harmony_path)
.try_exists()
.expect("couldn't check if path exists");
let harmony_bin_path: PathBuf;
match harmony_path {
true => {
harmony_bin_path = compile_harmony(
cli_args.compile_method,
cli_args.compile_platform,
cli_args.harmony_path.clone(),
)
.await
}
false => todo!("implement autodetect code"),
}
match cli_args.command {
Some(command) => match command {
Commands::Check(args) => {
let check_script_str =
format!("{}/{}", cli_args.harmony_path, args.check_script_path);
let check_script = Path::new(&check_script_str);
match check_script
.try_exists()
.expect("couldn't check if path exists")
{
true => (),
false => todo!("implement couldn't find path logic"),
};
let check_output = Command::new(check_script)
.output()
.expect("failed to run check script");
info!(
"check stdout: {}, check stderr: {}",
String::from_utf8(check_output.stdout).expect("couldn't parse from utf8"),
String::from_utf8(check_output.stderr).expect("couldn't parse from utf8")
);
}
Commands::Deploy(args) => {
if args.staging {
todo!("implement staging deployment");
}
if args.prod {
todo!("implement prod deployment");
}
let deploy_output = Command::new(harmony_bin_path)
.arg("-y")
.arg("-a")
.output()
.expect("failed to run harmony deploy");
println!(
"deploy output: {}",
String::from_utf8(deploy_output.stdout).expect("couldn't parse from utf8")
);
}
Commands::All(_args) => todo!(
"take all previous match arms and turn them into separate functions, and call them all one after the other"
),
Commands::Compile => return,
},
None => todo!("run interactively, ask for info on CLI"),
}
}
#[derive(Clone, Debug, clap::ValueEnum)]
enum CompileMethod {
LocalCargo,
Docker,
}
async fn compile_harmony(
method: Option<CompileMethod>,
platform: Option<String>,
harmony_location: String,
) -> PathBuf {
let platform = match platform {
Some(p) => p,
None => current_platform::CURRENT_PLATFORM.to_string(),
};
let cargo_exists = Command::new("which")
.arg("cargo")
.status()
.expect("couldn't get `which cargo` status")
.success();
let method = match method {
Some(m) => m,
None => {
if cargo_exists {
return compile_cargo(platform, harmony_location).await;
} else {
return compile_docker(platform, harmony_location).await;
}
}
};
match method {
CompileMethod::LocalCargo => return compile_cargo(platform, harmony_location).await,
CompileMethod::Docker => return compile_docker(platform, harmony_location).await,
};
}
// TODO: make sure this works with cargo workspaces
async fn compile_cargo(platform: String, harmony_location: String) -> PathBuf {
let metadata = MetadataCommand::new()
.manifest_path(format!("{}/Cargo.toml", harmony_location))
.exec()
.unwrap();
let mut cargo_build = Command::new("cargo")
.current_dir(&harmony_location)
.args(vec![
"build",
format!("--target={}", platform).as_str(),
"--release",
"--message-format=json-render-diagnostics",
])
.stdout(Stdio::piped())
.spawn()
.expect("run cargo command failed");
let mut artifacts: Vec<Artifact> = vec![];
let reader = std::io::BufReader::new(cargo_build.stdout.take().unwrap());
for message in cargo_metadata::Message::parse_stream(reader) {
match message.unwrap() {
Message::CompilerMessage(_msg) => (),
Message::CompilerArtifact(artifact) => {
if artifact.manifest_path
== metadata
.root_package()
.expect("failed to get root package")
.manifest_path
{
println!("{:?}", artifact);
artifacts.push(artifact);
}
}
Message::BuildScriptExecuted(_script) => (),
Message::BuildFinished(finished) => {
println!("{:?}", finished);
}
_ => (), // Unknown message
}
}
let bin = artifacts
.last()
.expect("no binaries built")
.filenames
.first()
.expect("couldn't get filename");
let bin_out;
if let Some(ext) = bin.extension() {
bin_out = PathBuf::from(format!("{}/harmony.{}", harmony_location, ext));
let _copy_res = fs::copy(&bin, &bin_out).await;
} else {
bin_out = PathBuf::from(format!("{}/harmony", harmony_location));
let _copy_res = fs::copy(&bin, &bin_out).await;
}
return bin_out;
}
async fn compile_docker(platform: String, harmony_location: String) -> PathBuf {
let docker_client =
bollard::Docker::connect_with_local_defaults().expect("couldn't connect to docker");
let mut filters = HashMap::new();
filters.insert(String::from("name"), vec![String::from("harmony_build")]);
let list_containers_options = ListContainersOptionsBuilder::new()
.all(true)
.filters(&filters)
.build();
let containers = &docker_client
.list_containers(Some(list_containers_options))
.await
.expect("list containers failed");
if containers.len() > 0 {
docker_client
.remove_container("harmony_build", None::<RemoveContainerOptions>)
.await
.expect("failed to remove container");
}
let options = Some(
CreateContainerOptionsBuilder::new()
.name("harmony_build")
.build(),
);
let config = ContainerCreateBody {
image: Some("hub.nationtech.io/harmony/harmony_composer".to_string()),
working_dir: Some("/mnt".to_string()),
cmd: Some(vec![
format!("--compile-platform={}", platform),
format!("--harmony-path=/mnt"),
"compile".to_string(),
]),
host_config: Some(HostConfig {
binds: Some(vec![format!("{}:/mnt", harmony_location)]),
..Default::default()
}),
..Default::default()
};
docker_client
.create_container(options, config)
.await
.expect("couldn't create container");
docker_client
.start_container("harmony_build", None::<StartContainerOptions>)
.await
.expect("couldn't start container");
let wait_options = WaitContainerOptions {
condition: "not-running".to_string(),
};
let mut wait = docker_client
.wait_container("harmony_build", Some(wait_options))
.boxed();
let mut logs_stream = docker_client.logs(
"harmony_build",
Some(LogsOptions {
follow: true,
stdout: true,
stderr: true,
tail: "all".to_string(),
..Default::default()
}),
);
while let Some(l) = logs_stream.next().await {
let l_str = l.expect("couldn't unwrap logoutput").to_string();
println!("{}", l_str);
}
// wait until container is no longer running
while let Some(_) = wait.next().await {}
// hack that should be cleaned up
if platform.contains("windows") {
return PathBuf::from(format!("{}/harmony.exe", harmony_location));
} else {
return PathBuf::from(format!("{}/harmony", harmony_location));
}
}

View File

@@ -14,6 +14,7 @@ quote = "1.0.37"
serde = "1.0.217"
serde_yaml = "0.9.34"
syn = "2.0.90"
cidr.workspace = true
[dev-dependencies]
serde = { version = "1.0.217", features = ["derive"] }

View File

@@ -132,3 +132,16 @@ pub fn ingress_path(input: TokenStream) -> TokenStream {
false => panic!("Invalid ingress path"),
}
}
#[proc_macro]
pub fn cidrv4(input: TokenStream) -> TokenStream {
let input = parse_macro_input!(input as LitStr);
let cidr_str = input.value();
if let Ok(_) = cidr_str.parse::<cidr::Ipv4Cidr>() {
let expanded = quote! { #cidr_str.parse::<cidr::Ipv4Cidr>().unwrap() };
return TokenStream::from(expanded);
}
panic!("Invalid IPv4 CIDR : {}", cidr_str);
}

146
opnsense-config/README.md Normal file
View File

@@ -0,0 +1,146 @@
## Supporting a new field in OPNSense `config.xml`
Two steps:
- Supporting the field in `opnsense-config-xml`
- Enabling Harmony to control the field
We'll use the `filename` field in the `dhcpcd` section of the file as an example.
### Supporting the field
As type checking if enforced, every field from `config.xml` must be known by the code. Each subsection of `config.xml` has its `.rs` file. For the `dhcpcd` section, we'll modify `opnsense-config-xml/src/data/dhcpd.rs`.
When a new field appears in the xml file, an error like this will be thrown and Harmony will panic :
```
Running `/home/stremblay/nt/dir/harmony/target/debug/example-nanodc`
Found unauthorized element filename
thread 'main' panicked at opnsense-config-xml/src/data/opnsense.rs:54:14:
OPNSense received invalid string, should be full XML: ()
```
Define the missing field (`filename`) in the `DhcpInterface` struct of `opnsense-config-xml/src/data/dhcpd.rs`:
```rust
pub struct DhcpInterface {
...
pub filename: Option<String>,
```
Harmony should now be fixed, build and run.
### Controlling the field
Define the `xml field setter` in `opnsense-config/src/modules/dhcpd.rs`.
```rust
impl<'a> DhcpConfig<'a> {
...
pub fn set_filename(&mut self, filename: &str) {
self.enable_netboot();
self.get_lan_dhcpd().filename = Some(filename.to_string());
}
...
```
Define the `value setter` in the `DhcpServer trait` in `domain/topology/network.rs`
```rust
#[async_trait]
pub trait DhcpServer: Send + Sync {
...
async fn set_filename(&self, filename: &str) -> Result<(), ExecutorError>;
...
```
Implement the `value setter` in each `DhcpServer` implementation.
`infra/opnsense/dhcp.rs`:
```rust
#[async_trait]
impl DhcpServer for OPNSenseFirewall {
...
async fn set_filename(&self, filename: &str) -> Result<(), ExecutorError> {
{
let mut writable_opnsense = self.opnsense_config.write().await;
writable_opnsense.dhcp().set_filename(filename);
debug!("OPNsense dhcp server set filename {filename}");
}
Ok(())
}
...
```
`domain/topology/ha_cluster.rs`
```rust
#[async_trait]
impl DhcpServer for DummyInfra {
...
async fn set_filename(&self, _filename: &str) -> Result<(), ExecutorError> {
unimplemented!("{}", UNIMPLEMENTED_DUMMY_INFRA)
}
...
```
Add the new field to the DhcpScore in `modules/dhcp.rs`
```rust
pub struct DhcpScore {
...
pub filename: Option<String>,
```
Define it in its implementation in `modules/okd/dhcp.rs`
```rust
impl OKDDhcpScore {
...
Self {
dhcp_score: DhcpScore {
...
filename: Some("undionly.kpxe".to_string()),
```
Define it in its implementation in `modules/okd/bootstrap_dhcp.rs`
```rust
impl OKDDhcpScore {
...
Self {
dhcp_score: DhcpScore::new(
...
Some("undionly.kpxe".to_string()),
```
Update the interpret (function called by the `execute` fn of the interpret) so it now updates the `filename` field value in `modules/dhcp.rs`
```rust
impl DhcpInterpret {
...
let filename_outcome = match &self.score.filename {
Some(filename) => {
let dhcp_server = Arc::new(topology.dhcp_server.clone());
dhcp_server.set_filename(&filename).await?;
Outcome::new(
InterpretStatus::SUCCESS,
format!("Dhcp Interpret Set filename to {filename}"),
)
}
None => Outcome::noop(),
};
if next_server_outcome.status == InterpretStatus::NOOP
&& boot_filename_outcome.status == InterpretStatus::NOOP
&& filename_outcome.status == InterpretStatus::NOOP
...
Ok(Outcome::new(
InterpretStatus::SUCCESS,
format!(
"Dhcp Interpret Set next boot to [{:?}], boot_filename to [{:?}], filename to [{:?}]",
self.score.boot_filename, self.score.boot_filename, self.score.filename
)
...
```