Compare commits

..

39 Commits

Author SHA1 Message Date
Ian Letourneau
387ae9f494 group related scores together 2025-08-09 23:20:45 -04:00
Ian Letourneau
336e1cfefe rename a few scores & interprets
Some checks failed
Run Check Script / check (pull_request) Has been cancelled
2025-08-09 23:18:15 -04:00
Ian Letourneau
403e199062 fix: improve usage of indicatif for tracking progress 2025-08-09 23:18:15 -04:00
29a261575b refactor: Interpret score with a provided method on Score (#100)
Some checks failed
Compile and package harmony_composer / package_harmony_composer (push) Successful in 6m49s
Run Check Script / check (push) Failing after 41s
First step in a direction to better orchestrate the core flow, even though it feels weird to move this logic into the `Score`. We'll refactor this as soon as we have a better solution.

Co-authored-by: Ian Letourneau <letourneau.ian@gmail.com>
Reviewed-on: #100
2025-08-09 22:56:23 +00:00
dcf8335240 Merge pull request 'refactor: Remove InterpretStatus/Error & Outcome from Topology' (#99) from remove-interpret-status-from-topology into master
Some checks are pending
Run Check Script / check (push) Waiting to run
Compile and package harmony_composer / package_harmony_composer (push) Waiting to run
Reviewed-on: #99
Reviewed-by: johnride <jg@nationtech.io>
2025-08-09 22:52:21 +00:00
Ian Letourneau
f876b5e67b refactor: Remove InterpretStatus/Error & Outcome from Topology
Some checks failed
Run Check Script / check (pull_request) Has been cancelled
2025-08-06 22:29:00 -04:00
440c1bce12 chore: reformat & clippy cleanup (#96)
Some checks failed
Run Check Script / check (pull_request) Has been cancelled
Run Check Script / check (push) Has been cancelled
Compile and package harmony_composer / package_harmony_composer (push) Has been cancelled
Clippy is now added to the `check` in the pipeline

Co-authored-by: Ian Letourneau <letourneau.ian@gmail.com>
Reviewed-on: #96
2025-08-06 15:57:14 +00:00
024084859e Monitor an application within a tenant (#86)
All checks were successful
Run Check Script / check (push) Successful in -45s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 4m35s
WIP: added implementation to deploy crd-alertmanagerconfigs
Co-authored-by: Ian Letourneau <letourneau.ian@gmail.com>
Reviewed-on: #86
Co-authored-by: Willem <wrolleman@nationtech.io>
Co-committed-by: Willem <wrolleman@nationtech.io>
2025-08-04 21:42:01 +00:00
54990cd1a5 fix(cli): simplify running the CLI by hiding the maestro inside the implemtation (#93)
All checks were successful
Run Check Script / check (push) Successful in -46s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 4m36s
Co-authored-by: Ian Letourneau <letourneau.ian@gmail.com>
Reviewed-on: #93
2025-08-04 20:59:07 +00:00
06aab1f57f fix(cli): reduce noise & better track progress within Harmony (#91)
All checks were successful
Run Check Script / check (push) Successful in -37s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 9m6s
Introduce a way to instrument what happens within Harmony and around Harmony (e.g. in the CLI or in Composer).

The goal is to provide visual feedback to the end users and inform them of the progress of their tasks (e.g. deployment) as clearly as possible. It is important to also let them know of the outcome of their tasks (what was created, where to access stuff, etc.).

<img src="https://media.discordapp.net/attachments/1295353830300713062/1400289618636574741/demo.gif?ex=688c18d5&is=688ac755&hm=2c70884aacb08f7bd15cbb65a7562a174846906718aa15294bbb238e64febbce&=" />

## Changes

### Instrumentation architecture
Extensibility and ease of use is key here, while preserving type safety as much as possible.

The proposed API is quite simple:
```rs
// Emit an event
instrumentation::instrument(
    HarmonyEvent::TopologyPrepared {
        topology: "k8s-anywhere",
        outcome: Outcome::success("yay")
    }
);

// Consume events
instrumentation::subscribe("Harmony CLI Logger", async |event| {
    match event {
        HarmonyEvent::TopologyPrepared { name, outcome } => todo!(),
    }
});
```

#### Current limitations
* this API is not very extensible, but it could be easily changed to allow end users to define custom events in addition to Harmony core events
* we use a tokio broadcast channel behind the scene so only in process communication can happen, but it could be easily changed to a more flexible communication mechanism as implementation details are hidden

### `harmony_composer` VS `harmony_cli`
As Harmony Composer launches commands from Harmony (CLI), they both live in different processes. And because of this, we cannot easily make all the logging happens in one place (Harmony Composer) and get rid of Harmony CLI. At least not without introducing additional complexity such as communication through a server, unix socket, etc.

So for the time being, it was decided to preserve both `harmony_composer` and `harmony_cli` and let them independently log their stuff and handle their own responsibilities:
* `harmony_composer`: takes care only of setting up & packaging a project, delegates everything else to `harmony_cli`
* `harmony_cli`: takes care of configuring & running Harmony

### Logging & prompts
* [indicatif](https://github.com/console-rs/indicatif) is used to create progress bars and track progress within Harmony, Harmony CLI, and Harmony Composer
* [inquire](https://github.com/mikaelmello/inquire) is preserved, but was removed from `harmony` (core) as UI concerns shouldn't go that deep
  * note: for now the only prompt we had was simply deleted, we'll have to find a better way to prompt stuff in the future

## Todos
* [ ] Update/Create ADRs
* [ ] Continue instrumentation for missing branches
* [ ] Allow instrumentation to emit and subscribe to custom events

Co-authored-by: Ian Letourneau <letourneau.ian@gmail.com>
Reviewed-on: #91
Reviewed-by: johnride <jg@nationtech.io>
2025-07-31 19:35:33 +00:00
1ab66af718 Merge pull request 'refactor(topo/k8s_anywhere): simplify local installation of k3d' (#90) from simply-k3d-installation into master
Some checks failed
Run Check Script / check (push) Failing after -1m14s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 3m23s
Reviewed-on: #90
Reviewed-by: wjro <wrolleman@nationtech.io>
2025-07-31 13:22:25 +00:00
Ian Letourneau
0fff4ef566 refactor(topo/k8s_anywhere): simplify local installation of k3d
All checks were successful
Run Check Script / check (pull_request) Successful in -37s
A Maestro was initialized with a new inventory simply to provide a
localhost topology to install K3D locally. But in practice, the K3D
installation wasn't actually using the topology nor the inventory.

Directly installing K3D within the K8s Anywhere topology makes things
simpler and actually enforce the topology to provide the capabilities
required to install K3D.
2025-07-27 11:50:48 -04:00
d95e84d6fc Merge pull request 'fix(apps/rust): build & push using image tag instead of local VS remote image name' (#87) from fix-image-tag into master
All checks were successful
Run Check Script / check (push) Successful in -37s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 13m2s
Reviewed-on: #87
Reviewed-by: johnride <jg@nationtech.io>
2025-07-27 14:10:19 +00:00
a47be890de Merge branch 'master' into fix-image-tag
All checks were successful
Run Check Script / check (pull_request) Successful in -38s
2025-07-27 14:09:24 +00:00
ee8dfa4a93 Merge pull request 'chore: cleanup of unnecessary files & adjust gitignores' (#88) from quick-cleanup into master
Some checks failed
Run Check Script / check (push) Successful in -37s
Compile and package harmony_composer / package_harmony_composer (push) Has been cancelled
Reviewed-on: #88
Reviewed-by: johnride <jg@nationtech.io>
2025-07-27 14:08:57 +00:00
5d41cc8380 Merge branch 'master' into quick-cleanup
All checks were successful
Run Check Script / check (pull_request) Successful in -34s
2025-07-27 14:07:55 +00:00
cef745b642 Merge pull request 'log(composer): Log check_path_str value when error' (#77) from log/composer into master
All checks were successful
Run Check Script / check (push) Successful in -31s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 11m59s
Reviewed-on: #77
2025-07-21 18:04:57 +00:00
d9959378a6 log(composer): Log check_path_str value when error
All checks were successful
Run Check Script / check (pull_request) Successful in -37s
2025-07-21 09:15:41 -04:00
Ian Letourneau
07f1151e4c chore: cleanup of unncessary files & adjust gitignores
All checks were successful
Run Check Script / check (pull_request) Successful in -31s
2025-07-20 20:03:26 -04:00
Ian Letourneau
f7625f0484 fix(rust): push only the actual image tag
All checks were successful
Run Check Script / check (pull_request) Successful in -22s
2025-07-16 13:51:02 -04:00
tahahawa
537da5800f uncomment docker image build
All checks were successful
Run Check Script / check (push) Successful in 2m49s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 5m20s
2025-07-11 10:34:37 -04:00
3be2fa246c fix: unjank the demo (#85)
Some checks failed
Run Check Script / check (push) Has been cancelled
Compile and package harmony_composer / package_harmony_composer (push) Has been cancelled
Co-authored-by: tahahawa <tahahawa@gmail.com>
Reviewed-on: #85
Reviewed-by: wjro <wrolleman@nationtech.io>
2025-07-11 14:32:16 +00:00
9452cf5616 Merge pull request 'fix/argoApplication' (#84) from fix/argoApplication into master
All checks were successful
Run Check Script / check (push) Successful in 1m41s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 4m4s
Reviewed-on: #84
2025-07-05 01:19:05 +00:00
9b7456e148 Merge pull request 'feat/monitoring-application-feature' (#83) from feat/monitoring-application-feature into master
Some checks are pending
Compile and package harmony_composer / package_harmony_composer (push) Has started running
Run Check Script / check (push) Successful in 2s
Reviewed-on: #83
Reviewed-by: johnride <jg@nationtech.io>
2025-07-05 01:16:08 +00:00
98f3f82ad5 refact: Rename HttpScore into StaticFileHttpScore and add minimal documentation
All checks were successful
Run Check Script / check (pull_request) Successful in 1m43s
2025-07-04 21:05:32 -04:00
3eca409f8d Merge remote-tracking branch 'origin/feat/monitoring-application-feature' into fix/argoApplication 2025-07-04 16:44:03 -04:00
c11a31c7a9 wip: Fix ArgoApplication 2025-07-04 16:43:10 -04:00
1a6d72dc17 fix: uncommented example
All checks were successful
Run Check Script / check (pull_request) Successful in 1m37s
2025-07-04 16:30:13 -04:00
df9e21807e fix: git conflict
All checks were successful
Run Check Script / check (pull_request) Successful in -6s
2025-07-04 16:22:39 -04:00
b1bf4fd4d5 fix: cargo fmt
All checks were successful
Run Check Script / check (pull_request) Successful in 1m40s
2025-07-04 16:14:47 -04:00
f702ecd8c9 fix: deploys a lighter weight prometheus and grafana which is limited to their respective namespaces 2025-07-04 16:13:41 -04:00
a19b52e690 fix: properly append YAML in correct places in argoapplication (#80)
All checks were successful
Run Check Script / check (push) Successful in -7s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 3m56s
Co-authored-by: tahahawa <tahahawa@gmail.com>
Reviewed-on: #80
2025-07-04 15:32:02 +00:00
b73f2e76d0 Merge pull request 'refact: Make RustWebappScore generic, it is now Application score and takes an application and list of features to attach to the application' (#81) from refact/application 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 3m38s
Reviewed-on: #81
Reviewed-by: wjro <wrolleman@nationtech.io>
2025-07-04 14:31:38 +00:00
b4534c6ee0 refact: Make RustWebappScore generic, it is now Application score and takes an application and list of features to attach to the application
All checks were successful
Run Check Script / check (pull_request) Successful in -8s
2025-07-04 10:27:16 -04:00
6149249a6c feat: create Argo interpret and kube client apply_yaml to install Argo Applications. Very messy implementation though, must be refactored/improved
All checks were successful
Run Check Script / check (push) Successful in -5s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 4m13s
2025-07-04 09:49:43 -04:00
d9935e20cb Merge pull request 'feat: harmony now defaults to using local k3d cluster. Also created OCICompliant: Application trait to make building images cleaner' (#76) from feat/oci into master
All checks were successful
Run Check Script / check (push) Successful in -9s
Compile and package harmony_composer / package_harmony_composer (push) Successful in 4m4s
Reviewed-on: #76
2025-07-03 19:37:46 +00:00
7b0f3b79b1 Merge remote-tracking branch 'origin/master' into feat/oci
All checks were successful
Run Check Script / check (pull_request) Successful in -8s
2025-07-03 15:36:52 -04:00
e6612245a5 Merge pull request 'feat/cd/localdeploymentdemo' (#79) from feat/cd/localdeploymentdemo into feat/oci
All checks were successful
Run Check Script / check (pull_request) Successful in -9s
Reviewed-on: #79
2025-07-03 19:31:45 +00:00
b4f5b91a57 feat: WIP argocd_score (#78)
Some checks are pending
Compile and package harmony_composer / package_harmony_composer (push) Waiting to run
Run Check Script / check (push) Successful in -8s
Co-authored-by: tahahawa <tahahawa@gmail.com>
Reviewed-on: #78
Reviewed-by: johnride <jg@nationtech.io>
Co-authored-by: Taha Hawa <taha@taha.dev>
Co-committed-by: Taha Hawa <taha@taha.dev>
2025-07-03 19:30:00 +00:00
130 changed files with 6064 additions and 888 deletions

29
.gitignore vendored
View File

@@ -1,4 +1,25 @@
target
private_repos
log/
*.tgz
### General ###
private_repos/
### Harmony ###
harmony.log
### Helm ###
# Chart dependencies
**/charts/*.tgz
### Rust ###
# Generated by Cargo
# will have compiled files and executables
debug/
target/
# Remove Cargo.lock from gitignore if creating an executable, leave it for libraries
# More information here https://doc.rust-lang.org/cargo/guide/cargo-toml-vs-cargo-lock.html
Cargo.lock
# These are backup files generated by rustfmt
**/*.rs.bk
# MSVC Windows builds of rustc generate these, which store debugging information
*.pdb

213
Cargo.lock generated
View File

@@ -96,6 +96,12 @@ dependencies = [
"libc",
]
[[package]]
name = "ansi_term"
version = "0.10.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6b3568b48b7cefa6b8ce125f9bb4989e52fbcc29ebea88df04cc7c5f12f70455"
[[package]]
name = "anstream"
version = "0.6.19"
@@ -650,6 +656,19 @@ dependencies = [
"crossbeam-utils",
]
[[package]]
name = "console"
version = "0.16.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2e09ced7ebbccb63b4c65413d821f2e00ce54c5ca4514ddc6b3c892fdbcbc69d"
dependencies = [
"encode_unicode",
"libc",
"once_cell",
"unicode-width 0.2.0",
"windows-sys 0.60.2",
]
[[package]]
name = "const-oid"
version = "0.9.6"
@@ -1137,6 +1156,12 @@ dependencies = [
"serde",
]
[[package]]
name = "encode_unicode"
version = "1.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "34aa73646ffb006b8f5147f3dc182bd4bcb190227ce861fc4a4844bf8e3cb2c0"
[[package]]
name = "encoding_rs"
version = "0.8.35"
@@ -1240,6 +1265,18 @@ dependencies = [
name = "example"
version = "0.0.0"
[[package]]
name = "example-application-monitoring-with-tenant"
version = "0.1.0"
dependencies = [
"env_logger",
"harmony",
"harmony_cli",
"logging",
"tokio",
"url",
]
[[package]]
name = "example-cli"
version = "0.1.0"
@@ -1355,6 +1392,7 @@ dependencies = [
name = "example-rust"
version = "0.1.0"
dependencies = [
"base64 0.22.1",
"env_logger",
"harmony",
"harmony_cli",
@@ -1427,6 +1465,18 @@ version = "0.2.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "28dea519a9695b9977216879a3ebfddf92f1c08c05d984f8996aecd6ecdc811d"
[[package]]
name = "filetime"
version = "0.2.25"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "35c0522e981e68cbfa8c3f978441a5f34b30b96e146b33cd3359176b50fe8586"
dependencies = [
"cfg-if",
"libc",
"libredox",
"windows-sys 0.59.0",
]
[[package]]
name = "flate2"
version = "1.1.2"
@@ -1726,6 +1776,8 @@ name = "harmony"
version = "0.1.0"
dependencies = [
"async-trait",
"base64 0.22.1",
"bollard",
"chrono",
"cidr",
"convert_case",
@@ -1742,27 +1794,32 @@ dependencies = [
"helm-wrapper-rs",
"hex",
"http 1.3.1",
"inquire",
"k3d-rs",
"k8s-openapi",
"kube",
"kube-derive",
"lazy_static",
"libredfish",
"log",
"non-blank-string-rs",
"once_cell",
"opnsense-config",
"opnsense-config-xml",
"pretty_assertions",
"rand 0.9.1",
"reqwest 0.11.27",
"russh",
"rust-ipmi",
"schemars 0.8.22",
"semver",
"serde",
"serde-value",
"serde_json",
"serde_with",
"serde_yaml",
"similar",
"strum 0.27.1",
"tar",
"temp-dir",
"temp-file",
"tempfile",
@@ -1778,10 +1835,15 @@ version = "0.1.0"
dependencies = [
"assert_cmd",
"clap",
"console",
"env_logger",
"harmony",
"harmony_tui",
"indicatif",
"indicatif-log-bridge",
"inquire",
"lazy_static",
"log",
"tokio",
]
@@ -1792,10 +1854,16 @@ dependencies = [
"bollard",
"cargo_metadata",
"clap",
"console",
"current_platform",
"env_logger",
"futures-util",
"harmony_cli",
"indicatif",
"indicatif-log-bridge",
"lazy_static",
"log",
"once_cell",
"serde_json",
"tokio",
]
@@ -2391,6 +2459,29 @@ dependencies = [
"serde",
]
[[package]]
name = "indicatif"
version = "0.18.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "70a646d946d06bedbbc4cac4c218acf4bbf2d87757a784857025f4d447e4e1cd"
dependencies = [
"console",
"portable-atomic",
"unicode-width 0.2.0",
"unit-prefix",
"web-time",
]
[[package]]
name = "indicatif-log-bridge"
version = "0.2.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "63703cf9069b85dbe6fe26e1c5230d013dee99d3559cd3d02ba39e099ef7ab02"
dependencies = [
"indicatif",
"log",
]
[[package]]
name = "indoc"
version = "2.0.6"
@@ -2598,6 +2689,7 @@ dependencies = [
"k8s-openapi",
"kube-client",
"kube-core",
"kube-derive",
"kube-runtime",
]
@@ -2651,12 +2743,27 @@ dependencies = [
"http 1.3.1",
"json-patch",
"k8s-openapi",
"schemars 0.8.22",
"serde",
"serde-value",
"serde_json",
"thiserror 2.0.12",
]
[[package]]
name = "kube-derive"
version = "1.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "079fc8c1c397538628309cfdee20696ebdcc26745f9fb17f89b78782205bd995"
dependencies = [
"darling",
"proc-macro2",
"quote",
"serde",
"serde_json",
"syn",
]
[[package]]
name = "kube-runtime"
version = "1.1.0"
@@ -2726,6 +2833,7 @@ checksum = "1580801010e535496706ba011c15f8532df6b42297d2e471fec38ceadd8c0638"
dependencies = [
"bitflags 2.9.1",
"libc",
"redox_syscall",
]
[[package]]
@@ -2771,6 +2879,15 @@ dependencies = [
"log",
]
[[package]]
name = "logging"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "461a8beca676e8ab1bd468c92e9b4436d6368e11e96ae038209e520cfe665e46"
dependencies = [
"ansi_term",
]
[[package]]
name = "lru"
version = "0.12.5"
@@ -4068,6 +4185,18 @@ dependencies = [
"windows-sys 0.59.0",
]
[[package]]
name = "schemars"
version = "0.8.22"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3fbf2ae1b8bc8e02df939598064d22402220cd5bbcca1c76f7d6a310974d5615"
dependencies = [
"dyn-clone",
"schemars_derive",
"serde",
"serde_json",
]
[[package]]
name = "schemars"
version = "0.9.0"
@@ -4080,6 +4209,30 @@ dependencies = [
"serde_json",
]
[[package]]
name = "schemars"
version = "1.0.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "82d20c4491bc164fa2f6c5d44565947a52ad80b9505d8e36f8d54c27c739fcd0"
dependencies = [
"dyn-clone",
"ref-cast",
"serde",
"serde_json",
]
[[package]]
name = "schemars_derive"
version = "0.8.22"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "32e265784ad618884abaea0600a9adf15393368d840e0222d101a072f3f7534d"
dependencies = [
"proc-macro2",
"quote",
"serde_derive_internals",
"syn",
]
[[package]]
name = "scopeguard"
version = "1.2.0"
@@ -4212,6 +4365,17 @@ dependencies = [
"syn",
]
[[package]]
name = "serde_derive_internals"
version = "0.29.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "18d26a20a969b9e3fdf2fc2d9f21eda6c40e2de84c9408bb5d3b05d499aae711"
dependencies = [
"proc-macro2",
"quote",
"syn",
]
[[package]]
name = "serde_json"
version = "1.0.140"
@@ -4280,22 +4444,36 @@ dependencies = [
[[package]]
name = "serde_with"
version = "3.13.0"
version = "3.14.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bf65a400f8f66fb7b0552869ad70157166676db75ed8181f8104ea91cf9d0b42"
checksum = "f2c45cd61fefa9db6f254525d46e392b852e0e61d9a1fd36e5bd183450a556d5"
dependencies = [
"base64 0.22.1",
"chrono",
"hex",
"indexmap 1.9.3",
"indexmap 2.10.0",
"schemars",
"schemars 0.9.0",
"schemars 1.0.4",
"serde",
"serde_derive",
"serde_json",
"serde_with_macros",
"time",
]
[[package]]
name = "serde_with_macros"
version = "3.14.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "de90945e6565ce0d9a25098082ed4ee4002e047cb59892c318d66821e14bb30f"
dependencies = [
"darling",
"proc-macro2",
"quote",
"syn",
]
[[package]]
name = "serde_yaml"
version = "0.9.34+deprecated"
@@ -4668,6 +4846,17 @@ version = "1.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "55937e1799185b12863d447f42597ed69d9928686b8d88a1df17376a097d8369"
[[package]]
name = "tar"
version = "0.4.44"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1d863878d212c87a19c1a610eb53bb01fe12951c0501cf5a0d65f724914a667a"
dependencies = [
"filetime",
"libc",
"xattr",
]
[[package]]
name = "temp-dir"
version = "0.1.16"
@@ -5125,6 +5314,12 @@ version = "0.2.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ebc1c04c71510c7f702b52b7c350734c9ff1295c464a03335b00bb84fc54f853"
[[package]]
name = "unit-prefix"
version = "0.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "323402cff2dd658f39ca17c789b502021b3f18707c91cdf22e3838e1b4023817"
[[package]]
name = "universal-hash"
version = "0.5.1"
@@ -5713,6 +5908,16 @@ dependencies = [
"tap",
]
[[package]]
name = "xattr"
version = "1.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "af3a19837351dc82ba89f8a125e22a3c475f05aba604acc023d62b2739ae2909"
dependencies = [
"libc",
"rustix 1.0.7",
]
[[package]]
name = "xml-rs"
version = "0.8.26"

View File

@@ -52,3 +52,7 @@ convert_case = "0.8"
chrono = "0.4"
similar = "2"
uuid = { version = "1.11", features = ["v4", "fast-rng", "macro-diagnostics"] }
pretty_assertions = "1.4.1"
bollard = "0.19.1"
base64 = "0.22.1"
tar = "0.4.44"

View File

@@ -13,6 +13,7 @@ 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 rustup component add clippy
RUN apt update
@@ -22,4 +23,4 @@ RUN apt install -y nodejs docker.io mingw-w64
COPY --from=build /app/target/release/harmony_composer .
ENTRYPOINT ["/app/harmony_composer"]
ENTRYPOINT ["/app/harmony_composer"]

View File

@@ -1,5 +1,6 @@
# Harmony : Open-source infrastructure orchestration that treats your platform like first-class code.
*By [NationTech](https://nationtech.io)*
# Harmony : Open-source infrastructure orchestration that treats your platform like first-class code
_By [NationTech](https://nationtech.io)_
[![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)
@@ -23,11 +24,11 @@ From a **developer laptop** to a **global production cluster**, a single **sourc
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.
| 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. |
| 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. |
These principles surface as simple, ergonomic Rust APIs that let teams focus on their product while trusting the platform underneath.
@@ -63,22 +64,20 @@ async fn main() {
},
};
// 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, …)
// 2. 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();
// 3. Run your scores on the desired topology & inventory
harmony_cli::run(
Inventory::autoload(), // auto-detect hardware / kube-config
K8sAnywhereTopology::from_env(), // local k3d, CI, staging, prod…
vec![
Box::new(lamp_stack),
Box::new(monitoring)
],
None
).await.unwrap();
}
```
@@ -94,13 +93,13 @@ Harmony analyses the code, shows an execution plan in a TUI, and applies it once
## 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. |
| 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.
@@ -112,9 +111,9 @@ A visual overview is in the diagram below.
Prerequisites:
* Rust
* Docker (if you deploy locally)
* `kubectl` / `helm` for Kubernetes-based topologies
- Rust
- Docker (if you deploy locally)
- `kubectl` / `helm` for Kubernetes-based topologies
```bash
git clone https://git.nationtech.io/nationtech/harmony
@@ -126,15 +125,15 @@ cargo build --release # builds the CLI, TUI and libraries
## 5 · Learning More
* **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)
- **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)
* **Extending Harmony** write new Scores / Interprets, add hardware like OPNsense firewalls, or embed Harmony in your own tooling (`/docs`).
- **Extending Harmony** write new Scores / Interprets, add hardware like OPNsense firewalls, or embed Harmony in your own tooling (`/docs`).
* **Community** discussions and roadmap live in [GitLab issues](https://git.nationtech.io/nationtech/harmony/-/issues). PRs, ideas, and feedback are welcome!
- **Community** discussions and roadmap live in [GitLab issues](https://git.nationtech.io/nationtech/harmony/-/issues). PRs, ideas, and feedback are welcome!
---
@@ -148,4 +147,4 @@ See [LICENSE](LICENSE) for the full text.
---
*Made with ❤️ & 🦀 by the NationTech and the Harmony community*
_Made with ❤️ & 🦀 by the NationTech and the Harmony community_

View File

@@ -1,5 +1,7 @@
#!/bin/sh
set -e
cargo check --all-targets --all-features --keep-going
cargo fmt --check
cargo clippy
cargo test

View File

@@ -0,0 +1,14 @@
[package]
name = "example-application-monitoring-with-tenant"
edition = "2024"
version.workspace = true
readme.workspace = true
license.workspace = true
[dependencies]
env_logger.workspace = true
harmony = { version = "0.1.0", path = "../../harmony" }
harmony_cli = { version = "0.1.0", path = "../../harmony_cli" }
logging = "0.1.0"
tokio.workspace = true
url.workspace = true

View File

@@ -0,0 +1,55 @@
use std::{path::PathBuf, str::FromStr, sync::Arc};
use harmony::{
data::Id,
inventory::Inventory,
modules::{
application::{ApplicationScore, RustWebFramework, RustWebapp, features::Monitoring},
monitoring::alert_channel::webhook_receiver::WebhookReceiver,
tenant::TenantScore,
},
topology::{K8sAnywhereTopology, Url, tenant::TenantConfig},
};
#[tokio::main]
async fn main() {
//TODO there is a bug where the application is deployed into the namespace matching the
//application name and the tenant is created in the namesapce matching the tenant name
//in order for the application to be deployed in the tenant namespace the application.name and
//the TenantConfig.name must match
let tenant = TenantScore {
config: TenantConfig {
id: Id::from_str("test-tenant-id").unwrap(),
name: "example-monitoring".to_string(),
..Default::default()
},
};
let application = Arc::new(RustWebapp {
name: "example-monitoring".to_string(),
domain: Url::Url(url::Url::parse("https://rustapp.harmony.example.com").unwrap()),
project_root: PathBuf::from("./examples/rust/webapp"),
framework: Some(RustWebFramework::Leptos),
});
let webhook_receiver = WebhookReceiver {
name: "sample-webhook-receiver".to_string(),
url: Url::Url(url::Url::parse("https://webhook-doesnt-exist.com").unwrap()),
};
let app = ApplicationScore {
features: vec![Box::new(Monitoring {
alert_receiver: vec![Box::new(webhook_receiver)],
application: application.clone(),
})],
application,
};
harmony_cli::run(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
vec![Box::new(tenant), Box::new(app)],
None,
)
.await
.unwrap();
}

View File

@@ -1,20 +1,21 @@
use harmony::{
inventory::Inventory,
maestro::Maestro,
modules::dummy::{ErrorScore, PanicScore, SuccessScore},
topology::LocalhostTopology,
};
#[tokio::main]
async fn main() {
let inventory = Inventory::autoload();
let topology = LocalhostTopology::new();
let mut maestro = Maestro::initialize(inventory, topology).await.unwrap();
maestro.register_all(vec![
Box::new(SuccessScore {}),
Box::new(ErrorScore {}),
Box::new(PanicScore {}),
]);
harmony_cli::init(maestro, None).await.unwrap();
harmony_cli::run(
Inventory::autoload(),
LocalhostTopology::new(),
vec![
Box::new(SuccessScore {}),
Box::new(ErrorScore {}),
Box::new(PanicScore {}),
],
None,
)
.await
.unwrap();
}

View File

@@ -125,40 +125,47 @@ spec:
name: nginx"#,
)
.unwrap();
return deployment;
deployment
}
fn nginx_deployment_2() -> Deployment {
let mut pod_template = PodTemplateSpec::default();
pod_template.metadata = Some(ObjectMeta {
labels: Some(BTreeMap::from([(
"app".to_string(),
"nginx-test".to_string(),
)])),
..Default::default()
});
pod_template.spec = Some(PodSpec {
containers: vec![Container {
name: "nginx".to_string(),
image: Some("nginx".to_string()),
let pod_template = PodTemplateSpec {
metadata: Some(ObjectMeta {
labels: Some(BTreeMap::from([(
"app".to_string(),
"nginx-test".to_string(),
)])),
..Default::default()
}],
..Default::default()
});
let mut spec = DeploymentSpec::default();
spec.template = pod_template;
spec.selector = LabelSelector {
match_expressions: None,
match_labels: Some(BTreeMap::from([(
"app".to_string(),
"nginx-test".to_string(),
)])),
}),
spec: Some(PodSpec {
containers: vec![Container {
name: "nginx".to_string(),
image: Some("nginx".to_string()),
..Default::default()
}],
..Default::default()
}),
};
let mut deployment = Deployment::default();
deployment.spec = Some(spec);
deployment.metadata.name = Some("nginx-test".to_string());
let spec = DeploymentSpec {
template: pod_template,
selector: LabelSelector {
match_expressions: None,
match_labels: Some(BTreeMap::from([(
"app".to_string(),
"nginx-test".to_string(),
)])),
},
..Default::default()
};
deployment
Deployment {
spec: Some(spec),
metadata: ObjectMeta {
name: Some("nginx-test".to_string()),
..Default::default()
},
..Default::default()
}
}
fn nginx_deployment() -> Deployment {

View File

@@ -1,7 +1,6 @@
use harmony::{
data::Version,
inventory::Inventory,
maestro::Maestro,
modules::lamp::{LAMPConfig, LAMPScore},
topology::{K8sAnywhereTopology, Url},
};
@@ -24,7 +23,7 @@ async fn main() {
// This config can be extended as needed for more complicated configurations
config: LAMPConfig {
project_root: "./php".into(),
database_size: format!("4Gi").into(),
database_size: "4Gi".to_string().into(),
..Default::default()
},
};
@@ -43,15 +42,13 @@ async fn main() {
// 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(
harmony_cli::run(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
vec![Box::new(lamp_stack)],
None,
)
.await
.unwrap();
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();
}
// That's it, end of the infra as code.

View File

@@ -2,7 +2,6 @@ use std::collections::HashMap;
use harmony::{
inventory::Inventory,
maestro::Maestro,
modules::{
monitoring::{
alert_channel::discord_alert_channel::DiscordWebhook,
@@ -51,8 +50,8 @@ async fn main() {
let service_monitor_endpoint = ServiceMonitorEndpoint {
port: Some("80".to_string()),
path: "/metrics".to_string(),
scheme: HTTPScheme::HTTP,
path: Some("/metrics".to_string()),
scheme: Some(HTTPScheme::HTTP),
..Default::default()
};
@@ -74,13 +73,13 @@ async fn main() {
rules: vec![Box::new(additional_rules), Box::new(additional_rules2)],
service_monitors: vec![service_monitor],
};
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
harmony_cli::run(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
vec![Box::new(alerting_score)],
None,
)
.await
.unwrap();
maestro.register_all(vec![Box::new(alerting_score)]);
harmony_cli::init(maestro, None).await.unwrap();
}

View File

@@ -1,9 +1,8 @@
use std::collections::HashMap;
use std::{collections::HashMap, str::FromStr};
use harmony::{
data::Id,
inventory::Inventory,
maestro::Maestro,
modules::{
monitoring::{
alert_channel::discord_alert_channel::DiscordWebhook,
@@ -29,7 +28,7 @@ use harmony::{
async fn main() {
let tenant = TenantScore {
config: TenantConfig {
id: Id::from_string("1234".to_string()),
id: Id::from_str("1234").unwrap(),
name: "test-tenant".to_string(),
resource_limits: ResourceLimits {
cpu_request_cores: 6.0,
@@ -54,8 +53,8 @@ async fn main() {
let service_monitor_endpoint = ServiceMonitorEndpoint {
port: Some("80".to_string()),
path: "/metrics".to_string(),
scheme: HTTPScheme::HTTP,
path: Some("/metrics".to_string()),
scheme: Some(HTTPScheme::HTTP),
..Default::default()
};
@@ -78,13 +77,13 @@ async fn main() {
rules: vec![Box::new(additional_rules)],
service_monitors: vec![service_monitor],
};
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
harmony_cli::run(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
vec![Box::new(tenant), Box::new(alerting_score)],
None,
)
.await
.unwrap();
maestro.register_all(vec![Box::new(tenant), Box::new(alerting_score)]);
harmony_cli::init(maestro, None).await.unwrap();
}

View File

@@ -10,7 +10,7 @@ use harmony::{
inventory::Inventory,
maestro::Maestro,
modules::{
http::HttpScore,
http::StaticFilesHttpScore,
ipxe::IpxeScore,
okd::{
bootstrap_dhcp::OKDBootstrapDhcpScore,
@@ -126,7 +126,7 @@ async fn main() {
harmony::modules::okd::load_balancer::OKDLoadBalancerScore::new(&topology);
let tftp_score = TftpScore::new(Url::LocalFolder("./data/watchguard/tftpboot".to_string()));
let http_score = HttpScore::new(Url::LocalFolder(
let http_score = StaticFilesHttpScore::new(Url::LocalFolder(
"./data/watchguard/pxe-http-files".to_string(),
));
let ipxe_score = IpxeScore::new();

View File

@@ -1,19 +1,18 @@
use harmony::{
inventory::Inventory, maestro::Maestro, modules::monitoring::ntfy::ntfy::NtfyScore,
topology::K8sAnywhereTopology,
inventory::Inventory, modules::monitoring::ntfy::ntfy::NtfyScore, topology::K8sAnywhereTopology,
};
#[tokio::main]
async fn main() {
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
harmony_cli::run(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
vec![Box::new(NtfyScore {
namespace: "monitoring".to_string(),
host: "localhost".to_string(),
})],
None,
)
.await
.unwrap();
maestro.register_all(vec![Box::new(NtfyScore {
namespace: "monitoring".to_string(),
})]);
harmony_cli::init(maestro, None).await.unwrap();
}

View File

@@ -11,7 +11,7 @@ use harmony::{
maestro::Maestro,
modules::{
dummy::{ErrorScore, PanicScore, SuccessScore},
http::HttpScore,
http::StaticFilesHttpScore,
okd::{dhcp::OKDDhcpScore, dns::OKDDnsScore, load_balancer::OKDLoadBalancerScore},
opnsense::OPNsenseShellCommandScore,
tftp::TftpScore,
@@ -81,7 +81,7 @@ async fn main() {
let load_balancer_score = OKDLoadBalancerScore::new(&topology);
let tftp_score = TftpScore::new(Url::LocalFolder("./data/watchguard/tftpboot".to_string()));
let http_score = HttpScore::new(Url::LocalFolder(
let http_score = StaticFilesHttpScore::new(Url::LocalFolder(
"./data/watchguard/pxe-http-files".to_string(),
));
let mut maestro = Maestro::initialize(inventory, topology).await.unwrap();

3
examples/rust/.gitignore vendored Normal file
View File

@@ -0,0 +1,3 @@
Dockerfile.harmony
.harmony_generated
harmony

View File

@@ -12,3 +12,4 @@ tokio = { workspace = true }
log = { workspace = true }
env_logger = { workspace = true }
url = { workspace = true }
base64.workspace = true

View File

@@ -2,35 +2,57 @@ use std::{path::PathBuf, sync::Arc};
use harmony::{
inventory::Inventory,
maestro::Maestro,
modules::application::{
RustWebFramework, RustWebapp, RustWebappScore, features::ContinuousDelivery,
modules::{
application::{
ApplicationScore, RustWebFramework, RustWebapp,
features::{ContinuousDelivery, Monitoring},
},
monitoring::alert_channel::{
discord_alert_channel::DiscordWebhook, webhook_receiver::WebhookReceiver,
},
},
topology::{K8sAnywhereTopology, Url},
};
#[tokio::main]
async fn main() {
env_logger::init();
let application = RustWebapp {
let application = Arc::new(RustWebapp {
name: "harmony-example-rust-webapp".to_string(),
project_root: PathBuf::from("./examples/rust/webapp"),
framework: Some(RustWebFramework::Leptos),
};
// TODO RustWebappScore should simply take a RustWebApp as config
let app = RustWebappScore {
name: "Example Rust Webapp".to_string(),
domain: Url::Url(url::Url::parse("https://rustapp.harmony.example.com").unwrap()),
features: vec![Box::new(ContinuousDelivery {
application: Arc::new(application.clone()),
})],
project_root: PathBuf::from("./webapp"), // Relative from 'harmony-path' param
framework: Some(RustWebFramework::Leptos),
});
let discord_receiver = DiscordWebhook {
name: "test-discord".to_string(),
url: Url::Url(url::Url::parse("https://discord.doesnt.exist.com").unwrap()),
};
let webhook_receiver = WebhookReceiver {
name: "sample-webhook-receiver".to_string(),
url: Url::Url(url::Url::parse("https://webhook-doesnt-exist.com").unwrap()),
};
let app = ApplicationScore {
features: vec![
Box::new(ContinuousDelivery {
application: application.clone(),
}),
Box::new(Monitoring {
application: application.clone(),
alert_receiver: vec![Box::new(discord_receiver), Box::new(webhook_receiver)],
}),
// TODO add backups, multisite ha, etc
],
application,
};
let topology = K8sAnywhereTopology::from_env();
let mut maestro = Maestro::initialize(Inventory::autoload(), topology)
.await
.unwrap();
maestro.register_all(vec![Box::new(app)]);
harmony_cli::init(maestro, None).await.unwrap();
harmony_cli::run(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
vec![Box::new(app)],
None,
)
.await
.unwrap();
}

View File

@@ -1,16 +0,0 @@
FROM rust:bookworm as builder
RUN apt-get update && apt-get install -y --no-install-recommends clang wget && wget https://github.com/cargo-bins/cargo-binstall/releases/latest/download/cargo-binstall-x86_64-unknown-linux-musl.tgz && tar -xvf cargo-binstall-x86_64-unknown-linux-musl.tgz && cp cargo-binstall /usr/local/cargo/bin && rm cargo-binstall-x86_64-unknown-linux-musl.tgz cargo-binstall && apt-get clean && rm -rf /var/lib/apt/lists/*
RUN cargo binstall cargo-leptos -y
RUN rustup target add wasm32-unknown-unknown
WORKDIR /app
COPY . .
RUN cargo leptos build --release -vv
FROM debian:bookworm-slim
RUN groupadd -r appgroup && useradd -r -s /bin/false -g appgroup appuser
ENV LEPTOS_SITE_ADDR=0.0.0.0:3000
EXPOSE 3000/tcp
WORKDIR /home/appuser
COPY --from=builder /app/target/site/pkg /home/appuser/pkg
COPY --from=builder /app/target/release/harmony-example-rust-webapp /home/appuser/harmony-example-rust-webapp
USER appuser
CMD /home/appuser/harmony-example-rust-webapp

View File

@@ -1,7 +1,8 @@
use std::str::FromStr;
use harmony::{
data::Id,
inventory::Inventory,
maestro::Maestro,
modules::tenant::TenantScore,
topology::{K8sAnywhereTopology, tenant::TenantConfig},
};
@@ -10,21 +11,20 @@ use harmony::{
async fn main() {
let tenant = TenantScore {
config: TenantConfig {
id: Id::from_str("test-tenant-id"),
id: Id::from_str("test-tenant-id").unwrap(),
name: "testtenant".to_string(),
..Default::default()
},
};
let mut maestro = Maestro::<K8sAnywhereTopology>::initialize(
harmony_cli::run(
Inventory::autoload(),
K8sAnywhereTopology::from_env(),
vec![Box::new(tenant)],
None,
)
.await
.unwrap();
maestro.register_all(vec![Box::new(tenant)]);
harmony_cli::init(maestro, None).await.unwrap();
}
// TODO write tests

View File

@@ -13,7 +13,7 @@ reqwest = { version = "0.11", features = ["blocking", "json"] }
russh = "0.45.0"
rust-ipmi = "0.1.1"
semver = "1.0.23"
serde = { version = "1.0.209", features = ["derive"] }
serde = { version = "1.0.209", features = ["derive", "rc"] }
serde_json = "1.0.127"
tokio.workspace = true
derive-new.workspace = true
@@ -27,12 +27,11 @@ harmony_macros = { path = "../harmony_macros" }
harmony_types = { path = "../harmony_types" }
uuid.workspace = true
url.workspace = true
kube.workspace = true
kube = { workspace = true, features = ["derive"] }
k8s-openapi.workspace = true
serde_yaml.workspace = true
http.workspace = true
serde-value.workspace = true
inquire.workspace = true
helm-wrapper-rs = "0.4.0"
non-blank-string-rs = "1.0.4"
k3d-rs = { path = "../k3d" }
@@ -44,12 +43,12 @@ 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",
"domain-name-without-special-chars",
"domain-name-length-limited-to-255",
"punycode",
"serde",
"domain-label-cannot-start-or-end-with-hyphen",
"domain-label-length-limited-to-63",
"domain-name-without-special-chars",
"domain-name-length-limited-to-255",
"punycode",
"serde",
] }
temp-dir = "0.1.14"
dyn-clone = "1.0.19"
@@ -58,3 +57,13 @@ futures-util = "0.3.31"
tokio-util = "0.7.15"
strum = { version = "0.27.1", features = ["derive"] }
tempfile = "3.20.0"
serde_with = "3.14.0"
schemars = "0.8.22"
kube-derive = "1.1.0"
bollard.workspace = true
tar.workspace = true
base64.workspace = true
once_cell = "1.21.3"
[dev-dependencies]
pretty_assertions.workspace = true

View File

@@ -11,5 +11,5 @@ lazy_static! {
pub static ref REGISTRY_PROJECT: String =
std::env::var("HARMONY_REGISTRY_PROJECT").unwrap_or_else(|_| "harmony".to_string());
pub static ref DRY_RUN: bool =
std::env::var("HARMONY_DRY_RUN").map_or(true, |value| value.parse().unwrap_or(true));
std::env::var("HARMONY_DRY_RUN").is_ok_and(|value| value.parse().unwrap_or(false));
}

View File

@@ -1,5 +1,6 @@
use rand::distr::Alphanumeric;
use rand::distr::SampleString;
use std::str::FromStr;
use std::time::SystemTime;
use std::time::UNIX_EPOCH;
@@ -23,13 +24,13 @@ pub struct Id {
value: String,
}
impl Id {
pub fn from_string(value: String) -> Self {
Self { value }
}
impl FromStr for Id {
type Err = ();
pub fn from_str(value: &str) -> Self {
Self::from_string(value.to_string())
fn from_str(s: &str) -> Result<Self, Self::Err> {
Ok(Id {
value: s.to_string(),
})
}
}

View File

@@ -47,7 +47,7 @@ impl serde::Serialize for Version {
impl std::fmt::Display for Version {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
return self.value.fmt(f);
self.value.fmt(f)
}
}

View File

@@ -35,10 +35,9 @@ impl PhysicalHost {
pub fn cluster_mac(&self) -> MacAddress {
self.network
.get(0)
.first()
.expect("Cluster physical host should have a network interface")
.mac_address
.clone()
}
pub fn cpu(mut self, cpu_count: Option<u64>) -> Self {

View File

@@ -0,0 +1,69 @@
use log::debug;
use once_cell::sync::Lazy;
use tokio::sync::broadcast;
use super::{
interpret::{InterpretError, Outcome},
topology::TopologyStatus,
};
#[derive(Debug, Clone)]
pub enum HarmonyEvent {
HarmonyStarted,
HarmonyFinished,
InterpretExecutionStarted {
execution_id: String,
topology: String,
interpret: String,
score: String,
message: String,
},
InterpretExecutionFinished {
execution_id: String,
topology: String,
interpret: String,
score: String,
outcome: Result<Outcome, InterpretError>,
},
TopologyStateChanged {
topology: String,
status: TopologyStatus,
message: Option<String>,
},
}
static HARMONY_EVENT_BUS: Lazy<broadcast::Sender<HarmonyEvent>> = Lazy::new(|| {
// TODO: Adjust channel capacity
let (tx, _rx) = broadcast::channel(100);
tx
});
pub fn instrument(event: HarmonyEvent) -> Result<(), &'static str> {
match HARMONY_EVENT_BUS.send(event) {
Ok(_) => Ok(()),
Err(_) => Err("send error: no subscribers"),
}
}
pub async fn subscribe<F, Fut>(name: &str, mut handler: F)
where
F: FnMut(HarmonyEvent) -> Fut + Send + 'static,
Fut: Future<Output = bool> + Send,
{
let mut rx = HARMONY_EVENT_BUS.subscribe();
debug!("[{name}] Service started. Listening for events...");
loop {
match rx.recv().await {
Ok(event) => {
if !handler(event).await {
debug!("[{name}] Handler requested exit.");
break;
}
}
Err(broadcast::error::RecvError::Lagged(n)) => {
debug!("[{name}] Lagged behind by {n} messages.");
}
Err(_) => break,
}
}
}

View File

@@ -7,6 +7,7 @@ use super::{
data::{Id, Version},
executors::ExecutorError,
inventory::Inventory,
topology::PreparationError,
};
pub enum InterpretName {
@@ -22,6 +23,15 @@ pub enum InterpretName {
K3dInstallation,
TenantInterpret,
Application,
ArgoCD,
Alerting,
Ntfy,
HelmChart,
HelmCommand,
K8sResource,
Lamp,
ApplicationMonitoring,
K8sPrometheusCrdAlerting,
}
impl std::fmt::Display for InterpretName {
@@ -39,6 +49,15 @@ impl std::fmt::Display for InterpretName {
InterpretName::K3dInstallation => f.write_str("K3dInstallation"),
InterpretName::TenantInterpret => f.write_str("Tenant"),
InterpretName::Application => f.write_str("Application"),
InterpretName::ArgoCD => f.write_str("ArgoCD"),
InterpretName::Alerting => f.write_str("Alerting"),
InterpretName::Ntfy => f.write_str("Ntfy"),
InterpretName::HelmChart => f.write_str("HelmChart"),
InterpretName::HelmCommand => f.write_str("HelmCommand"),
InterpretName::K8sResource => f.write_str("K8sResource"),
InterpretName::Lamp => f.write_str("LAMP"),
InterpretName::ApplicationMonitoring => f.write_str("ApplicationMonitoring"),
InterpretName::K8sPrometheusCrdAlerting => f.write_str("K8sPrometheusCrdAlerting"),
}
}
}
@@ -111,6 +130,14 @@ impl std::fmt::Display for InterpretError {
}
impl Error for InterpretError {}
impl From<PreparationError> for InterpretError {
fn from(value: PreparationError) -> Self {
Self {
msg: format!("InterpretError : {value}"),
}
}
}
impl From<ExecutorError> for InterpretError {
fn from(value: ExecutorError) -> Self {
Self {

View File

@@ -1,12 +1,14 @@
use std::sync::{Arc, Mutex, RwLock};
use std::sync::{Arc, RwLock};
use log::{info, warn};
use log::{debug, warn};
use crate::topology::TopologyStatus;
use super::{
interpret::{InterpretError, InterpretStatus, Outcome},
interpret::{InterpretError, Outcome},
inventory::Inventory,
score::Score,
topology::Topology,
topology::{PreparationError, PreparationOutcome, Topology, TopologyState},
};
type ScoreVec<T> = Vec<Box<dyn Score<T>>>;
@@ -15,7 +17,7 @@ pub struct Maestro<T: Topology> {
inventory: Inventory,
topology: T,
scores: Arc<RwLock<ScoreVec<T>>>,
topology_preparation_result: Mutex<Option<Outcome>>,
topology_state: TopologyState,
}
impl<T: Topology> Maestro<T> {
@@ -23,36 +25,46 @@ impl<T: Topology> Maestro<T> {
///
/// This should rarely be used. Most of the time Maestro::initialize should be used instead.
pub fn new_without_initialization(inventory: Inventory, topology: T) -> Self {
let topology_name = topology.name().to_string();
Self {
inventory,
topology,
scores: Arc::new(RwLock::new(Vec::new())),
topology_preparation_result: None.into(),
topology_state: TopologyState::new(topology_name),
}
}
pub async fn initialize(inventory: Inventory, topology: T) -> Result<Self, InterpretError> {
let instance = Self::new_without_initialization(inventory, topology);
pub async fn initialize(inventory: Inventory, topology: T) -> Result<Self, PreparationError> {
let mut instance = Self::new_without_initialization(inventory, topology);
instance.prepare_topology().await?;
Ok(instance)
}
/// Ensures the associated Topology is ready for operations.
/// Delegates the readiness check and potential setup actions to the Topology.
pub async fn prepare_topology(&self) -> Result<Outcome, InterpretError> {
info!("Ensuring topology '{}' is ready...", self.topology.name());
let outcome = self.topology.ensure_ready().await?;
info!(
"Topology '{}' readiness check complete: {}",
self.topology.name(),
outcome.status
);
async fn prepare_topology(&mut self) -> Result<PreparationOutcome, PreparationError> {
self.topology_state.prepare();
self.topology_preparation_result
.lock()
.unwrap()
.replace(outcome.clone());
Ok(outcome)
let result = self.topology.ensure_ready().await;
match result {
Ok(outcome) => {
match outcome.clone() {
PreparationOutcome::Success { details } => {
self.topology_state.success(details);
}
PreparationOutcome::Noop => {
self.topology_state.noop();
}
};
Ok(outcome)
}
Err(err) => {
self.topology_state.error(err.to_string());
Err(err)
}
}
}
pub fn register_all(&mut self, mut scores: ScoreVec<T>) {
@@ -61,15 +73,7 @@ impl<T: Topology> Maestro<T> {
}
fn is_topology_initialized(&self) -> bool {
let result = self.topology_preparation_result.lock().unwrap();
if let Some(outcome) = result.as_ref() {
match outcome.status {
InterpretStatus::SUCCESS => return true,
_ => return false,
}
} else {
false
}
self.topology_state.status == TopologyStatus::Success
}
pub async fn interpret(&self, score: Box<dyn Score<T>>) -> Result<Outcome, InterpretError> {
@@ -80,11 +84,9 @@ impl<T: Topology> Maestro<T> {
self.topology.name(),
);
}
info!("Running score {score:?}");
let interpret = score.create_interpret();
info!("Launching interpret {interpret:?}");
let result = interpret.execute(&self.inventory, &self.topology).await;
info!("Got result {result:?}");
debug!("Interpreting score {score:?}");
let result = score.interpret(&self.inventory, &self.topology).await;
debug!("Got result {result:?}");
result
}

View File

@@ -3,6 +3,7 @@ pub mod data;
pub mod executors;
pub mod filter;
pub mod hardware;
pub mod instrumentation;
pub mod interpret;
pub mod inventory;
pub mod maestro;

View File

@@ -1,22 +1,62 @@
use std::collections::BTreeMap;
use async_trait::async_trait;
use serde::Serialize;
use serde_value::Value;
use super::{interpret::Interpret, topology::Topology};
use super::{
data::Id,
instrumentation::{self, HarmonyEvent},
interpret::{Interpret, InterpretError, Outcome},
inventory::Inventory,
topology::Topology,
};
#[async_trait]
pub trait Score<T: Topology>:
std::fmt::Debug + ScoreToString<T> + Send + Sync + CloneBoxScore<T> + SerializeScore<T>
{
fn create_interpret(&self) -> Box<dyn Interpret<T>>;
async fn interpret(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
let id = Id::default();
let interpret = self.create_interpret();
instrumentation::instrument(HarmonyEvent::InterpretExecutionStarted {
execution_id: id.clone().to_string(),
topology: topology.name().into(),
interpret: interpret.get_name().to_string(),
score: self.name(),
message: format!("{} running...", interpret.get_name()),
})
.unwrap();
let result = interpret.execute(inventory, topology).await;
instrumentation::instrument(HarmonyEvent::InterpretExecutionFinished {
execution_id: id.clone().to_string(),
topology: topology.name().into(),
interpret: interpret.get_name().to_string(),
score: self.name(),
outcome: result.clone(),
})
.unwrap();
result
}
fn name(&self) -> String;
#[doc(hidden)]
fn create_interpret(&self) -> Box<dyn Interpret<T>>;
}
pub trait SerializeScore<T: Topology> {
fn serialize(&self) -> Value;
}
impl<'de, S, T> SerializeScore<T> for S
impl<S, T> SerializeScore<T> for S
where
T: Topology,
S: Score<T> + Serialize,
@@ -24,7 +64,7 @@ where
fn serialize(&self) -> Value {
// TODO not sure if this is the right place to handle the error or it should bubble
// up?
serde_value::to_value(&self).expect("Score should serialize successfully")
serde_value::to_value(self).expect("Score should serialize successfully")
}
}

View File

@@ -0,0 +1,59 @@
////////////////////
/// Working idea
///
///
trait ScoreWithDep<T> {
fn create_interpret(&self) -> Box<dyn Interpret<T>>;
fn name(&self) -> String;
fn get_dependencies(&self) -> Vec<TypeId>; // Force T to impl Installer<TypeId> or something
// like that
}
struct PrometheusAlertScore;
impl <T> ScoreWithDep<T> for PrometheusAlertScore {
fn create_interpret(&self) -> Box<dyn Interpret<T>> {
todo!()
}
fn name(&self) -> String {
todo!()
}
fn get_dependencies(&self) -> Vec<TypeId> {
// We have to find a way to constrait here so at compile time we are only allowed to return
// TypeId for types which can be installed by T
//
// This means, for example that T must implement HelmCommand if the impl <T: HelmCommand> Installable<T> for
// KubePrometheus calls for HelmCommand.
vec![TypeId::of::<KubePrometheus>()]
}
}
trait Installable{}
struct KubePrometheus;
impl Installable for KubePrometheus;
struct Maestro<T> {
topology: T
}
impl <T>Maestro<T> {
fn execute_store(&self, score: ScoreWithDep<T>) {
score.get_dependencies().iter().for_each(|dep| {
self.topology.ensure_dependency_ready(dep);
});
}
}
struct TopologyWithDep {
}
impl TopologyWithDep {
fn ensure_dependency_ready(&self, type_id: TypeId) -> Result<(), String> {
self.installer
}
}

View File

@@ -4,8 +4,6 @@ use harmony_types::net::MacAddress;
use log::info;
use crate::executors::ExecutorError;
use crate::interpret::InterpretError;
use crate::interpret::Outcome;
use super::DHCPStaticEntry;
use super::DhcpServer;
@@ -19,6 +17,8 @@ use super::K8sclient;
use super::LoadBalancer;
use super::LoadBalancerService;
use super::LogicalHost;
use super::PreparationError;
use super::PreparationOutcome;
use super::Router;
use super::TftpServer;
@@ -48,7 +48,7 @@ impl Topology for HAClusterTopology {
fn name(&self) -> &str {
"HAClusterTopology"
}
async fn ensure_ready(&self) -> Result<Outcome, InterpretError> {
async fn ensure_ready(&self) -> Result<PreparationOutcome, PreparationError> {
todo!(
"ensure_ready, not entirely sure what it should do here, probably something like verify that the hosts are reachable and all services are up and ready."
)
@@ -244,10 +244,12 @@ impl Topology for DummyInfra {
todo!()
}
async fn ensure_ready(&self) -> Result<Outcome, InterpretError> {
async fn ensure_ready(&self) -> Result<PreparationOutcome, PreparationError> {
let dummy_msg = "This is a dummy infrastructure that does nothing";
info!("{dummy_msg}");
Ok(Outcome::success(dummy_msg.to_string()))
Ok(PreparationOutcome::Success {
details: dummy_msg.into(),
})
}
}

View File

@@ -1,11 +1,8 @@
use derive_new::new;
use futures_util::StreamExt;
use k8s_openapi::{
ClusterResourceScope, NamespaceResourceScope,
api::{apps::v1::Deployment, core::v1::Pod},
};
use kube::runtime::conditions;
use kube::runtime::wait::await_condition;
use kube::{
Client, Config, Error, Resource,
api::{Api, AttachParams, ListParams, Patch, PatchParams, ResourceExt},
@@ -13,15 +10,29 @@ use kube::{
core::ErrorResponse,
runtime::reflector::Lookup,
};
use kube::{api::DynamicObject, runtime::conditions};
use kube::{
api::{ApiResource, GroupVersionKind},
runtime::wait::await_condition,
};
use log::{debug, error, trace};
use serde::de::DeserializeOwned;
use similar::{DiffableStr, TextDiff};
use serde::{Serialize, de::DeserializeOwned};
use similar::TextDiff;
#[derive(new, Clone)]
pub struct K8sClient {
client: Client,
}
impl Serialize for K8sClient {
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{
todo!()
}
}
impl std::fmt::Debug for K8sClient {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
// This is a poor man's debug implementation for now as kube::Client does not provide much
@@ -55,13 +66,13 @@ impl K8sClient {
}
let establish = await_condition(api, name.as_str(), conditions::is_deployment_completed());
let t = if let Some(t) = timeout { t } else { 300 };
let t = timeout.unwrap_or(300);
let res = tokio::time::timeout(std::time::Duration::from_secs(t), establish).await;
if let Ok(r) = res {
return Ok(());
if res.is_ok() {
Ok(())
} else {
return Err("timed out while waiting for deployment".to_string());
Err("timed out while waiting for deployment".to_string())
}
}
@@ -100,7 +111,7 @@ impl K8sClient {
.await;
match res {
Err(e) => return Err(e.to_string()),
Err(e) => Err(e.to_string()),
Ok(mut process) => {
let status = process
.take_status()
@@ -110,13 +121,9 @@ impl K8sClient {
if let Some(s) = status.status {
debug!("Status: {}", s);
if s == "Success" {
return Ok(());
} else {
return Err(s);
}
if s == "Success" { Ok(()) } else { Err(s) }
} else {
return Err("Couldn't get inner status of pod exec".to_string());
Err("Couldn't get inner status of pod exec".to_string())
}
}
}
@@ -157,8 +164,9 @@ impl K8sClient {
trace!("Received current value {current:#?}");
// The resource exists, so we calculate and display a diff.
println!("\nPerforming dry-run for resource: '{}'", name);
let mut current_yaml = serde_yaml::to_value(&current)
.expect(&format!("Could not serialize current value : {current:#?}"));
let mut current_yaml = serde_yaml::to_value(&current).unwrap_or_else(|_| {
panic!("Could not serialize current value : {current:#?}")
});
if current_yaml.is_mapping() && current_yaml.get("status").is_some() {
let map = current_yaml.as_mapping_mut().unwrap();
let removed = map.remove_entry("status");
@@ -225,7 +233,7 @@ impl K8sClient {
}
}
pub async fn apply_many<K>(&self, resource: &Vec<K>, ns: Option<&str>) -> Result<Vec<K>, Error>
pub async fn apply_many<K>(&self, resource: &[K], ns: Option<&str>) -> Result<Vec<K>, Error>
where
K: Resource + Clone + std::fmt::Debug + DeserializeOwned + serde::Serialize,
<K as Resource>::Scope: ApplyStrategy<K>,
@@ -239,6 +247,70 @@ impl K8sClient {
Ok(result)
}
pub async fn apply_yaml_many(
&self,
#[allow(clippy::ptr_arg)] yaml: &Vec<serde_yaml::Value>,
ns: Option<&str>,
) -> Result<(), Error> {
for y in yaml.iter() {
self.apply_yaml(y, ns).await?;
}
Ok(())
}
pub async fn apply_yaml(
&self,
yaml: &serde_yaml::Value,
ns: Option<&str>,
) -> Result<(), Error> {
let obj: DynamicObject = serde_yaml::from_value(yaml.clone()).expect("TODO do not unwrap");
let name = obj.metadata.name.as_ref().expect("YAML must have a name");
let api_version = yaml
.get("apiVersion")
.expect("couldn't get apiVersion from YAML")
.as_str()
.expect("couldn't get apiVersion as str");
let kind = yaml
.get("kind")
.expect("couldn't get kind from YAML")
.as_str()
.expect("couldn't get kind as str");
let split: Vec<&str> = api_version.splitn(2, "/").collect();
let g = split[0];
let v = split[1];
let gvk = GroupVersionKind::gvk(g, v, kind);
let api_resource = ApiResource::from_gvk(&gvk);
let namespace = match ns {
Some(n) => n,
None => obj
.metadata
.namespace
.as_ref()
.expect("YAML must have a namespace"),
};
// 5. Create a dynamic API client for this resource type.
let api: Api<DynamicObject> =
Api::namespaced_with(self.client.clone(), namespace, &api_resource);
// 6. Apply the object to the cluster using Server-Side Apply.
// This will create the resource if it doesn't exist, or update it if it does.
println!(
"Applying Argo Application '{}' in namespace '{}'...",
name, namespace
);
let patch_params = PatchParams::apply("harmony"); // Use a unique field manager name
let result = api.patch(name, &patch_params, &Patch::Apply(&obj)).await?;
println!("Successfully applied '{}'.", result.name_any());
Ok(())
}
pub(crate) async fn from_kubeconfig(path: &str) -> Option<K8sClient> {
let k = match Kubeconfig::read_from(path) {
Ok(k) => k,

View File

@@ -1,30 +1,40 @@
use std::{process::Command, sync::Arc};
use async_trait::async_trait;
use inquire::Confirm;
use log::{debug, info, warn};
use serde::Serialize;
use tokio::sync::OnceCell;
use crate::{
executors::ExecutorError,
interpret::{InterpretError, Outcome},
interpret::InterpretStatus,
inventory::Inventory,
maestro::Maestro,
modules::k3d::K3DInstallationScore,
topology::LocalhostTopology,
modules::{
k3d::K3DInstallationScore,
monitoring::kube_prometheus::crd::{
crd_alertmanager_config::CRDPrometheus,
prometheus_operator::prometheus_operator_helm_chart_score,
},
prometheus::{
k8s_prometheus_alerting_score::K8sPrometheusCRDAlertingScore,
prometheus::PrometheusApplicationMonitoring,
},
},
score::Score,
};
use super::{
DeploymentTarget, HelmCommand, K8sclient, MultiTargetTopology, Topology,
DeploymentTarget, HelmCommand, K8sclient, MultiTargetTopology, PreparationError,
PreparationOutcome, Topology,
k8s::K8sClient,
oberservability::monitoring::AlertReceiver,
tenant::{TenantConfig, TenantManager, k8s::K8sTenantManager},
};
#[derive(Clone, Debug)]
struct K8sState {
client: Arc<K8sClient>,
_source: K8sSource,
source: K8sSource,
message: String,
}
@@ -58,8 +68,42 @@ impl K8sclient for K8sAnywhereTopology {
}
}
#[async_trait]
impl PrometheusApplicationMonitoring<CRDPrometheus> for K8sAnywhereTopology {
async fn install_prometheus(
&self,
sender: &CRDPrometheus,
inventory: &Inventory,
receivers: Option<Vec<Box<dyn AlertReceiver<CRDPrometheus>>>>,
) -> Result<PreparationOutcome, PreparationError> {
let po_result = self.ensure_prometheus_operator(sender).await?;
if po_result == PreparationOutcome::Noop {
debug!("Skipping Prometheus CR installation due to missing operator.");
return Ok(po_result);
}
let result = self
.get_k8s_prometheus_application_score(sender.clone(), receivers)
.await
.interpret(inventory, self)
.await;
match result {
Ok(outcome) => match outcome.status {
InterpretStatus::SUCCESS => Ok(PreparationOutcome::Success {
details: outcome.message,
}),
InterpretStatus::NOOP => Ok(PreparationOutcome::Noop),
_ => Err(PreparationError::new(outcome.message)),
},
Err(err) => Err(PreparationError::new(err.to_string())),
}
}
}
impl Serialize for K8sAnywhereTopology {
fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{
@@ -84,6 +128,19 @@ impl K8sAnywhereTopology {
}
}
async fn get_k8s_prometheus_application_score(
&self,
sender: CRDPrometheus,
receivers: Option<Vec<Box<dyn AlertReceiver<CRDPrometheus>>>>,
) -> K8sPrometheusCRDAlertingScore {
K8sPrometheusCRDAlertingScore {
sender,
receivers: receivers.unwrap_or_default(),
service_monitors: vec![],
prometheus_rules: vec![],
}
}
fn is_helm_available(&self) -> Result<(), String> {
let version_result = Command::new("helm")
.arg("version")
@@ -94,9 +151,8 @@ impl K8sAnywhereTopology {
return Err("Failed to run 'helm -version'".to_string());
}
// Print the version output
let version_output = String::from_utf8_lossy(&version_result.stdout);
println!("Helm version: {}", version_output.trim());
debug!("Helm version: {}", version_output.trim());
Ok(())
}
@@ -113,33 +169,42 @@ impl K8sAnywhereTopology {
K3DInstallationScore::default()
}
async fn try_install_k3d(&self) -> Result<(), InterpretError> {
let maestro = Maestro::initialize(Inventory::autoload(), LocalhostTopology::new()).await?;
let k3d_score = self.get_k3d_installation_score();
maestro.interpret(Box::new(k3d_score)).await?;
Ok(())
async fn try_install_k3d(&self) -> Result<(), PreparationError> {
let result = self
.get_k3d_installation_score()
.interpret(&Inventory::empty(), self)
.await;
match result {
Ok(outcome) => match outcome.status {
InterpretStatus::SUCCESS => Ok(()),
InterpretStatus::NOOP => Ok(()),
_ => Err(PreparationError::new(outcome.message)),
},
Err(err) => Err(PreparationError::new(err.to_string())),
}
}
async fn try_get_or_install_k8s_client(&self) -> Result<Option<K8sState>, InterpretError> {
async fn try_get_or_install_k8s_client(&self) -> Result<Option<K8sState>, PreparationError> {
let k8s_anywhere_config = &self.config;
// TODO this deserves some refactoring, it is becoming a bit hard to figure out
// be careful when making modifications here
if k8s_anywhere_config.use_local_k3d {
info!("Using local k3d cluster because of use_local_k3d set to true");
debug!("Using local k3d cluster because of use_local_k3d set to true");
} else {
if let Some(kubeconfig) = &k8s_anywhere_config.kubeconfig {
debug!("Loading kubeconfig {kubeconfig}");
match self.try_load_kubeconfig(&kubeconfig).await {
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}"),
}));
}
None => {
return Err(InterpretError::new(format!(
return Err(PreparationError::new(format!(
"Failed to load kubeconfig from {kubeconfig}"
)));
}
@@ -158,22 +223,13 @@ impl K8sAnywhereTopology {
}
if !k8s_anywhere_config.autoinstall {
debug!("Autoinstall confirmation prompt");
let confirmation = Confirm::new( "Harmony autoinstallation is not activated, do you wish to launch autoinstallation? : ")
.with_default(false)
.prompt()
.expect("Unexpected prompt error");
debug!("Autoinstall confirmation {confirmation}");
if !confirmation {
warn!(
"Installation cancelled, K8sAnywhere could not initialize a valid Kubernetes client"
);
return Ok(None);
}
warn!(
"Installation cancelled, K8sAnywhere could not initialize a valid Kubernetes client"
);
return Ok(None);
}
info!("Starting K8sAnywhere installation");
debug!("Starting K8sAnywhere installation");
self.try_install_k3d().await?;
let k3d_score = self.get_k3d_installation_score();
// I feel like having to rely on the k3d_rs crate here is a smell
@@ -185,8 +241,8 @@ impl K8sAnywhereTopology {
let state = match k3d.get_client().await {
Ok(client) => K8sState {
client: Arc::new(K8sClient::new(client)),
_source: K8sSource::LocalK3d,
message: "Successfully installed K3D cluster and acquired client".to_string(),
source: K8sSource::LocalK3d,
message: "K8s client ready".to_string(),
},
Err(_) => todo!(),
};
@@ -195,12 +251,13 @@ impl K8sAnywhereTopology {
}
async fn ensure_k8s_tenant_manager(&self) -> Result<(), String> {
if let Some(_) = self.tenant_manager.get() {
if self.tenant_manager.get().is_some() {
return Ok(());
}
self.tenant_manager
.get_or_try_init(async || -> Result<K8sTenantManager, String> {
// TOOD: checker si K8s ou K3d/s tenant manager (ref. issue https://git.nationtech.io/NationTech/harmony/issues/94)
let k8s_client = self.k8s_client().await?;
Ok(K8sTenantManager::new(k8s_client))
})
@@ -217,6 +274,55 @@ impl K8sAnywhereTopology {
)),
}
}
async fn ensure_prometheus_operator(
&self,
sender: &CRDPrometheus,
) -> Result<PreparationOutcome, PreparationError> {
let status = Command::new("sh")
.args(["-c", "kubectl get crd -A | grep -i prometheuses"])
.status()
.map_err(|e| PreparationError::new(format!("could not connect to cluster: {}", e)))?;
if !status.success() {
if let Some(Some(k8s_state)) = self.k8s_state.get() {
match k8s_state.source {
K8sSource::LocalK3d => {
debug!("installing prometheus operator");
let op_score =
prometheus_operator_helm_chart_score(sender.namespace.clone());
let result = op_score.interpret(&Inventory::empty(), self).await;
return match result {
Ok(outcome) => match outcome.status {
InterpretStatus::SUCCESS => Ok(PreparationOutcome::Success {
details: "installed prometheus operator".into(),
}),
InterpretStatus::NOOP => Ok(PreparationOutcome::Noop),
_ => Err(PreparationError::new(
"failed to install prometheus operator (unknown error)".into(),
)),
},
Err(err) => Err(PreparationError::new(err.to_string())),
};
}
K8sSource::Kubeconfig => {
debug!("unable to install prometheus operator, contact cluster admin");
return Ok(PreparationOutcome::Noop);
}
}
} else {
warn!("Unable to detect k8s_state. Skipping Prometheus Operator install.");
return Ok(PreparationOutcome::Noop);
}
}
debug!("Prometheus operator is already present, skipping install");
Ok(PreparationOutcome::Success {
details: "prometheus operator present in cluster".into(),
})
}
}
#[derive(Clone, Debug)]
@@ -237,7 +343,7 @@ pub struct K8sAnywhereConfig {
///
/// When enabled, autoinstall will setup a K3D cluster on the localhost. https://k3d.io/stable/
///
/// Default: false
/// Default: true
pub autoinstall: bool,
/// Whether to use local k3d cluster.
@@ -246,7 +352,7 @@ pub struct K8sAnywhereConfig {
///
/// default: true
pub use_local_k3d: bool,
harmony_profile: String,
pub harmony_profile: String,
}
impl K8sAnywhereConfig {
@@ -256,7 +362,7 @@ impl K8sAnywhereConfig {
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)),
.map_or_else(|_| true, |v| v.parse().ok().unwrap_or(false)),
// TODO harmony_profile should be managed at a more core level than this
harmony_profile: std::env::var("HARMONY_PROFILE").map_or_else(
|_| "dev".to_string(),
@@ -274,26 +380,25 @@ impl Topology for K8sAnywhereTopology {
"K8sAnywhereTopology"
}
async fn ensure_ready(&self) -> Result<Outcome, InterpretError> {
async fn ensure_ready(&self) -> Result<PreparationOutcome, PreparationError> {
let k8s_state = self
.k8s_state
.get_or_try_init(|| self.try_get_or_install_k8s_client())
.await?;
let k8s_state: &K8sState = k8s_state.as_ref().ok_or(InterpretError::new(
"No K8s client could be found or installed".to_string(),
let k8s_state: &K8sState = k8s_state.as_ref().ok_or(PreparationError::new(
"no K8s client could be found or installed".to_string(),
))?;
self.ensure_k8s_tenant_manager()
.await
.map_err(|e| InterpretError::new(e))?;
.map_err(PreparationError::new)?;
match self.is_helm_available() {
Ok(()) => Ok(Outcome::success(format!(
"{} + helm available",
k8s_state.message.clone()
))),
Err(e) => Err(InterpretError::new(format!("helm unavailable: {}", e))),
Ok(()) => Ok(PreparationOutcome::Success {
details: format!("{} + helm available", k8s_state.message.clone()),
}),
Err(e) => Err(PreparationError::new(format!("helm unavailable: {}", e))),
}
}
}

View File

@@ -1,9 +1,7 @@
use async_trait::async_trait;
use derive_new::new;
use crate::interpret::{InterpretError, Outcome};
use super::{HelmCommand, Topology};
use super::{HelmCommand, PreparationError, PreparationOutcome, Topology};
#[derive(new)]
pub struct LocalhostTopology;
@@ -14,10 +12,10 @@ impl Topology for LocalhostTopology {
"LocalHostTopology"
}
async fn ensure_ready(&self) -> Result<Outcome, InterpretError> {
Ok(Outcome::success(
"Localhost is Chuck Norris, always ready.".to_string(),
))
async fn ensure_ready(&self) -> Result<PreparationOutcome, PreparationError> {
Ok(PreparationOutcome::Success {
details: "Localhost is Chuck Norris, always ready.".into(),
})
}
}

View File

@@ -6,6 +6,7 @@ mod k8s_anywhere;
mod localhost;
pub mod oberservability;
pub mod tenant;
use derive_new::new;
pub use k8s_anywhere::*;
pub use localhost::*;
pub mod k8s;
@@ -26,10 +27,13 @@ pub use tftp::*;
mod helm_command;
pub use helm_command::*;
use super::{
executors::ExecutorError,
instrumentation::{self, HarmonyEvent},
};
use std::error::Error;
use std::net::IpAddr;
use super::interpret::{InterpretError, Outcome};
/// Represents a logical view of an infrastructure environment providing specific capabilities.
///
/// A Topology acts as a self-contained "package" responsible for managing access
@@ -57,9 +61,128 @@ pub trait Topology: Send + Sync {
/// * **Internal Orchestration:** For complex topologies, this method might manage dependencies on other sub-topologies, ensuring *their* `ensure_ready` is called first. Using nested `Maestros` to run setup `Scores` against these sub-topologies is the recommended pattern for non-trivial bootstrapping, allowing reuse of Harmony's core orchestration logic.
///
/// # Returns
/// - `Ok(Outcome)`: Indicates the topology is now ready. The `Outcome` status might be `SUCCESS` if actions were taken, or `NOOP` if it was already ready. The message should provide context.
/// - `Err(TopologyError)`: Indicates the topology could not reach a ready state due to configuration issues, discovery failures, bootstrap errors, or unsupported environments.
async fn ensure_ready(&self) -> Result<Outcome, InterpretError>;
/// - `Ok(PreparationOutcome)`: Indicates the topology is now ready. The `Outcome` status might be `SUCCESS` if actions were taken, or `NOOP` if it was already ready. The message should provide context.
/// - `Err(PreparationError)`: Indicates the topology could not reach a ready state due to configuration issues, discovery failures, bootstrap errors, or unsupported environments.
async fn ensure_ready(&self) -> Result<PreparationOutcome, PreparationError>;
}
#[derive(Debug, Clone, PartialEq, Eq)]
pub enum PreparationOutcome {
Success { details: String },
Noop,
}
#[derive(Debug, Clone, new)]
pub struct PreparationError {
msg: String,
}
impl std::fmt::Display for PreparationError {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.write_str(&self.msg)
}
}
impl Error for PreparationError {}
impl From<ExecutorError> for PreparationError {
fn from(value: ExecutorError) -> Self {
Self {
msg: format!("InterpretError : {value}"),
}
}
}
impl From<kube::Error> for PreparationError {
fn from(value: kube::Error) -> Self {
Self {
msg: format!("PreparationError : {value}"),
}
}
}
impl From<String> for PreparationError {
fn from(value: String) -> Self {
Self {
msg: format!("PreparationError : {value}"),
}
}
}
#[derive(Clone, Debug, PartialEq)]
pub enum TopologyStatus {
Queued,
Preparing,
Success,
Noop,
Error,
}
pub struct TopologyState {
pub topology: String,
pub status: TopologyStatus,
}
impl TopologyState {
pub fn new(topology: String) -> Self {
let instance = Self {
topology,
status: TopologyStatus::Queued,
};
instrumentation::instrument(HarmonyEvent::TopologyStateChanged {
topology: instance.topology.clone(),
status: instance.status.clone(),
message: None,
})
.unwrap();
instance
}
pub fn prepare(&mut self) {
self.status = TopologyStatus::Preparing;
instrumentation::instrument(HarmonyEvent::TopologyStateChanged {
topology: self.topology.clone(),
status: self.status.clone(),
message: None,
})
.unwrap();
}
pub fn success(&mut self, message: String) {
self.status = TopologyStatus::Success;
instrumentation::instrument(HarmonyEvent::TopologyStateChanged {
topology: self.topology.clone(),
status: self.status.clone(),
message: Some(message),
})
.unwrap();
}
pub fn noop(&mut self) {
self.status = TopologyStatus::Noop;
instrumentation::instrument(HarmonyEvent::TopologyStateChanged {
topology: self.topology.clone(),
status: self.status.clone(),
message: None,
})
.unwrap();
}
pub fn error(&mut self, message: String) {
self.status = TopologyStatus::Error;
instrumentation::instrument(HarmonyEvent::TopologyStateChanged {
topology: self.topology.clone(),
status: self.status.clone(),
message: Some(message),
})
.unwrap();
}
}
#[derive(Debug)]
@@ -88,7 +211,7 @@ impl Serialize for Url {
{
match self {
Url::LocalFolder(path) => serializer.serialize_str(path),
Url::Url(url) => serializer.serialize_str(&url.as_str()),
Url::Url(url) => serializer.serialize_str(url.as_str()),
}
}
}

View File

@@ -1,3 +1,5 @@
use std::any::Any;
use async_trait::async_trait;
use log::debug;
@@ -43,7 +45,7 @@ impl<S: AlertSender + Installable<T>, T: Topology> Interpret<T> for AlertingInte
}
fn get_name(&self) -> InterpretName {
todo!()
InterpretName::Alerting
}
fn get_version(&self) -> Version {
@@ -62,7 +64,9 @@ impl<S: AlertSender + Installable<T>, T: Topology> Interpret<T> for AlertingInte
#[async_trait]
pub trait AlertReceiver<S: AlertSender>: std::fmt::Debug + Send + Sync {
async fn install(&self, sender: &S) -> Result<Outcome, InterpretError>;
fn name(&self) -> String;
fn clone_box(&self) -> Box<dyn AlertReceiver<S>>;
fn as_any(&self) -> &dyn Any;
}
#[async_trait]
@@ -72,6 +76,6 @@ pub trait AlertRule<S: AlertSender>: std::fmt::Debug + Send + Sync {
}
#[async_trait]
pub trait ScrapeTarger<S: AlertSender> {
pub trait ScrapeTarget<S: AlertSender> {
async fn install(&self, sender: &S) -> Result<(), InterpretError>;
}

View File

@@ -27,11 +27,11 @@ pub struct UnmanagedRouter {
impl Router for UnmanagedRouter {
fn get_gateway(&self) -> IpAddress {
self.gateway.clone()
self.gateway
}
fn get_cidr(&self) -> Ipv4Cidr {
self.cidr.clone()
self.cidr
}
fn get_host(&self) -> LogicalHost {

View File

@@ -15,7 +15,7 @@ use k8s_openapi::{
apimachinery::pkg::util::intstr::IntOrString,
};
use kube::Resource;
use log::{debug, info, warn};
use log::debug;
use serde::de::DeserializeOwned;
use serde_json::json;
use tokio::sync::OnceCell;
@@ -43,8 +43,7 @@ impl K8sTenantManager {
}
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");
// TODO: Ensure constraints are applied to namespace (https://git.nationtech.io/NationTech/harmony/issues/98)
Ok(())
}
@@ -231,8 +230,13 @@ impl K8sTenantManager {
{
"to": [
{
//TODO this ip is from the docker network that k3d is running on
//since k3d does not deploy kube-api-server as a pod it needs to ahve the ip
//address opened up
//need to find a way to automatically detect the ip address from the docker
//network
"ipBlock": {
"cidr": "172.23.0.0/16",
"cidr": "172.18.0.0/16",
}
}
]
@@ -304,19 +308,19 @@ impl K8sTenantManager {
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())),
port: Some(IntOrString::Int((*port).into())),
..Default::default()
}],
super::PortSpecData::PortRange(start, end) => vec![NetworkPolicyPort {
port: Some(IntOrString::Int(start.clone().into())),
end_port: Some(end.clone().into()),
port: Some(IntOrString::Int((*start).into())),
end_port: Some((*end).into()),
protocol: None, // Not currently supported by Harmony
}],
super::PortSpecData::ListOfPorts(items) => items
.iter()
.map(|i| NetworkPolicyPort {
port: Some(IntOrString::Int(i.clone().into())),
port: Some(IntOrString::Int((*i).into())),
..Default::default()
})
.collect(),
@@ -361,19 +365,19 @@ impl K8sTenantManager {
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())),
port: Some(IntOrString::Int((*port).into())),
..Default::default()
}],
super::PortSpecData::PortRange(start, end) => vec![NetworkPolicyPort {
port: Some(IntOrString::Int(start.clone().into())),
end_port: Some(end.clone().into()),
port: Some(IntOrString::Int((*start).into())),
end_port: Some((*end).into()),
protocol: None, // Not currently supported by Harmony
}],
super::PortSpecData::ListOfPorts(items) => items
.iter()
.map(|i| NetworkPolicyPort {
port: Some(IntOrString::Int(i.clone().into())),
port: Some(IntOrString::Int((*i).into())),
..Default::default()
})
.collect(),
@@ -428,13 +432,14 @@ impl TenantManager for K8sTenantManager {
debug!("Creating network_policy for tenant {}", config.name);
self.apply_resource(network_policy, config).await?;
info!(
debug!(
"Success provisionning K8s tenant id {} name {}",
config.id, config.name
);
self.store_config(config);
Ok(())
}
async fn get_tenant_config(&self) -> Option<TenantConfig> {
self.k8s_tenant_config.get().cloned()
}

View File

@@ -60,7 +60,7 @@ impl DnsServer for OPNSenseFirewall {
}
fn get_ip(&self) -> IpAddress {
OPNSenseFirewall::get_ip(&self)
OPNSenseFirewall::get_ip(self)
}
fn get_host(&self) -> LogicalHost {

View File

@@ -48,7 +48,7 @@ impl HttpServer for OPNSenseFirewall {
async fn ensure_initialized(&self) -> Result<(), ExecutorError> {
let mut config = self.opnsense_config.write().await;
let caddy = config.caddy();
if let None = caddy.get_full_config() {
if caddy.get_full_config().is_none() {
info!("Http config not available in opnsense config, installing package");
config.install_package("os-caddy").await.map_err(|e| {
ExecutorError::UnexpectedError(format!(

View File

@@ -121,10 +121,12 @@ pub(crate) fn haproxy_xml_config_to_harmony_loadbalancer(
LoadBalancerService {
backend_servers,
listening_port: frontend.bind.parse().expect(&format!(
"HAProxy frontend address should be a valid SocketAddr, got {}",
frontend.bind
)),
listening_port: frontend.bind.parse().unwrap_or_else(|_| {
panic!(
"HAProxy frontend address should be a valid SocketAddr, got {}",
frontend.bind
)
}),
health_check,
}
})
@@ -167,28 +169,28 @@ pub(crate) fn get_health_check_for_backend(
None => return None,
};
let haproxy_health_check = match haproxy
let haproxy_health_check = haproxy
.healthchecks
.healthchecks
.iter()
.find(|h| &h.uuid == health_check_uuid)
{
Some(health_check) => health_check,
None => return None,
};
.find(|h| &h.uuid == health_check_uuid)?;
let binding = haproxy_health_check.health_check_type.to_uppercase();
let uppercase = binding.as_str();
match uppercase {
"TCP" => {
if let Some(checkport) = haproxy_health_check.checkport.content.as_ref() {
if checkport.len() > 0 {
return Some(HealthCheck::TCP(Some(checkport.parse().expect(&format!(
"HAProxy check port should be a valid port number, got {checkport}"
)))));
if !checkport.is_empty() {
return Some(HealthCheck::TCP(Some(checkport.parse().unwrap_or_else(
|_| {
panic!(
"HAProxy check port should be a valid port number, got {checkport}"
)
},
))));
}
}
return Some(HealthCheck::TCP(None));
Some(HealthCheck::TCP(None))
}
"HTTP" => {
let path: String = haproxy_health_check
@@ -355,16 +357,13 @@ mod tests {
// Create an HAProxy instance with servers
let mut haproxy = HAProxy::default();
let mut server = HAProxyServer::default();
server.uuid = "server1".to_string();
server.address = "192.168.1.1".to_string();
server.port = 80;
let server = HAProxyServer {
uuid: "server1".to_string(),
address: "192.168.1.1".to_string(),
port: 80,
..Default::default()
};
haproxy.servers.servers.push(server);
let mut server = HAProxyServer::default();
server.uuid = "server3".to_string();
server.address = "192.168.1.3".to_string();
server.port = 8080;
// Call the function
let result = get_servers_for_backend(&backend, &haproxy);
@@ -384,10 +383,12 @@ mod tests {
let backend = HAProxyBackend::default();
// Create an HAProxy instance with servers
let mut haproxy = HAProxy::default();
let mut server = HAProxyServer::default();
server.uuid = "server1".to_string();
server.address = "192.168.1.1".to_string();
server.port = 80;
let server = HAProxyServer {
uuid: "server1".to_string(),
address: "192.168.1.1".to_string(),
port: 80,
..Default::default()
};
haproxy.servers.servers.push(server);
// Call the function
let result = get_servers_for_backend(&backend, &haproxy);
@@ -402,10 +403,12 @@ mod tests {
backend.linked_servers.content = Some("server4,server5".to_string());
// Create an HAProxy instance with servers
let mut haproxy = HAProxy::default();
let mut server = HAProxyServer::default();
server.uuid = "server1".to_string();
server.address = "192.168.1.1".to_string();
server.port = 80;
let server = HAProxyServer {
uuid: "server1".to_string(),
address: "192.168.1.1".to_string(),
port: 80,
..Default::default()
};
haproxy.servers.servers.push(server);
// Call the function
let result = get_servers_for_backend(&backend, &haproxy);
@@ -416,20 +419,28 @@ mod tests {
#[test]
fn test_get_servers_for_backend_multiple_linked_servers() {
// Create a backend with multiple linked servers
#[allow(clippy::field_reassign_with_default)]
let mut backend = HAProxyBackend::default();
backend.linked_servers.content = Some("server1,server2".to_string());
//
// Create an HAProxy instance with matching servers
let mut haproxy = HAProxy::default();
let mut server = HAProxyServer::default();
server.uuid = "server1".to_string();
server.address = "some-hostname.test.mcd".to_string();
server.port = 80;
let server = HAProxyServer {
uuid: "server1".to_string(),
address: "some-hostname.test.mcd".to_string(),
port: 80,
..Default::default()
};
haproxy.servers.servers.push(server);
let mut server = HAProxyServer::default();
server.uuid = "server2".to_string();
server.address = "192.168.1.2".to_string();
server.port = 8080;
let server = HAProxyServer {
uuid: "server2".to_string(),
address: "192.168.1.2".to_string(),
port: 8080,
..Default::default()
};
haproxy.servers.servers.push(server);
// Call the function
let result = get_servers_for_backend(&backend, &haproxy);
// Check the result

View File

@@ -58,7 +58,7 @@ impl TftpServer for OPNSenseFirewall {
async fn ensure_initialized(&self) -> Result<(), ExecutorError> {
let mut config = self.opnsense_config.write().await;
let tftp = config.tftp();
if let None = tftp.get_full_config() {
if tftp.get_full_config().is_none() {
info!("Tftp config not available in opnsense config, installing package");
config.install_package("os-tftp").await.map_err(|e| {
ExecutorError::UnexpectedError(format!(

View File

@@ -3,7 +3,6 @@ use serde::Serialize;
use crate::topology::Topology;
use super::Application;
/// An ApplicationFeature provided by harmony, such as Backups, Monitoring, MultisiteAvailability,
/// ContinuousIntegration, ContinuousDelivery
#[async_trait]
@@ -14,7 +13,7 @@ pub trait ApplicationFeature<T: Topology>:
fn name(&self) -> String;
}
trait ApplicationFeatureClone<T: Topology> {
pub trait ApplicationFeatureClone<T: Topology> {
fn clone_box(&self) -> Box<dyn ApplicationFeature<T>>;
}
@@ -28,7 +27,7 @@ where
}
impl<T: Topology> Serialize for Box<dyn ApplicationFeature<T>> {
fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{

View File

@@ -0,0 +1,353 @@
use log::debug;
use serde::Serialize;
use serde_with::skip_serializing_none;
use serde_yaml::Value;
use crate::modules::application::features::CDApplicationConfig;
#[skip_serializing_none]
#[derive(Clone, Debug, Serialize)]
#[serde(rename_all = "camelCase")]
pub struct Helm {
pub pass_credentials: Option<bool>,
pub parameters: Vec<Value>,
pub file_parameters: Vec<Value>,
pub release_name: Option<String>,
pub value_files: Vec<String>,
pub ignore_missing_value_files: Option<bool>,
pub values: Option<String>,
pub values_object: Option<Value>,
pub skip_crds: Option<bool>,
pub skip_schema_validation: Option<bool>,
pub version: Option<String>,
pub kube_version: Option<String>,
pub api_versions: Vec<String>,
pub namespace: Option<String>,
}
#[skip_serializing_none]
#[derive(Clone, Debug, Serialize)]
#[serde(rename_all = "camelCase")]
pub struct Source {
// Using string for this because URL enforces a URL scheme at the beginning but Helm, ArgoCD, etc do not, and it can be counterproductive,
// as the only way I've found to get OCI working isn't by using oci:// but rather no scheme at all
#[serde(rename = "repoURL")]
pub repo_url: String,
pub target_revision: Option<String>,
pub chart: String,
pub helm: Helm,
pub path: String,
}
#[derive(Clone, Debug, Serialize)]
#[serde(rename_all = "camelCase")]
pub struct Automated {
pub prune: bool,
pub self_heal: bool,
pub allow_empty: bool,
}
#[derive(Clone, Debug, Serialize)]
#[serde(rename_all = "camelCase")]
pub struct Backoff {
pub duration: String,
pub factor: u32,
pub max_duration: String,
}
#[derive(Clone, Debug, Serialize)]
#[serde(rename_all = "camelCase")]
pub struct Retry {
pub limit: u32,
pub backoff: Backoff,
}
#[derive(Clone, Debug, Serialize)]
#[serde(rename_all = "camelCase")]
pub struct SyncPolicy {
pub automated: Automated,
pub sync_options: Vec<String>,
pub retry: Retry,
}
#[skip_serializing_none]
#[derive(Clone, Debug, Serialize)]
#[serde(rename_all = "camelCase")]
pub struct ArgoApplication {
pub name: String,
pub namespace: Option<String>,
pub project: String,
pub source: Source,
pub sync_policy: SyncPolicy,
pub revision_history_limit: u32,
}
impl Default for ArgoApplication {
fn default() -> Self {
Self {
name: Default::default(),
namespace: Default::default(),
project: Default::default(),
source: Source {
repo_url: "http://asdf".to_string(),
target_revision: None,
chart: "".to_string(),
helm: Helm {
pass_credentials: None,
parameters: vec![],
file_parameters: vec![],
release_name: None,
value_files: vec![],
ignore_missing_value_files: None,
values: None,
values_object: None,
skip_crds: None,
skip_schema_validation: None,
version: None,
kube_version: None,
api_versions: vec![],
namespace: None,
},
path: "".to_string(),
},
sync_policy: SyncPolicy {
automated: Automated {
prune: false,
self_heal: false,
allow_empty: false,
},
sync_options: vec![],
retry: Retry {
limit: 5,
backoff: Backoff {
duration: "5s".to_string(),
factor: 2,
max_duration: "3m".to_string(),
},
},
},
revision_history_limit: 10,
}
}
}
impl From<CDApplicationConfig> for ArgoApplication {
fn from(value: CDApplicationConfig) -> Self {
Self {
name: value.name,
namespace: Some(value.namespace),
project: "default".to_string(),
source: Source {
repo_url: value.helm_chart_repo_url,
target_revision: Some(value.version.to_string()),
chart: value.helm_chart_name.clone(),
path: value.helm_chart_name,
helm: Helm {
pass_credentials: None,
parameters: vec![],
file_parameters: vec![],
release_name: None,
value_files: vec![],
ignore_missing_value_files: None,
values: None,
values_object: value.values_overrides,
skip_crds: None,
skip_schema_validation: None,
version: None,
kube_version: None,
api_versions: vec![],
namespace: None,
},
},
sync_policy: SyncPolicy {
automated: Automated {
prune: false,
self_heal: false,
allow_empty: true,
},
sync_options: vec![],
retry: Retry {
limit: 5,
backoff: Backoff {
duration: "5s".to_string(),
factor: 2,
max_duration: "3m".to_string(),
},
},
},
..Self::default()
}
}
}
impl ArgoApplication {
pub fn to_yaml(&self) -> serde_yaml::Value {
let name = &self.name;
let namespace = if let Some(ns) = self.namespace.as_ref() {
ns
} else {
"argocd"
};
let project = &self.project;
let yaml_str = format!(
r#"
apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
name: {name}
# You'll usually want to add your resources to the argocd namespace.
namespace: {namespace}
spec:
# The project the application belongs to.
project: {project}
# Destination cluster and namespace to deploy the application
destination:
# cluster API URL
server: https://kubernetes.default.svc
# or cluster name
# name: in-cluster
# The namespace will only be set for namespace-scoped resources that have not set a value for .metadata.namespace
namespace: {namespace}
"#
);
let mut yaml_value: Value =
serde_yaml::from_str(yaml_str.as_str()).expect("couldn't parse string to YAML");
let spec = yaml_value
.get_mut("spec")
.expect("couldn't get spec from yaml")
.as_mapping_mut()
.expect("couldn't unwrap spec as mutable mapping");
let source =
serde_yaml::to_value(&self.source).expect("couldn't serialize source to value");
let sync_policy = serde_yaml::to_value(&self.sync_policy)
.expect("couldn't serialize sync_policy to value");
let revision_history_limit = serde_yaml::to_value(self.revision_history_limit)
.expect("couldn't serialize revision_history_limit to value");
spec.insert(
serde_yaml::to_value("source").expect("string to value failed"),
source,
);
spec.insert(
serde_yaml::to_value("syncPolicy").expect("string to value failed"),
sync_policy,
);
spec.insert(
serde_yaml::to_value("revisionHistoryLimit")
.expect("couldn't convert str to yaml value"),
revision_history_limit,
);
debug!("spec: {}", serde_yaml::to_string(spec).unwrap());
debug!(
"entire yaml_value: {}",
serde_yaml::to_string(&yaml_value).unwrap()
);
yaml_value
}
}
#[cfg(test)]
mod tests {
use pretty_assertions::assert_eq;
use crate::modules::application::features::{
ArgoApplication, Automated, Backoff, Helm, Retry, Source, SyncPolicy,
};
#[test]
fn test_argo_application_to_yaml_happy_path() {
let app = ArgoApplication {
name: "test".to_string(),
namespace: Some("test-ns".to_string()),
project: "test-project".to_string(),
source: Source {
repo_url: "http://test".to_string(),
target_revision: None,
chart: "test-chart".to_string(),
helm: Helm {
pass_credentials: None,
parameters: vec![],
file_parameters: vec![],
release_name: Some("test-release-neame".to_string()),
value_files: vec![],
ignore_missing_value_files: None,
values: None,
values_object: None,
skip_crds: None,
skip_schema_validation: None,
version: None,
kube_version: None,
api_versions: vec![],
namespace: None,
},
path: "".to_string(),
},
sync_policy: SyncPolicy {
automated: Automated {
prune: false,
self_heal: false,
allow_empty: false,
},
sync_options: vec![],
retry: Retry {
limit: 5,
backoff: Backoff {
duration: "5s".to_string(),
factor: 2,
max_duration: "3m".to_string(),
},
},
},
revision_history_limit: 10,
};
let expected_yaml_output = r#"apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
name: test
namespace: test-ns
spec:
project: test-project
destination:
server: https://kubernetes.default.svc
namespace: test-ns
source:
repoURL: http://test
chart: test-chart
helm:
parameters: []
fileParameters: []
releaseName: test-release-neame
valueFiles: []
apiVersions: []
path: ''
syncPolicy:
automated:
prune: false
selfHeal: false
allowEmpty: false
syncOptions: []
retry:
limit: 5
backoff:
duration: 5s
factor: 2
maxDuration: 3m
revisionHistoryLimit: 10"#;
assert_eq!(
expected_yaml_output.trim(),
serde_yaml::to_string(&app.clone().to_yaml())
.unwrap()
.trim()
);
}
}

View File

@@ -1,20 +1,20 @@
use std::{io::Write, process::Command, sync::Arc};
use async_trait::async_trait;
use log::{error, info};
use serde_json::Value;
use log::{debug, error};
use serde_yaml::Value;
use tempfile::NamedTempFile;
use crate::{
config::HARMONY_DATA_DIR,
data::Version,
inventory::Inventory,
modules::{
application::{Application, ApplicationFeature, HelmPackage, OCICompliant},
helm::chart::HelmChartScore,
modules::application::{
ApplicationFeature, HelmPackage, OCICompliant,
features::{ArgoApplication, ArgoHelmScore},
},
score::Score,
topology::{DeploymentTarget, HelmCommand, MultiTargetTopology, Topology, Url},
topology::{DeploymentTarget, HelmCommand, K8sclient, MultiTargetTopology, Topology},
};
/// ContinuousDelivery in Harmony provides this functionality :
@@ -63,7 +63,7 @@ impl<A: OCICompliant + HelmPackage> ContinuousDelivery<A> {
error!("TODO hardcoded k3d bin path is wrong");
let k3d_bin_path = (*HARMONY_DATA_DIR).join("k3d").join("k3d");
// --- 1. Import the container image into the k3d cluster ---
info!(
debug!(
"Importing image '{}' into k3d cluster 'harmony'",
image_name
);
@@ -80,7 +80,7 @@ impl<A: OCICompliant + HelmPackage> ContinuousDelivery<A> {
}
// --- 2. Get the kubeconfig for the k3d cluster and write it to a temp file ---
info!("Retrieving kubeconfig for k3d cluster 'harmony'");
debug!("Retrieving kubeconfig for k3d cluster 'harmony'");
let kubeconfig_output = Command::new(&k3d_bin_path)
.args(["kubeconfig", "get", "harmony"])
.output()
@@ -101,7 +101,7 @@ impl<A: OCICompliant + HelmPackage> ContinuousDelivery<A> {
let kubeconfig_path = temp_kubeconfig.path().to_str().unwrap();
// --- 3. Install or upgrade the Helm chart in the cluster ---
info!(
debug!(
"Deploying Helm chart '{}' to namespace '{}'",
chart_url, app_name
);
@@ -131,7 +131,7 @@ impl<A: OCICompliant + HelmPackage> ContinuousDelivery<A> {
));
}
info!("Successfully deployed '{}' to local k3d cluster.", app_name);
debug!("Successfully deployed '{}' to local k3d cluster.", app_name);
Ok(())
}
}
@@ -139,25 +139,25 @@ impl<A: OCICompliant + HelmPackage> ContinuousDelivery<A> {
#[async_trait]
impl<
A: OCICompliant + HelmPackage + Clone + 'static,
T: Topology + HelmCommand + MultiTargetTopology + 'static,
T: Topology + HelmCommand + MultiTargetTopology + K8sclient + 'static,
> ApplicationFeature<T> for ContinuousDelivery<A>
{
async fn ensure_installed(&self, topology: &T) -> Result<(), String> {
let image = self.application.image_name();
// TODO
error!(
"TODO reverse helm chart packaging and docker image build. I put helm package first for faster iterations"
);
// TODO Write CI/CD workflow files
// we can autotedect the CI type using the remote url (default to github action for github
// url, etc..)
// Or ask for it when unknown
let helm_chart = self.application.build_push_helm_package(&image).await?;
info!("Pushed new helm chart {helm_chart}");
debug!("Pushed new helm chart {helm_chart}");
// let image = self.application.build_push_oci_image().await?;
// info!("Pushed new docker image {image}");
error!("uncomment above");
error!("TODO Make building image configurable/skippable if image already exists (prompt)");
let image = self.application.build_push_oci_image().await?;
debug!("Pushed new docker image {image}");
info!("Installing ContinuousDelivery feature");
debug!("Installing ContinuousDelivery feature");
// TODO this is a temporary hack for demo purposes, the deployment target should be driven
// by the topology only and we should not have to know how to perform tasks like this for
// which the topology should be responsible.
@@ -177,32 +177,28 @@ impl<
.await?;
}
target => {
info!("Deploying to target {target:?}");
let cd_server = HelmChartScore {
namespace: todo!(
"ArgoCD Helm chart with proper understanding of Tenant, see how Will did it for Monitoring for now"
),
release_name: todo!("argocd helm chart whatever"),
chart_name: todo!(),
chart_version: todo!(),
values_overrides: todo!(),
values_yaml: todo!(),
create_namespace: todo!(),
install_only: todo!(),
repository: todo!(),
debug!("Deploying to target {target:?}");
let score = ArgoHelmScore {
namespace: "harmonydemo-staging".to_string(),
openshift: false,
domain: "argo.harmonydemo.apps.st.mcd".to_string(),
argo_apps: vec![ArgoApplication::from(CDApplicationConfig {
// helm pull oci://hub.nationtech.io/harmony/harmony-example-rust-webapp-chart --version 0.1.0
version: Version::from("0.1.0").unwrap(),
helm_chart_repo_url: "hub.nationtech.io/harmony".to_string(),
helm_chart_name: "harmony-example-rust-webapp-chart".to_string(),
values_overrides: None,
name: "harmony-demo-rust-webapp".to_string(),
namespace: "harmonydemo-staging".to_string(),
})],
};
let interpret = cd_server.create_interpret();
interpret.execute(&Inventory::empty(), topology);
score
.interpret(&Inventory::empty(), topology)
.await
.unwrap();
}
};
todo!("1. Create ArgoCD score that installs argo using helm chart, see if Taha's already done it
- [X] Package app (docker image, helm chart)
- [X] Push to registry
- [ ] Push only if staging or prod
- [ ] Deploy to local k3d when target is local
- [ ] Poke Argo
- [ ] Ensure app is up")
Ok(())
}
fn name(&self) -> String {
"ContinuousDelivery".to_string()
@@ -212,9 +208,12 @@ impl<
/// For now this is entirely bound to K8s / ArgoCD, will have to be revisited when we support
/// more CD systems
pub struct CDApplicationConfig {
version: Version,
helm_chart_url: Url,
values_overrides: Value,
pub version: Version,
pub helm_chart_repo_url: String,
pub helm_chart_name: String,
pub values_overrides: Option<Value>,
pub name: String,
pub namespace: String,
}
pub trait ContinuousDeliveryApplication {

View File

@@ -2,7 +2,7 @@ use async_trait::async_trait;
use log::info;
use crate::{
modules::application::{Application, ApplicationFeature},
modules::application::ApplicationFeature,
topology::{K8sclient, Topology},
};

File diff suppressed because it is too large Load Diff

View File

@@ -6,3 +6,9 @@ pub use monitoring::*;
mod continuous_delivery;
pub use continuous_delivery::*;
mod helm_argocd_score;
pub use helm_argocd_score::*;
mod argo_types;
pub use argo_types::*;

View File

@@ -1,19 +1,103 @@
use async_trait::async_trait;
use log::info;
use std::sync::Arc;
use crate::modules::application::{Application, ApplicationFeature};
use crate::modules::monitoring::application_monitoring::application_monitoring_score::ApplicationMonitoringScore;
use crate::modules::monitoring::kube_prometheus::crd::crd_alertmanager_config::CRDPrometheus;
use crate::{
modules::application::{Application, ApplicationFeature},
topology::{HelmCommand, Topology},
inventory::Inventory,
modules::monitoring::{
alert_channel::webhook_receiver::WebhookReceiver, ntfy::ntfy::NtfyScore,
},
score::Score,
topology::{HelmCommand, K8sclient, Topology, Url, tenant::TenantManager},
};
use crate::{
modules::prometheus::prometheus::PrometheusApplicationMonitoring,
topology::oberservability::monitoring::AlertReceiver,
};
use async_trait::async_trait;
use base64::{Engine as _, engine::general_purpose};
use log::{debug, info};
#[derive(Debug, Default, Clone)]
pub struct Monitoring {}
#[derive(Debug, Clone)]
pub struct Monitoring {
pub application: Arc<dyn Application>,
pub alert_receiver: Vec<Box<dyn AlertReceiver<CRDPrometheus>>>,
}
#[async_trait]
impl<T: Topology + HelmCommand + 'static> ApplicationFeature<T> for Monitoring {
async fn ensure_installed(&self, _topology: &T) -> Result<(), String> {
impl<
T: Topology
+ HelmCommand
+ 'static
+ TenantManager
+ K8sclient
+ std::fmt::Debug
+ PrometheusApplicationMonitoring<CRDPrometheus>,
> ApplicationFeature<T> for Monitoring
{
async fn ensure_installed(&self, topology: &T) -> Result<(), String> {
info!("Ensuring monitoring is available for application");
todo!("create and execute k8s prometheus score, depends on Will's work")
let namespace = topology
.get_tenant_config()
.await
.map(|ns| ns.name.clone())
.unwrap_or_else(|| self.application.name());
let mut alerting_score = ApplicationMonitoringScore {
sender: CRDPrometheus {
namespace: namespace.clone(),
client: topology.k8s_client().await.unwrap(),
},
application: self.application.clone(),
receivers: self.alert_receiver.clone(),
};
let ntfy = NtfyScore {
namespace: namespace.clone(),
host: "localhost".to_string(),
};
ntfy.interpret(&Inventory::empty(), topology)
.await
.expect("couldn't create interpret for ntfy");
let ntfy_default_auth_username = "harmony";
let ntfy_default_auth_password = "harmony";
let ntfy_default_auth_header = format!(
"Basic {}",
general_purpose::STANDARD.encode(format!(
"{ntfy_default_auth_username}:{ntfy_default_auth_password}"
))
);
debug!("ntfy_default_auth_header: {ntfy_default_auth_header}");
let ntfy_default_auth_param = general_purpose::STANDARD
.encode(ntfy_default_auth_header)
.replace("=", "");
debug!("ntfy_default_auth_param: {ntfy_default_auth_param}");
let ntfy_receiver = WebhookReceiver {
name: "ntfy-webhook".to_string(),
url: Url::Url(
url::Url::parse(
format!(
"http://ntfy.{}.svc.cluster.local/rust-web-app?auth={ntfy_default_auth_param}",
namespace.clone()
)
.as_str(),
)
.unwrap(),
),
};
alerting_score.receivers.push(Box::new(ntfy_receiver));
alerting_score
.interpret(&Inventory::empty(), topology)
.await
.unwrap();
Ok(())
}
fn name(&self) -> String {
"Monitoring".to_string()

View File

@@ -5,11 +5,12 @@ mod rust;
use std::sync::Arc;
pub use feature::*;
use log::info;
use log::debug;
pub use oci::*;
pub use rust::*;
use async_trait::async_trait;
use serde::Serialize;
use crate::{
data::{Id, Version},
@@ -23,20 +24,20 @@ pub trait Application: std::fmt::Debug + Send + Sync {
}
#[derive(Debug)]
pub struct ApplicationInterpret<T: Topology + std::fmt::Debug> {
pub struct ApplicationInterpret<A: Application, T: Topology + std::fmt::Debug> {
features: Vec<Box<dyn ApplicationFeature<T>>>,
application: Arc<Box<dyn Application>>,
application: Arc<A>,
}
#[async_trait]
impl<T: Topology + std::fmt::Debug> Interpret<T> for ApplicationInterpret<T> {
impl<A: Application, T: Topology + std::fmt::Debug> Interpret<T> for ApplicationInterpret<A, T> {
async fn execute(
&self,
_inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
let app_name = self.application.name();
info!(
debug!(
"Preparing {} features [{}] for application {app_name}",
self.features.len(),
self.features
@@ -46,7 +47,7 @@ impl<T: Topology + std::fmt::Debug> Interpret<T> for ApplicationInterpret<T> {
.join(", ")
);
for feature in self.features.iter() {
info!(
debug!(
"Installing feature {} for application {app_name}",
feature.name()
);
@@ -59,9 +60,7 @@ impl<T: Topology + std::fmt::Debug> Interpret<T> for ApplicationInterpret<T> {
}
};
}
todo!(
"Do I need to do anything more than this here?? I feel like the Application trait itself should expose something like ensure_ready but its becoming redundant. We'll see as this evolves."
)
Ok(Outcome::success("Application created".to_string()))
}
fn get_name(&self) -> InterpretName {
@@ -80,3 +79,12 @@ impl<T: Topology + std::fmt::Debug> Interpret<T> for ApplicationInterpret<T> {
todo!()
}
}
impl Serialize for dyn Application {
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{
todo!()
}
}

View File

@@ -1,14 +1,18 @@
use std::fs;
use std::path::PathBuf;
use std::path::{Path, PathBuf};
use std::process;
use std::sync::Arc;
use async_trait::async_trait;
use bollard::query_parameters::PushImageOptionsBuilder;
use bollard::{Docker, body_full};
use dockerfile_builder::Dockerfile;
use dockerfile_builder::instruction::{CMD, COPY, ENV, EXPOSE, FROM, RUN, USER, WORKDIR};
use dockerfile_builder::instruction_builder::CopyBuilder;
use log::{debug, error, info};
use futures_util::StreamExt;
use log::{debug, error, log_enabled};
use serde::Serialize;
use tar::Archive;
use crate::config::{REGISTRY_PROJECT, REGISTRY_URL};
use crate::{
@@ -19,23 +23,30 @@ use crate::{
use super::{Application, ApplicationFeature, ApplicationInterpret, HelmPackage, OCICompliant};
#[derive(Debug, Serialize, Clone)]
pub struct RustWebappScore<T: Topology + Clone + Serialize> {
pub name: String,
pub domain: Url,
pub struct ApplicationScore<A: Application + Serialize, T: Topology + Clone + Serialize>
where
Arc<A>: Serialize + Clone,
{
pub features: Vec<Box<dyn ApplicationFeature<T>>>,
pub application: RustWebapp,
pub application: Arc<A>,
}
impl<T: Topology + std::fmt::Debug + Clone + Serialize + 'static> Score<T> for RustWebappScore<T> {
impl<
A: Application + Serialize + Clone + 'static,
T: Topology + std::fmt::Debug + Clone + Serialize + 'static,
> Score<T> for ApplicationScore<A, T>
where
Arc<A>: Serialize,
{
fn create_interpret(&self) -> Box<dyn crate::interpret::Interpret<T>> {
Box::new(ApplicationInterpret {
features: self.features.clone(),
application: Arc::new(Box::new(self.application.clone())),
application: self.application.clone(),
})
}
fn name(&self) -> String {
format!("{}-RustWebapp", self.name)
format!("{} [ApplicationScore]", self.application.name())
}
}
@@ -47,6 +58,7 @@ pub enum RustWebFramework {
#[derive(Debug, Clone, Serialize)]
pub struct RustWebapp {
pub name: String,
pub domain: Url,
/// The path to the root of the Rust project to be containerized.
pub project_root: PathBuf,
pub framework: Option<RustWebFramework>,
@@ -61,19 +73,19 @@ impl Application for RustWebapp {
#[async_trait]
impl HelmPackage for RustWebapp {
async fn build_push_helm_package(&self, image_url: &str) -> Result<String, String> {
info!("Starting Helm chart build and push for '{}'", self.name);
debug!("Starting Helm chart build and push for '{}'", self.name);
// 1. Create the Helm chart files on disk.
let chart_dir = self
.create_helm_chart_files(image_url)
.map_err(|e| format!("Failed to create Helm chart files: {}", e))?;
info!("Successfully created Helm chart files in {:?}", chart_dir);
debug!("Successfully created Helm chart files in {:?}", chart_dir);
// 2. Package the chart into a .tgz archive.
let packaged_chart_path = self
.package_helm_chart(&chart_dir)
.map_err(|e| format!("Failed to package Helm chart: {}", e))?;
info!(
debug!(
"Successfully packaged Helm chart: {}",
packaged_chart_path.to_string_lossy()
);
@@ -82,7 +94,7 @@ impl HelmPackage for RustWebapp {
let oci_chart_url = self
.push_helm_chart(&packaged_chart_path)
.map_err(|e| format!("Failed to push Helm chart: {}", e))?;
info!("Successfully pushed Helm chart to: {}", oci_chart_url);
debug!("Successfully pushed Helm chart to: {}", oci_chart_url);
Ok(oci_chart_url)
}
@@ -95,24 +107,22 @@ impl OCICompliant for RustWebapp {
async fn build_push_oci_image(&self) -> Result<String, String> {
// This function orchestrates the build and push process.
// It's async to match the trait definition, though the underlying docker commands are blocking.
info!("Starting OCI image build and push for '{}'", self.name);
debug!("Starting OCI image build and push for '{}'", self.name);
// 1. Build the local image by calling the synchronous helper function.
let local_image_name = self.local_image_name();
self.build_docker_image(&local_image_name)
// 1. Build the image by calling the synchronous helper function.
let image_tag = self.image_name();
self.build_docker_image(&image_tag)
.await
.map_err(|e| format!("Failed to build Docker image: {}", e))?;
info!(
"Successfully built local Docker image: {}",
local_image_name
);
debug!("Successfully built Docker image: {}", image_tag);
let remote_image_name = self.image_name();
// 2. Push the image to the registry.
self.push_docker_image(&local_image_name, &remote_image_name)
self.push_docker_image(&image_tag)
.await
.map_err(|e| format!("Failed to push Docker image: {}", e))?;
info!("Successfully pushed Docker image to: {}", remote_image_name);
debug!("Successfully pushed Docker image to: {}", image_tag);
Ok(remote_image_name)
Ok(image_tag)
}
fn local_image_name(&self) -> String {
@@ -145,68 +155,74 @@ impl RustWebapp {
}
/// Builds the Docker image using the generated Dockerfile.
pub fn build_docker_image(
pub async fn build_docker_image(
&self,
image_name: &str,
) -> Result<String, Box<dyn std::error::Error>> {
info!("Generating Dockerfile for '{}'", self.name);
let dockerfile_path = self.build_dockerfile()?;
debug!("Generating Dockerfile for '{}'", self.name);
let _dockerfile_path = self.build_dockerfile()?;
info!(
"Building Docker image with file {} from root {}",
dockerfile_path.to_string_lossy(),
self.project_root.to_string_lossy()
let docker = Docker::connect_with_socket_defaults().unwrap();
let quiet = !log_enabled!(log::Level::Debug);
let build_image_options = bollard::query_parameters::BuildImageOptionsBuilder::default()
.dockerfile("Dockerfile.harmony")
.t(image_name)
.q(quiet)
.version(bollard::query_parameters::BuilderVersion::BuilderV1)
.platform("linux/x86_64");
let mut temp_tar_builder = tar::Builder::new(Vec::new());
temp_tar_builder
.append_dir_all("", self.project_root.clone())
.unwrap();
let archive = temp_tar_builder
.into_inner()
.expect("couldn't finish creating tar");
let archived_files = Archive::new(archive.as_slice())
.entries()
.unwrap()
.map(|entry| entry.unwrap().path().unwrap().into_owned())
.collect::<Vec<_>>();
debug!("files in docker tar: {:#?}", archived_files);
let mut image_build_stream = docker.build_image(
build_image_options.build(),
None,
Some(body_full(archive.into())),
);
let output = process::Command::new("docker")
.args([
"build",
"--file",
dockerfile_path.to_str().unwrap(),
"-t",
&image_name,
self.project_root.to_str().unwrap(),
])
.spawn()?
.wait_with_output()?;
self.check_output(&output, "Failed to build Docker image")?;
while let Some(msg) = image_build_stream.next().await {
println!("Message: {msg:?}");
}
Ok(image_name.to_string())
}
/// Tags and pushes a Docker image to the configured remote registry.
fn push_docker_image(
async fn push_docker_image(
&self,
image_name: &str,
full_tag: &str,
image_tag: &str,
) -> Result<String, Box<dyn std::error::Error>> {
info!("Pushing docker image {full_tag}");
debug!("Pushing docker image {image_tag}");
// Tag the image for the remote registry.
let output = process::Command::new("docker")
.args(["tag", image_name, &full_tag])
.spawn()?
.wait_with_output()?;
self.check_output(&output, "Tagging docker image failed")?;
debug!(
"docker tag output: stdout: {}, stderr: {}",
String::from_utf8_lossy(&output.stdout),
String::from_utf8_lossy(&output.stderr)
let docker = Docker::connect_with_socket_defaults().unwrap();
// let push_options = PushImageOptionsBuilder::new().tag(tag);
let mut push_image_stream = docker.push_image(
image_tag,
Some(PushImageOptionsBuilder::new().build()),
None,
);
// Push the image.
let output = process::Command::new("docker")
.args(["push", &full_tag])
.spawn()?
.wait_with_output()?;
self.check_output(&output, "Pushing docker image failed")?;
debug!(
"docker push output: stdout: {}, stderr: {}",
String::from_utf8_lossy(&output.stdout),
String::from_utf8_lossy(&output.stderr)
);
while let Some(msg) = push_image_stream.next().await {
println!("Message: {msg:?}");
}
Ok(full_tag.to_string())
Ok(image_tag.to_string())
}
/// Checks the output of a process command for success.
@@ -341,7 +357,11 @@ impl RustWebapp {
image_url: &str,
) -> Result<PathBuf, Box<dyn std::error::Error>> {
let chart_name = format!("{}-chart", self.name);
let chart_dir = self.project_root.join("helm").join(&chart_name);
let chart_dir = self
.project_root
.join(".harmony_generated")
.join("helm")
.join(&chart_name);
let templates_dir = chart_dir.join("templates");
fs::create_dir_all(&templates_dir)?;
@@ -408,7 +428,7 @@ ingress:
Expand the name of the chart.
*/}}
{{- define "chart.name" -}}
{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
{{- default .Chart.Name $.Values.nameOverride | trunc 63 | trimSuffix "-" }}
{{- end }}
{{/*
@@ -416,7 +436,7 @@ Create a default fully qualified app name.
We truncate at 63 chars because some Kubernetes name fields are limited to this (by the DNS naming spec).
*/}}
{{- define "chart.fullname" -}}
{{- $name := default .Chart.Name .Values.nameOverride }}
{{- $name := default .Chart.Name $.Values.nameOverride }}
{{- printf "%s-%s" .Release.Name $name | trunc 63 | trimSuffix "-" }}
{{- end }}
"#;
@@ -429,12 +449,12 @@ kind: Service
metadata:
name: {{ include "chart.fullname" . }}
spec:
type: {{ .Values.service.type }}
type: {{ $.Values.service.type }}
ports:
- port: {{ .Values.service.port }}
targetPort: 3000
- name: main
port: {{ $.Values.service.port | default 3000 }}
targetPort: {{ $.Values.service.port | default 3000 }}
protocol: TCP
name: http
selector:
app: {{ include "chart.name" . }}
"#;
@@ -447,7 +467,7 @@ kind: Deployment
metadata:
name: {{ include "chart.fullname" . }}
spec:
replicas: {{ .Values.replicaCount }}
replicas: {{ $.Values.replicaCount }}
selector:
matchLabels:
app: {{ include "chart.name" . }}
@@ -458,28 +478,28 @@ spec:
spec:
containers:
- name: {{ .Chart.Name }}
image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"
imagePullPolicy: {{ .Values.image.pullPolicy }}
image: "{{ $.Values.image.repository }}:{{ $.Values.image.tag | default .Chart.AppVersion }}"
imagePullPolicy: {{ $.Values.image.pullPolicy }}
ports:
- name: http
containerPort: 3000
- name: main
containerPort: {{ $.Values.service.port | default 3000 }}
protocol: TCP
"#;
fs::write(templates_dir.join("deployment.yaml"), deployment_yaml)?;
// Create templates/ingress.yaml
let ingress_yaml = r#"
{{- if .Values.ingress.enabled -}}
{{- if $.Values.ingress.enabled -}}
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: {{ include "chart.fullname" . }}
annotations:
{{- toYaml .Values.ingress.annotations | nindent 4 }}
{{- toYaml $.Values.ingress.annotations | nindent 4 }}
spec:
{{- if .Values.ingress.tls }}
{{- if $.Values.ingress.tls }}
tls:
{{- range .Values.ingress.tls }}
{{- range $.Values.ingress.tls }}
- hosts:
{{- range .hosts }}
- {{ . | quote }}
@@ -488,7 +508,7 @@ spec:
{{- end }}
{{- end }}
rules:
{{- range .Values.ingress.hosts }}
{{- range $.Values.ingress.hosts }}
- host: {{ .host | quote }}
http:
paths:
@@ -499,7 +519,7 @@ spec:
service:
name: {{ include "chart.fullname" $ }}
port:
number: 3000
number: {{ $.Values.service.port | default 3000 }}
{{- end }}
{{- end }}
{{- end }}
@@ -510,26 +530,26 @@ spec:
}
/// Packages a Helm chart directory into a .tgz file.
fn package_helm_chart(
&self,
chart_dir: &PathBuf,
) -> Result<PathBuf, Box<dyn std::error::Error>> {
fn package_helm_chart(&self, chart_dir: &Path) -> Result<PathBuf, Box<dyn std::error::Error>> {
let chart_dirname = chart_dir.file_name().expect("Should find a chart dirname");
info!(
debug!(
"Launching `helm package {}` cli with CWD {}",
chart_dirname.to_string_lossy(),
&self.project_root.join("helm").to_string_lossy()
&self
.project_root
.join(".harmony_generated")
.join("helm")
.to_string_lossy()
);
let output = process::Command::new("helm")
.args(["package", chart_dirname.to_str().unwrap()])
.current_dir(&self.project_root.join("helm")) // Run package from the parent dir
.current_dir(self.project_root.join(".harmony_generated").join("helm")) // Run package from the parent dir
.output()?;
self.check_output(&output, "Failed to package Helm chart")?;
// Helm prints the path of the created chart to stdout.
let tgz_name = String::from_utf8(output.stdout)?
.trim()
.split_whitespace()
.last()
.unwrap_or_default()
@@ -539,20 +559,24 @@ spec:
}
// The output from helm is relative, so we join it with the execution directory.
Ok(self.project_root.join("helm").join(tgz_name))
Ok(self
.project_root
.join(".harmony_generated")
.join("helm")
.join(tgz_name))
}
/// Pushes a packaged Helm chart to an OCI registry.
fn push_helm_chart(
&self,
packaged_chart_path: &PathBuf,
packaged_chart_path: &Path,
) -> Result<String, Box<dyn std::error::Error>> {
// The chart name is the file stem of the .tgz file
let chart_file_name = packaged_chart_path.file_stem().unwrap().to_str().unwrap();
let oci_push_url = format!("oci://{}/{}", *REGISTRY_URL, *REGISTRY_PROJECT);
let oci_pull_url = format!("{oci_push_url}/{}-chart", self.name);
info!(
debug!(
"Pushing Helm chart {} to {}",
packaged_chart_path.to_string_lossy(),
oci_push_url

View File

@@ -41,6 +41,6 @@ impl<T: Topology + HelmCommand> Score<T> for CertManagerHelmScore {
}
fn name(&self) -> String {
format!("CertManagerHelmScore")
"CertManagerHelmScore".to_string()
}
}

View File

@@ -111,7 +111,7 @@ impl DhcpInterpret {
let boot_filename_outcome = match &self.score.boot_filename {
Some(boot_filename) => {
dhcp_server.set_boot_filename(&boot_filename).await?;
dhcp_server.set_boot_filename(boot_filename).await?;
Outcome::new(
InterpretStatus::SUCCESS,
format!("Dhcp Interpret Set boot filename to {boot_filename}"),
@@ -122,7 +122,7 @@ impl DhcpInterpret {
let filename_outcome = match &self.score.filename {
Some(filename) => {
dhcp_server.set_filename(&filename).await?;
dhcp_server.set_filename(filename).await?;
Outcome::new(
InterpretStatus::SUCCESS,
format!("Dhcp Interpret Set filename to {filename}"),
@@ -133,7 +133,7 @@ impl DhcpInterpret {
let filename64_outcome = match &self.score.filename64 {
Some(filename64) => {
dhcp_server.set_filename64(&filename64).await?;
dhcp_server.set_filename64(filename64).await?;
Outcome::new(
InterpretStatus::SUCCESS,
format!("Dhcp Interpret Set filename64 to {filename64}"),
@@ -144,7 +144,7 @@ impl DhcpInterpret {
let filenameipxe_outcome = match &self.score.filenameipxe {
Some(filenameipxe) => {
dhcp_server.set_filenameipxe(&filenameipxe).await?;
dhcp_server.set_filenameipxe(filenameipxe).await?;
Outcome::new(
InterpretStatus::SUCCESS,
format!("Dhcp Interpret Set filenameipxe to {filenameipxe}"),
@@ -209,7 +209,7 @@ impl<T: DhcpServer> Interpret<T> for DhcpInterpret {
Ok(Outcome::new(
InterpretStatus::SUCCESS,
format!("Dhcp Interpret execution successful"),
"Dhcp Interpret execution successful".to_string(),
))
}
}

View File

@@ -112,7 +112,7 @@ impl<T: Topology + DnsServer> Interpret<T> for DnsInterpret {
Ok(Outcome::new(
InterpretStatus::SUCCESS,
format!("Dns Interpret execution successful"),
"Dns Interpret execution successful".to_string(),
))
}
}

View File

@@ -55,7 +55,7 @@ impl<T: Topology + HelmCommand> Score<T> for HelmChartScore {
}
fn name(&self) -> String {
format!("{} {} HelmChartScore", self.release_name, self.chart_name)
format!("{} [HelmChartScore]", self.release_name)
}
}
@@ -90,14 +90,10 @@ impl HelmChartInterpret {
);
match add_output.status.success() {
true => {
return Ok(());
}
false => {
return Err(InterpretError::new(format!(
"Failed to add helm repository!\n{full_output}"
)));
}
true => Ok(()),
false => Err(InterpretError::new(format!(
"Failed to add helm repository!\n{full_output}"
))),
}
}
}
@@ -212,7 +208,7 @@ impl<T: Topology + HelmCommand> Interpret<T> for HelmChartInterpret {
}
let res = helm_executor.install_or_upgrade(
&ns,
ns,
&self.score.release_name,
&self.score.chart_name,
self.score.chart_version.as_ref(),
@@ -229,24 +225,27 @@ impl<T: Topology + HelmCommand> Interpret<T> for HelmChartInterpret {
match status {
helm_wrapper_rs::HelmDeployStatus::Deployed => Ok(Outcome::new(
InterpretStatus::SUCCESS,
"Helm Chart deployed".to_string(),
format!("Helm Chart {} deployed", self.score.release_name),
)),
helm_wrapper_rs::HelmDeployStatus::PendingInstall => Ok(Outcome::new(
InterpretStatus::RUNNING,
"Helm Chart Pending install".to_string(),
format!("Helm Chart {} pending install...", self.score.release_name),
)),
helm_wrapper_rs::HelmDeployStatus::PendingUpgrade => Ok(Outcome::new(
InterpretStatus::RUNNING,
"Helm Chart pending upgrade".to_string(),
)),
helm_wrapper_rs::HelmDeployStatus::Failed => Err(InterpretError::new(
"Failed to install helm chart".to_string(),
format!("Helm Chart {} pending upgrade...", self.score.release_name),
)),
helm_wrapper_rs::HelmDeployStatus::Failed => Err(InterpretError::new(format!(
"Helm Chart {} installation failed",
self.score.release_name
))),
}
}
fn get_name(&self) -> InterpretName {
todo!()
InterpretName::HelmChart
}
fn get_version(&self) -> Version {
todo!()
}

View File

@@ -77,14 +77,11 @@ impl HelmCommandExecutor {
)?;
}
let out = match self.clone().run_command(
let out = self.clone().run_command(
self.chart
.clone()
.helm_args(self.globals.chart_home.clone().unwrap()),
) {
Ok(out) => out,
Err(e) => return Err(e),
};
)?;
// TODO: don't use unwrap here
let s = String::from_utf8(out.stdout).unwrap();
@@ -98,14 +95,11 @@ impl HelmCommandExecutor {
}
pub fn version(self) -> Result<String, std::io::Error> {
let out = match self.run_command(vec![
let out = self.run_command(vec![
"version".to_string(),
"-c".to_string(),
"--short".to_string(),
]) {
Ok(out) => out,
Err(e) => return Err(e),
};
])?;
// TODO: don't use unwrap
Ok(String::from_utf8(out.stdout).unwrap())
@@ -129,15 +123,11 @@ impl HelmCommandExecutor {
None => PathBuf::from(TempDir::new()?.path()),
};
match self.chart.values_inline {
Some(yaml_str) => {
let tf: TempFile;
tf = temp_file::with_contents(yaml_str.as_bytes());
self.chart
.additional_values_files
.push(PathBuf::from(tf.path()));
}
None => (),
if let Some(yaml_str) = self.chart.values_inline {
let tf: TempFile = temp_file::with_contents(yaml_str.as_bytes());
self.chart
.additional_values_files
.push(PathBuf::from(tf.path()));
};
self.env.insert(
@@ -180,9 +170,9 @@ impl HelmChart {
match self.repo {
Some(r) => {
if r.starts_with("oci://") {
args.push(String::from(
args.push(
r.trim_end_matches("/").to_string() + "/" + self.name.clone().as_str(),
));
);
} else {
args.push("--repo".to_string());
args.push(r.to_string());
@@ -193,12 +183,9 @@ impl HelmChart {
None => args.push(self.name),
};
match self.version {
Some(v) => {
args.push("--version".to_string());
args.push(v.to_string());
}
None => (),
if let Some(v) = self.version {
args.push("--version".to_string());
args.push(v.to_string());
}
args
@@ -362,7 +349,7 @@ impl<T: Topology + K8sclient + HelmCommand> Interpret<T> for HelmChartInterpretV
}
fn get_name(&self) -> InterpretName {
todo!()
InterpretName::HelmCommand
}
fn get_version(&self) -> Version {
todo!()

View File

@@ -10,14 +10,25 @@ use crate::{
topology::{HttpServer, Topology, Url},
};
/// Configure an HTTP server that is provided by the Topology
///
/// This Score will let you easily specify a file path to be served by the HTTP server
///
/// For example, if you have a folder of assets at `/var/www/assets` simply do :
///
/// ```rust,ignore
/// StaticFilesHttpScore {
/// files_to_serve: url!("file:///var/www/assets"),
/// }
/// ```
#[derive(Debug, new, Clone, Serialize)]
pub struct HttpScore {
pub struct StaticFilesHttpScore {
files_to_serve: Url,
}
impl<T: Topology + HttpServer> Score<T> for HttpScore {
impl<T: Topology + HttpServer> Score<T> for StaticFilesHttpScore {
fn create_interpret(&self) -> Box<dyn Interpret<T>> {
Box::new(HttpInterpret::new(self.clone()))
Box::new(StaticFilesHttpInterpret::new(self.clone()))
}
fn name(&self) -> String {
@@ -26,12 +37,12 @@ impl<T: Topology + HttpServer> Score<T> for HttpScore {
}
#[derive(Debug, new, Clone)]
pub struct HttpInterpret {
score: HttpScore,
pub struct StaticFilesHttpInterpret {
score: StaticFilesHttpScore,
}
#[async_trait]
impl<T: Topology + HttpServer> Interpret<T> for HttpInterpret {
impl<T: Topology + HttpServer> Interpret<T> for StaticFilesHttpInterpret {
async fn execute(
&self,
_inventory: &Inventory,

View File

@@ -1,7 +1,7 @@
use std::path::PathBuf;
use async_trait::async_trait;
use log::info;
use log::debug;
use serde::Serialize;
use crate::{
@@ -36,7 +36,7 @@ impl<T: Topology> Score<T> for K3DInstallationScore {
}
fn name(&self) -> String {
todo!()
"K3dInstallationScore".into()
}
}
@@ -56,14 +56,15 @@ impl<T: Topology> Interpret<T> for K3dInstallationInterpret {
self.score.installation_path.clone(),
Some(self.score.cluster_name.clone()),
);
match k3d.ensure_installed().await {
Ok(_client) => {
let msg = format!("k3d cluster {} is installed ", self.score.cluster_name);
info!("{msg}");
let msg = format!("k3d cluster '{}' installed ", self.score.cluster_name);
debug!("{msg}");
Ok(Outcome::success(msg))
}
Err(msg) => Err(InterpretError::new(format!(
"K3dInstallationInterpret failed to ensure k3d is installed : {msg}"
"failed to ensure k3d is installed : {msg}"
))),
}
}

View File

@@ -89,7 +89,7 @@ where
))
}
fn get_name(&self) -> InterpretName {
todo!()
InterpretName::K8sResource
}
fn get_version(&self) -> Version {
todo!()

View File

@@ -128,13 +128,12 @@ impl<T: Topology + K8sclient + HelmCommand> Interpret<T> for LAMPInterpret {
info!("Deploying score {deployment_score:#?}");
deployment_score
.create_interpret()
.execute(inventory, topology)
.await?;
deployment_score.interpret(inventory, topology).await?;
info!("LAMP deployment_score {deployment_score:?}");
let ingress_path = ingress_path!("/");
let lamp_ingress = K8sIngressScore {
name: fqdn!("lamp-ingress"),
host: fqdn!("test"),
@@ -144,17 +143,14 @@ impl<T: Topology + K8sclient + HelmCommand> Interpret<T> for LAMPInterpret {
.as_str()
),
port: 8080,
path: Some(ingress_path!("/")),
path: Some(ingress_path),
path_type: None,
namespace: self
.get_namespace()
.map(|nbs| fqdn!(nbs.to_string().as_str())),
};
lamp_ingress
.create_interpret()
.execute(inventory, topology)
.await?;
lamp_ingress.interpret(inventory, topology).await?;
info!("LAMP lamp_ingress {lamp_ingress:?}");
@@ -164,7 +160,7 @@ impl<T: Topology + K8sclient + HelmCommand> Interpret<T> for LAMPInterpret {
}
fn get_name(&self) -> InterpretName {
todo!()
InterpretName::Lamp
}
fn get_version(&self) -> Version {
@@ -213,7 +209,7 @@ impl LAMPInterpret {
repository: None,
};
score.create_interpret().execute(inventory, topology).await
score.interpret(inventory, topology).await
}
fn build_dockerfile(&self, score: &LAMPScore) -> Result<PathBuf, Box<dyn std::error::Error>> {
let mut dockerfile = Dockerfile::new();

View File

@@ -1,12 +1,24 @@
use std::any::Any;
use std::collections::BTreeMap;
use async_trait::async_trait;
use k8s_openapi::api::core::v1::Secret;
use kube::api::ObjectMeta;
use serde::Serialize;
use serde_json::json;
use serde_yaml::{Mapping, Value};
use crate::modules::monitoring::kube_prometheus::crd::crd_alertmanager_config::{
AlertmanagerConfig, AlertmanagerConfigSpec, CRDPrometheus,
};
use crate::{
interpret::{InterpretError, Outcome},
modules::monitoring::kube_prometheus::{
prometheus::{Prometheus, PrometheusReceiver},
types::{AlertChannelConfig, AlertManagerChannelConfig},
modules::monitoring::{
kube_prometheus::{
prometheus::{KubePrometheus, KubePrometheusReceiver},
types::{AlertChannelConfig, AlertManagerChannelConfig},
},
prometheus::prometheus::{Prometheus, PrometheusReceiver},
},
topology::{Url, oberservability::monitoring::AlertReceiver},
};
@@ -17,14 +29,98 @@ pub struct DiscordWebhook {
pub url: Url,
}
#[async_trait]
impl AlertReceiver<CRDPrometheus> for DiscordWebhook {
async fn install(&self, sender: &CRDPrometheus) -> Result<Outcome, InterpretError> {
let ns = sender.namespace.clone();
let secret_name = format!("{}-secret", self.name.clone());
let webhook_key = format!("{}", self.url.clone());
let mut string_data = BTreeMap::new();
string_data.insert("webhook-url".to_string(), webhook_key.clone());
let secret = Secret {
metadata: kube::core::ObjectMeta {
name: Some(secret_name.clone()),
..Default::default()
},
string_data: Some(string_data),
type_: Some("Opaque".to_string()),
..Default::default()
};
let _ = sender.client.apply(&secret, Some(&ns)).await;
let spec = AlertmanagerConfigSpec {
data: json!({
"route": {
"receiver": self.name,
},
"receivers": [
{
"name": self.name,
"discordConfigs": [
{
"apiURL": {
"name": secret_name,
"key": "webhook-url",
},
"title": "{{ template \"discord.default.title\" . }}",
"message": "{{ template \"discord.default.message\" . }}"
}
]
}
]
}),
};
let alertmanager_configs = AlertmanagerConfig {
metadata: ObjectMeta {
name: Some(self.name.clone()),
labels: Some(std::collections::BTreeMap::from([(
"alertmanagerConfig".to_string(),
"enabled".to_string(),
)])),
namespace: Some(ns),
..Default::default()
},
spec,
};
sender
.client
.apply(&alertmanager_configs, Some(&sender.namespace))
.await?;
Ok(Outcome::success(format!(
"installed crd-alertmanagerconfigs for {}",
self.name
)))
}
fn name(&self) -> String {
"discord-webhook".to_string()
}
fn clone_box(&self) -> Box<dyn AlertReceiver<CRDPrometheus>> {
Box::new(self.clone())
}
fn as_any(&self) -> &dyn Any {
self
}
}
#[async_trait]
impl AlertReceiver<Prometheus> for DiscordWebhook {
async fn install(&self, sender: &Prometheus) -> Result<Outcome, InterpretError> {
sender.install_receiver(self).await
}
fn name(&self) -> String {
"discord-webhook".to_string()
}
fn clone_box(&self) -> Box<dyn AlertReceiver<Prometheus>> {
Box::new(self.clone())
}
fn as_any(&self) -> &dyn Any {
self
}
}
#[async_trait]
@@ -37,6 +133,32 @@ impl PrometheusReceiver for DiscordWebhook {
}
}
#[async_trait]
impl AlertReceiver<KubePrometheus> for DiscordWebhook {
async fn install(&self, sender: &KubePrometheus) -> Result<Outcome, InterpretError> {
sender.install_receiver(self).await
}
fn clone_box(&self) -> Box<dyn AlertReceiver<KubePrometheus>> {
Box::new(self.clone())
}
fn name(&self) -> String {
"discord-webhook".to_string()
}
fn as_any(&self) -> &dyn Any {
self
}
}
#[async_trait]
impl KubePrometheusReceiver 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 {

View File

@@ -1,12 +1,23 @@
use std::any::Any;
use async_trait::async_trait;
use kube::api::ObjectMeta;
use log::debug;
use serde::Serialize;
use serde_json::json;
use serde_yaml::{Mapping, Value};
use crate::{
interpret::{InterpretError, Outcome},
modules::monitoring::kube_prometheus::{
prometheus::{Prometheus, PrometheusReceiver},
types::{AlertChannelConfig, AlertManagerChannelConfig},
modules::monitoring::{
kube_prometheus::{
crd::crd_alertmanager_config::{
AlertmanagerConfig, AlertmanagerConfigSpec, CRDPrometheus,
},
prometheus::{KubePrometheus, KubePrometheusReceiver},
types::{AlertChannelConfig, AlertManagerChannelConfig},
},
prometheus::prometheus::{Prometheus, PrometheusReceiver},
},
topology::{Url, oberservability::monitoring::AlertReceiver},
};
@@ -17,14 +28,81 @@ pub struct WebhookReceiver {
pub url: Url,
}
#[async_trait]
impl AlertReceiver<CRDPrometheus> for WebhookReceiver {
async fn install(&self, sender: &CRDPrometheus) -> Result<Outcome, InterpretError> {
let spec = AlertmanagerConfigSpec {
data: json!({
"route": {
"receiver": self.name,
},
"receivers": [
{
"name": self.name,
"webhookConfigs": [
{
"url": self.url,
}
]
}
]
}),
};
let alertmanager_configs = AlertmanagerConfig {
metadata: ObjectMeta {
name: Some(self.name.clone()),
labels: Some(std::collections::BTreeMap::from([(
"alertmanagerConfig".to_string(),
"enabled".to_string(),
)])),
namespace: Some(sender.namespace.clone()),
..Default::default()
},
spec,
};
debug!(
"alert manager configs: \n{:#?}",
alertmanager_configs.clone()
);
sender
.client
.apply(&alertmanager_configs, Some(&sender.namespace))
.await?;
Ok(Outcome::success(format!(
"installed crd-alertmanagerconfigs for {}",
self.name
)))
}
fn name(&self) -> String {
"webhook-receiver".to_string()
}
fn clone_box(&self) -> Box<dyn AlertReceiver<CRDPrometheus>> {
Box::new(self.clone())
}
fn as_any(&self) -> &dyn Any {
self
}
}
#[async_trait]
impl AlertReceiver<Prometheus> for WebhookReceiver {
async fn install(&self, sender: &Prometheus) -> Result<Outcome, InterpretError> {
sender.install_receiver(self).await
}
fn name(&self) -> String {
"webhook-receiver".to_string()
}
fn clone_box(&self) -> Box<dyn AlertReceiver<Prometheus>> {
Box::new(self.clone())
}
fn as_any(&self) -> &dyn Any {
self
}
}
#[async_trait]
@@ -36,6 +114,31 @@ impl PrometheusReceiver for WebhookReceiver {
self.get_config().await
}
}
#[async_trait]
impl AlertReceiver<KubePrometheus> for WebhookReceiver {
async fn install(&self, sender: &KubePrometheus) -> Result<Outcome, InterpretError> {
sender.install_receiver(self).await
}
fn name(&self) -> String {
"webhook-receiver".to_string()
}
fn clone_box(&self) -> Box<dyn AlertReceiver<KubePrometheus>> {
Box::new(self.clone())
}
fn as_any(&self) -> &dyn Any {
self
}
}
#[async_trait]
impl KubePrometheusReceiver 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 {

View File

@@ -5,17 +5,30 @@ use serde::Serialize;
use crate::{
interpret::{InterpretError, Outcome},
modules::monitoring::kube_prometheus::{
prometheus::{Prometheus, PrometheusRule},
types::{AlertGroup, AlertManagerAdditionalPromRules},
modules::monitoring::{
kube_prometheus::{
prometheus::{KubePrometheus, KubePrometheusRule},
types::{AlertGroup, AlertManagerAdditionalPromRules},
},
prometheus::prometheus::{Prometheus, PrometheusRule},
},
topology::oberservability::monitoring::AlertRule,
};
#[async_trait]
impl AlertRule<KubePrometheus> for AlertManagerRuleGroup {
async fn install(&self, sender: &KubePrometheus) -> Result<Outcome, InterpretError> {
sender.install_rule(self).await
}
fn clone_box(&self) -> Box<dyn AlertRule<KubePrometheus>> {
Box::new(self.clone())
}
}
#[async_trait]
impl AlertRule<Prometheus> for AlertManagerRuleGroup {
async fn install(&self, sender: &Prometheus) -> Result<Outcome, InterpretError> {
sender.install_rule(&self).await
sender.install_rule(self).await
}
fn clone_box(&self) -> Box<dyn AlertRule<Prometheus>> {
Box::new(self.clone())
@@ -41,6 +54,25 @@ impl PrometheusRule for AlertManagerRuleGroup {
}
}
}
#[async_trait]
impl KubePrometheusRule 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 {

View File

@@ -0,0 +1,91 @@
use std::sync::Arc;
use async_trait::async_trait;
use serde::Serialize;
use crate::{
data::{Id, Version},
interpret::{Interpret, InterpretError, InterpretName, InterpretStatus, Outcome},
inventory::Inventory,
modules::{
application::Application,
monitoring::kube_prometheus::crd::crd_alertmanager_config::CRDPrometheus,
prometheus::prometheus::PrometheusApplicationMonitoring,
},
score::Score,
topology::{PreparationOutcome, Topology, oberservability::monitoring::AlertReceiver},
};
#[derive(Debug, Clone, Serialize)]
pub struct ApplicationMonitoringScore {
pub sender: CRDPrometheus,
pub application: Arc<dyn Application>,
pub receivers: Vec<Box<dyn AlertReceiver<CRDPrometheus>>>,
}
impl<T: Topology + PrometheusApplicationMonitoring<CRDPrometheus>> Score<T>
for ApplicationMonitoringScore
{
fn create_interpret(&self) -> Box<dyn Interpret<T>> {
Box::new(ApplicationMonitoringInterpret {
score: self.clone(),
})
}
fn name(&self) -> String {
format!(
"{} monitoring [ApplicationMonitoringScore]",
self.application.name()
)
}
}
#[derive(Debug)]
pub struct ApplicationMonitoringInterpret {
score: ApplicationMonitoringScore,
}
#[async_trait]
impl<T: Topology + PrometheusApplicationMonitoring<CRDPrometheus>> Interpret<T>
for ApplicationMonitoringInterpret
{
async fn execute(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
let result = topology
.install_prometheus(
&self.score.sender,
inventory,
Some(self.score.receivers.clone()),
)
.await;
match result {
Ok(outcome) => match outcome {
PreparationOutcome::Success { details: _ } => {
Ok(Outcome::success("Prometheus installed".into()))
}
PreparationOutcome::Noop => Ok(Outcome::noop()),
},
Err(err) => Err(InterpretError::from(err)),
}
}
fn get_name(&self) -> InterpretName {
InterpretName::ApplicationMonitoring
}
fn get_version(&self) -> Version {
todo!()
}
fn get_status(&self) -> InterpretStatus {
todo!()
}
fn get_children(&self) -> Vec<Id> {
todo!()
}
}

View File

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

View File

@@ -0,0 +1,27 @@
use non_blank_string_rs::NonBlankString;
use std::str::FromStr;
use crate::modules::helm::chart::HelmChartScore;
pub fn grafana_helm_chart_score(ns: &str) -> HelmChartScore {
let values = r#"
rbac:
namespaced: true
sidecar:
dashboards:
enabled: true
"#
.to_string();
HelmChartScore {
namespace: Some(NonBlankString::from_str(ns).unwrap()),
release_name: NonBlankString::from_str("grafana").unwrap(),
chart_name: NonBlankString::from_str("oci://ghcr.io/grafana/helm-charts/grafana").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 @@
pub mod helm_grafana;

View File

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

View File

@@ -0,0 +1,50 @@
use std::sync::Arc;
use kube::CustomResource;
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use crate::topology::{
k8s::K8sClient,
oberservability::monitoring::{AlertReceiver, AlertSender},
};
#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[kube(
group = "monitoring.coreos.com",
version = "v1alpha1",
kind = "AlertmanagerConfig",
plural = "alertmanagerconfigs",
namespaced
)]
pub struct AlertmanagerConfigSpec {
#[serde(flatten)]
pub data: serde_json::Value,
}
#[derive(Debug, Clone, Serialize)]
pub struct CRDPrometheus {
pub namespace: String,
pub client: Arc<K8sClient>,
}
impl AlertSender for CRDPrometheus {
fn name(&self) -> String {
"CRDAlertManager".to_string()
}
}
impl Clone for Box<dyn AlertReceiver<CRDPrometheus>> {
fn clone(&self) -> Self {
self.clone_box()
}
}
impl Serialize for Box<dyn AlertReceiver<CRDPrometheus>> {
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{
todo!()
}
}

View File

@@ -0,0 +1,52 @@
use kube::CustomResource;
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use super::crd_prometheuses::LabelSelector;
/// Rust CRD for `Alertmanager` from Prometheus Operator
#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[kube(
group = "monitoring.coreos.com",
version = "v1",
kind = "Alertmanager",
plural = "alertmanagers",
namespaced
)]
#[serde(rename_all = "camelCase")]
pub struct AlertmanagerSpec {
/// Number of replicas for HA
pub replicas: i32,
/// Selectors for AlertmanagerConfig CRDs
#[serde(default, skip_serializing_if = "Option::is_none")]
pub alertmanager_config_selector: Option<LabelSelector>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub alertmanager_config_namespace_selector: Option<LabelSelector>,
/// Optional pod template metadata (annotations, labels)
#[serde(default, skip_serializing_if = "Option::is_none")]
pub pod_metadata: Option<LabelSelector>,
/// Optional topology spread settings
#[serde(default, skip_serializing_if = "Option::is_none")]
pub version: Option<String>,
}
impl Default for AlertmanagerSpec {
fn default() -> Self {
AlertmanagerSpec {
replicas: 1,
// Match all AlertmanagerConfigs in the same namespace
alertmanager_config_namespace_selector: None,
// Empty selector matches all AlertmanagerConfigs in that namespace
alertmanager_config_selector: Some(LabelSelector::default()),
pod_metadata: None,
version: None,
}
}
}

View File

@@ -0,0 +1,25 @@
use crate::modules::prometheus::alerts::k8s::{
deployment::alert_deployment_unavailable,
pod::{alert_container_restarting, alert_pod_not_ready, pod_failed},
pvc::high_pvc_fill_rate_over_two_days,
service::alert_service_down,
};
use super::crd_prometheus_rules::Rule;
pub fn build_default_application_rules() -> Vec<Rule> {
let pod_failed: Rule = pod_failed().into();
let container_restarting: Rule = alert_container_restarting().into();
let pod_not_ready: Rule = alert_pod_not_ready().into();
let service_down: Rule = alert_service_down().into();
let deployment_unavailable: Rule = alert_deployment_unavailable().into();
let high_pvc_fill_rate: Rule = high_pvc_fill_rate_over_two_days().into();
vec![
pod_failed,
container_restarting,
pod_not_ready,
service_down,
deployment_unavailable,
high_pvc_fill_rate,
]
}

View File

@@ -0,0 +1,153 @@
use std::collections::BTreeMap;
use kube::CustomResource;
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use super::crd_prometheuses::LabelSelector;
#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[kube(
group = "grafana.integreatly.org",
version = "v1beta1",
kind = "Grafana",
plural = "grafanas",
namespaced
)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaSpec {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub config: Option<GrafanaConfig>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub admin_user: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub admin_password: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub ingress: Option<GrafanaIngress>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub persistence: Option<GrafanaPersistence>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub resources: Option<ResourceRequirements>,
}
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaConfig {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub log: Option<GrafanaLogConfig>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub security: Option<GrafanaSecurityConfig>,
}
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaLogConfig {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub mode: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub level: Option<String>,
}
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaSecurityConfig {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub admin_user: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub admin_password: Option<String>,
}
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaIngress {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub enabled: Option<bool>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub hosts: Option<Vec<String>>,
}
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaPersistence {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub enabled: Option<bool>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub storage_class_name: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub size: Option<String>,
}
// ------------------------------------------------------------------------------------------------
#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[kube(
group = "grafana.integreatly.org",
version = "v1beta1",
kind = "GrafanaDashboard",
plural = "grafanadashboards",
namespaced
)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaDashboardSpec {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub resync_period: Option<String>,
pub instance_selector: LabelSelector,
pub json: String,
}
// ------------------------------------------------------------------------------------------------
#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[kube(
group = "grafana.integreatly.org",
version = "v1beta1",
kind = "GrafanaDatasource",
plural = "grafanadatasources",
namespaced
)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaDatasourceSpec {
pub instance_selector: LabelSelector,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub allow_cross_namespace_import: Option<bool>,
pub datasource: GrafanaDatasourceConfig,
}
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct GrafanaDatasourceConfig {
pub access: String,
pub database: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub json_data: Option<BTreeMap<String, String>>,
pub name: String,
pub r#type: String,
pub url: String,
}
// ------------------------------------------------------------------------------------------------
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema, Default)]
#[serde(rename_all = "camelCase")]
pub struct ResourceRequirements {
#[serde(default, skip_serializing_if = "BTreeMap::is_empty")]
pub limits: BTreeMap<String, String>,
#[serde(default, skip_serializing_if = "BTreeMap::is_empty")]
pub requests: BTreeMap<String, String>,
}

View File

@@ -0,0 +1,57 @@
use std::collections::BTreeMap;
use kube::CustomResource;
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use crate::modules::monitoring::alert_rule::prometheus_alert_rule::PrometheusAlertRule;
#[derive(CustomResource, Debug, Serialize, Deserialize, Clone, JsonSchema)]
#[kube(
group = "monitoring.coreos.com",
version = "v1",
kind = "PrometheusRule",
plural = "prometheusrules",
namespaced
)]
#[serde(rename_all = "camelCase")]
pub struct PrometheusRuleSpec {
pub groups: Vec<RuleGroup>,
}
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
pub struct RuleGroup {
pub name: String,
pub rules: Vec<Rule>,
}
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct Rule {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub alert: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub expr: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub for_: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub labels: Option<std::collections::BTreeMap<String, String>>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub annotations: Option<std::collections::BTreeMap<String, String>>,
}
impl From<PrometheusAlertRule> for Rule {
fn from(value: PrometheusAlertRule) -> Self {
Rule {
alert: Some(value.alert),
expr: Some(value.expr),
for_: value.r#for,
labels: Some(value.labels.into_iter().collect::<BTreeMap<_, _>>()),
annotations: Some(value.annotations.into_iter().collect::<BTreeMap<_, _>>()),
}
}
}

View File

@@ -0,0 +1,118 @@
use std::collections::BTreeMap;
use kube::CustomResource;
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use crate::modules::monitoring::kube_prometheus::types::Operator;
#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[kube(
group = "monitoring.coreos.com",
version = "v1",
kind = "Prometheus",
plural = "prometheuses",
namespaced
)]
#[serde(rename_all = "camelCase")]
pub struct PrometheusSpec {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub alerting: Option<PrometheusSpecAlerting>,
pub service_account_name: String,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub service_monitor_namespace_selector: Option<LabelSelector>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub service_monitor_selector: Option<LabelSelector>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub service_discovery_role: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub pod_monitor_selector: Option<LabelSelector>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub rule_selector: Option<LabelSelector>,
#[serde(default, skip_serializing_if = "Option::is_none")]
pub rule_namespace_selector: Option<LabelSelector>,
}
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema, Default)]
#[serde(rename_all = "camelCase")]
pub struct NamespaceSelector {
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub match_names: Vec<String>,
}
/// Contains alerting configuration, specifically Alertmanager endpoints.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema, Default)]
pub struct PrometheusSpecAlerting {
#[serde(skip_serializing_if = "Option::is_none")]
pub alertmanagers: Option<Vec<AlertmanagerEndpoints>>,
}
/// Represents an Alertmanager endpoint configuration used by Prometheus.
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema, Default)]
pub struct AlertmanagerEndpoints {
/// Name of the Alertmanager Service.
#[serde(skip_serializing_if = "Option::is_none")]
pub name: Option<String>,
/// Namespace of the Alertmanager Service.
#[serde(skip_serializing_if = "Option::is_none")]
pub namespace: Option<String>,
/// Port to access on the Alertmanager Service (e.g. "web").
#[serde(skip_serializing_if = "Option::is_none")]
pub port: Option<String>,
/// Scheme to use for connecting (e.g. "http").
#[serde(skip_serializing_if = "Option::is_none")]
pub scheme: Option<String>,
// Other fields like `tls_config`, `path_prefix`, etc., can be added if needed.
}
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema, Default)]
#[serde(rename_all = "camelCase")]
pub struct LabelSelector {
#[serde(default, skip_serializing_if = "BTreeMap::is_empty")]
pub match_labels: BTreeMap<String, String>,
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub match_expressions: Vec<LabelSelectorRequirement>,
}
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct LabelSelectorRequirement {
pub key: String,
pub operator: Operator,
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub values: Vec<String>,
}
impl Default for PrometheusSpec {
fn default() -> Self {
PrometheusSpec {
alerting: None,
service_account_name: "prometheus".into(),
// null means "only my namespace"
service_monitor_namespace_selector: None,
// empty selector means match all ServiceMonitors in that namespace
service_monitor_selector: Some(LabelSelector::default()),
service_discovery_role: Some("Endpoints".into()),
pod_monitor_selector: None,
rule_selector: None,
rule_namespace_selector: Some(LabelSelector::default()),
}
}
}

View File

@@ -0,0 +1,203 @@
pub fn build_default_dashboard(namespace: &str) -> String {
let dashboard = format!(
r#"{{
"annotations": {{
"list": []
}},
"editable": true,
"gnetId": null,
"graphTooltip": 0,
"id": null,
"iteration": 171105,
"panels": [
{{
"datasource": "$datasource",
"fieldConfig": {{
"defaults": {{
"unit": "short"
}},
"overrides": []
}},
"gridPos": {{
"h": 6,
"w": 6,
"x": 0,
"y": 0
}},
"id": 1,
"options": {{
"reduceOptions": {{
"calcs": ["lastNotNull"],
"fields": "",
"values": false
}}
}},
"pluginVersion": "9.0.0",
"targets": [
{{
"expr": "sum(kube_pod_status_phase{{namespace=\"{namespace}\", phase=\"Running\"}})",
"legendFormat": "",
"refId": "A"
}}
],
"title": "Pods in Namespace",
"type": "stat"
}},
{{
"datasource": "$datasource",
"fieldConfig": {{
"defaults": {{
"unit": "short"
}},
"overrides": []
}},
"gridPos": {{
"h": 6,
"w": 6,
"x": 6,
"y": 0
}},
"id": 2,
"options": {{
"reduceOptions": {{
"calcs": ["lastNotNull"],
"fields": "",
"values": false
}}
}},
"pluginVersion": "9.0.0",
"targets": [
{{
"expr": "sum(kube_pod_status_phase{{phase=\"Failed\", namespace=\"{namespace}\"}})",
"legendFormat": "",
"refId": "A"
}}
],
"title": "Pods in Failed State",
"type": "stat"
}},
{{
"datasource": "$datasource",
"fieldConfig": {{
"defaults": {{
"unit": "percentunit"
}},
"overrides": []
}},
"gridPos": {{
"h": 6,
"w": 12,
"x": 0,
"y": 6
}},
"id": 3,
"options": {{
"reduceOptions": {{
"calcs": ["lastNotNull"],
"fields": "",
"values": false
}}
}},
"pluginVersion": "9.0.0",
"targets": [
{{
"expr": "sum(kube_deployment_status_replicas_available{{namespace=\"{namespace}\"}}) / sum(kube_deployment_spec_replicas{{namespace=\"{namespace}\"}})",
"legendFormat": "",
"refId": "A"
}}
],
"title": "Deployment Health (Available / Desired)",
"type": "stat"
}},
{{
"datasource": "$datasource",
"fieldConfig": {{
"defaults": {{
"unit": "short"
}},
"overrides": []
}},
"gridPos": {{
"h": 6,
"w": 12,
"x": 0,
"y": 12
}},
"id": 4,
"options": {{
"reduceOptions": {{
"calcs": ["lastNotNull"],
"fields": "",
"values": false
}}
}},
"pluginVersion": "9.0.0",
"targets": [
{{
"expr": "sum by(pod) (rate(kube_pod_container_status_restarts_total{{namespace=\"{namespace}\"}}[5m]))",
"legendFormat": "{{{{pod}}}}",
"refId": "A"
}}
],
"title": "Container Restarts (per pod)",
"type": "timeseries"
}},
{{
"datasource": "$datasource",
"fieldConfig": {{
"defaults": {{
"unit": "short"
}},
"overrides": []
}},
"gridPos": {{
"h": 6,
"w": 12,
"x": 0,
"y": 18
}},
"id": 5,
"options": {{
"reduceOptions": {{
"calcs": ["lastNotNull"],
"fields": "",
"values": false
}}
}},
"pluginVersion": "9.0.0",
"targets": [
{{
"expr": "sum(ALERTS{{alertstate=\"firing\", namespace=\"{namespace}\"}}) or vector(0)",
"legendFormat": "",
"refId": "A"
}}
],
"title": "Firing Alerts in Namespace",
"type": "stat"
}}
],
"schemaVersion": 36,
"templating": {{
"list": [
{{
"name": "datasource",
"type": "datasource",
"pluginId": "prometheus",
"label": "Prometheus",
"query": "prometheus",
"refresh": 1,
"hide": 0,
"current": {{
"selected": true,
"text": "Prometheus",
"value": "Prometheus"
}}
}}
]
}},
"title": "Tenant Namespace Overview",
"version": 1
}}"#
);
dashboard
}

View File

@@ -0,0 +1,20 @@
use std::str::FromStr;
use non_blank_string_rs::NonBlankString;
use crate::modules::helm::chart::HelmChartScore;
pub fn grafana_operator_helm_chart_score(ns: String) -> HelmChartScore {
HelmChartScore {
namespace: Some(NonBlankString::from_str(&ns).unwrap()),
release_name: NonBlankString::from_str("grafana_operator").unwrap(),
chart_name: NonBlankString::from_str("oci://ghcr.io/grafana/helm-charts/grafana-operator")
.unwrap(),
chart_version: None,
values_overrides: None,
values_yaml: None,
create_namespace: true,
install_only: true,
repository: None,
}
}

View File

@@ -0,0 +1,11 @@
pub mod crd_alertmanager_config;
pub mod crd_alertmanagers;
pub mod crd_default_rules;
pub mod crd_grafana;
pub mod crd_prometheus_rules;
pub mod crd_prometheuses;
pub mod grafana_default_dashboard;
pub mod grafana_operator;
pub mod prometheus_operator;
pub mod role;
pub mod service_monitor;

View File

@@ -0,0 +1,22 @@
use std::str::FromStr;
use non_blank_string_rs::NonBlankString;
use crate::modules::helm::chart::HelmChartScore;
pub fn prometheus_operator_helm_chart_score(ns: String) -> HelmChartScore {
HelmChartScore {
namespace: Some(NonBlankString::from_str(&ns).unwrap()),
release_name: NonBlankString::from_str("prometheus-operator").unwrap(),
chart_name: NonBlankString::from_str(
"oci://hub.nationtech.io/harmony/nt-prometheus-operator",
)
.unwrap(),
chart_version: None,
values_overrides: None,
values_yaml: None,
create_namespace: true,
install_only: true,
repository: None,
}
}

View File

@@ -0,0 +1,62 @@
use k8s_openapi::api::{
core::v1::ServiceAccount,
rbac::v1::{PolicyRule, Role, RoleBinding, RoleRef, Subject},
};
use kube::api::ObjectMeta;
pub fn build_prom_role(role_name: String, namespace: String) -> Role {
Role {
metadata: ObjectMeta {
name: Some(role_name),
namespace: Some(namespace),
..Default::default()
},
rules: Some(vec![PolicyRule {
api_groups: Some(vec!["".into()]), // core API group
resources: Some(vec!["services".into(), "endpoints".into(), "pods".into()]),
verbs: vec!["get".into(), "list".into(), "watch".into()],
..Default::default()
}]),
}
}
pub fn build_prom_rolebinding(
role_name: String,
namespace: String,
service_account_name: String,
) -> RoleBinding {
RoleBinding {
metadata: ObjectMeta {
name: Some(format!("{}-rolebinding", role_name)),
namespace: Some(namespace.clone()),
..Default::default()
},
role_ref: RoleRef {
api_group: "rbac.authorization.k8s.io".into(),
kind: "Role".into(),
name: role_name,
},
subjects: Some(vec![Subject {
kind: "ServiceAccount".into(),
name: service_account_name,
namespace: Some(namespace.clone()),
..Default::default()
}]),
}
}
pub fn build_prom_service_account(
service_account_name: String,
namespace: String,
) -> ServiceAccount {
ServiceAccount {
automount_service_account_token: None,
image_pull_secrets: None,
metadata: ObjectMeta {
name: Some(service_account_name),
namespace: Some(namespace),
..Default::default()
},
secrets: None,
}
}

View File

@@ -0,0 +1,87 @@
use std::collections::HashMap;
use kube::CustomResource;
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use crate::modules::monitoring::kube_prometheus::types::{
HTTPScheme, MatchExpression, NamespaceSelector, Operator, Selector,
ServiceMonitor as KubeServiceMonitor, ServiceMonitorEndpoint,
};
/// This is the top-level struct for the ServiceMonitor Custom Resource.
/// The `#[derive(CustomResource)]` macro handles all the boilerplate for you,
/// including the `impl Resource`.
#[derive(CustomResource, Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[kube(
group = "monitoring.coreos.com",
version = "v1",
kind = "ServiceMonitor",
plural = "servicemonitors",
namespaced
)]
#[serde(rename_all = "camelCase")]
pub struct ServiceMonitorSpec {
/// A label selector to select services to monitor.
pub selector: Selector,
/// A list of endpoints on the selected services to be monitored.
pub endpoints: Vec<ServiceMonitorEndpoint>,
/// Selector to select which namespaces the Kubernetes Endpoints objects
/// are discovered from.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub namespace_selector: Option<NamespaceSelector>,
/// The label to use to retrieve the job name from.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub job_label: Option<String>,
/// Pod-based target labels to transfer from the Kubernetes Pod onto the target.
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub pod_target_labels: Vec<String>,
/// TargetLabels transfers labels on the Kubernetes Service object to the target.
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub target_labels: Vec<String>,
}
impl Default for ServiceMonitorSpec {
fn default() -> Self {
let labels = HashMap::new();
Self {
selector: Selector {
match_labels: { labels },
match_expressions: vec![MatchExpression {
key: "app.kubernetes.io/name".into(),
operator: Operator::Exists,
values: vec![],
}],
},
endpoints: vec![ServiceMonitorEndpoint {
port: Some("http".to_string()),
path: Some("/metrics".into()),
interval: Some("30s".into()),
scheme: Some(HTTPScheme::HTTP),
..Default::default()
}],
namespace_selector: None, // only the same namespace
job_label: Some("app".into()),
pod_target_labels: vec![],
target_labels: vec![],
}
}
}
impl From<KubeServiceMonitor> for ServiceMonitorSpec {
fn from(value: KubeServiceMonitor) -> Self {
Self {
selector: value.selector,
endpoints: value.endpoints,
namespace_selector: value.namespace_selector,
job_label: value.job_label,
pod_target_labels: value.pod_target_labels,
target_labels: value.target_labels,
}
}
}

View File

@@ -27,6 +27,12 @@ pub struct KubePrometheusConfig {
pub alert_rules: Vec<AlertManagerAdditionalPromRules>,
pub additional_service_monitors: Vec<ServiceMonitor>,
}
impl Default for KubePrometheusConfig {
fn default() -> Self {
Self::new()
}
}
impl KubePrometheusConfig {
pub fn new() -> Self {
Self {
@@ -35,18 +41,18 @@ impl KubePrometheusConfig {
windows_monitoring: false,
alert_manager: true,
grafana: true,
node_exporter: false,
node_exporter: true,
prometheus: true,
kubernetes_service_monitors: true,
kubernetes_api_server: false,
kubernetes_api_server: true,
kubelet: true,
kube_controller_manager: false,
kube_etcd: false,
kube_proxy: false,
kube_controller_manager: true,
kube_etcd: true,
kube_proxy: true,
kube_state_metrics: true,
prometheus_operator: true,
core_dns: false,
kube_scheduler: false,
core_dns: true,
kube_scheduler: true,
alert_receiver_configs: vec![],
alert_rules: vec![],
additional_service_monitors: vec![],

View File

@@ -12,8 +12,8 @@ use crate::modules::{
helm::chart::HelmChartScore,
monitoring::kube_prometheus::types::{
AlertGroup, AlertManager, AlertManagerAdditionalPromRules, AlertManagerConfig,
AlertManagerRoute, AlertManagerSpec, AlertManagerValues, ConfigReloader, Limits,
PrometheusConfig, Requests, Resources,
AlertManagerConfigSelector, AlertManagerRoute, AlertManagerSpec, AlertManagerValues,
ConfigReloader, Limits, PrometheusConfig, Requests, Resources,
},
};
@@ -35,7 +35,7 @@ pub fn kube_prometheus_helm_chart_score(
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_operator = config.prometheus_operator.to_string();
let prometheus = config.prometheus.to_string();
let resource_limit = Resources {
limits: Limits {
@@ -64,15 +64,18 @@ pub fn kube_prometheus_helm_chart_score(
indent_lines(&yaml, indent_level + 2)
)
}
let resource_section = resource_block(&resource_limit, 2);
let _resource_section = resource_block(&resource_limit, 2);
let mut values = format!(
r#"
global:
rbac:
create: true
prometheus:
enabled: {prometheus}
prometheusSpec:
resources:
requests:
requests:
cpu: 100m
memory: 500Mi
limits:
@@ -118,7 +121,7 @@ defaultRules:
windowsMonitoring:
enabled: {windows_monitoring}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -127,13 +130,13 @@ windowsMonitoring:
grafana:
enabled: {grafana}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
cpu: 200m
memory: 250Mi
initChownData:
initChownData:
resources:
requests:
cpu: 10m
@@ -154,7 +157,7 @@ kubernetesServiceMonitors:
kubeApiServer:
enabled: {kubernetes_api_server}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -163,7 +166,7 @@ kubeApiServer:
kubelet:
enabled: {kubelet}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -172,7 +175,7 @@ kubelet:
kubeControllerManager:
enabled: {kube_controller_manager}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -181,7 +184,7 @@ kubeControllerManager:
coreDns:
enabled: {core_dns}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -190,7 +193,7 @@ coreDns:
kubeEtcd:
enabled: {kube_etcd}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -199,7 +202,7 @@ kubeEtcd:
kubeScheduler:
enabled: {kube_scheduler}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -208,7 +211,7 @@ kubeScheduler:
kubeProxy:
enabled: {kube_proxy}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -218,7 +221,7 @@ kubeStateMetrics:
enabled: {kube_state_metrics}
kube-state-metrics:
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -227,7 +230,7 @@ kube-state-metrics:
nodeExporter:
enabled: {node_exporter}
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -235,16 +238,16 @@ nodeExporter:
memory: 250Mi
prometheus-node-exporter:
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
cpu: 200m
memory: 250Mi
prometheusOperator:
enabled: {prometheus_operator}
enabled: true
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -252,7 +255,7 @@ prometheusOperator:
memory: 200Mi
prometheusConfigReloader:
resources:
requests:
requests:
cpu: 100m
memory: 150Mi
limits:
@@ -264,7 +267,7 @@ prometheusOperator:
limits:
cpu: 10m
memory: 100Mi
requests:
requests:
cpu: 10m
memory: 100Mi
patch:
@@ -272,7 +275,7 @@ prometheusOperator:
limits:
cpu: 10m
memory: 100Mi
requests:
requests:
cpu: 10m
memory: 100Mi
"#,
@@ -329,6 +332,11 @@ prometheusOperator:
.push(receiver.channel_receiver.clone());
}
let mut labels = BTreeMap::new();
labels.insert("alertmanagerConfig".to_string(), "enabled".to_string());
let alert_manager_config_selector = AlertManagerConfigSelector {
match_labels: labels,
};
let alert_manager_values = AlertManagerValues {
alertmanager: AlertManager {
enabled: config.alert_manager,
@@ -344,6 +352,8 @@ prometheusOperator:
cpu: "100m".to_string(),
},
},
alert_manager_config_selector,
replicas: 2,
},
init_config_reloader: ConfigReloader {
resources: Resources {

View File

@@ -2,7 +2,7 @@ use std::sync::{Arc, Mutex};
use serde::Serialize;
use super::{helm::config::KubePrometheusConfig, prometheus::Prometheus};
use super::{helm::config::KubePrometheusConfig, prometheus::KubePrometheus};
use crate::{
modules::monitoring::kube_prometheus::types::ServiceMonitor,
score::Score,
@@ -15,8 +15,8 @@ use crate::{
#[derive(Clone, Debug, Serialize)]
pub struct HelmPrometheusAlertingScore {
pub receivers: Vec<Box<dyn AlertReceiver<Prometheus>>>,
pub rules: Vec<Box<dyn AlertRule<Prometheus>>>,
pub receivers: Vec<Box<dyn AlertReceiver<KubePrometheus>>>,
pub rules: Vec<Box<dyn AlertRule<KubePrometheus>>>,
pub service_monitors: Vec<ServiceMonitor>,
}
@@ -28,7 +28,7 @@ impl<T: Topology + HelmCommand + TenantManager> Score<T> for HelmPrometheusAlert
.expect("couldn't lock config")
.additional_service_monitors = self.service_monitors.clone();
Box::new(AlertingInterpret {
sender: Prometheus::new(),
sender: KubePrometheus { config },
receivers: self.receivers.clone(),
rules: self.rules.clone(),
})

View File

@@ -1,3 +1,4 @@
pub mod crd;
pub mod helm;
pub mod helm_prometheus_alert_score;
pub mod prometheus;

View File

@@ -10,7 +10,7 @@ use crate::{
modules::monitoring::alert_rule::prometheus_alert_rule::AlertManagerRuleGroup,
score,
topology::{
HelmCommand, K8sAnywhereTopology, Topology,
HelmCommand, Topology,
installable::Installable,
oberservability::monitoring::{AlertReceiver, AlertRule, AlertSender},
tenant::TenantManager,
@@ -27,14 +27,14 @@ use super::{
};
#[async_trait]
impl AlertSender for Prometheus {
impl AlertSender for KubePrometheus {
fn name(&self) -> String {
"HelmKubePrometheus".to_string()
}
}
#[async_trait]
impl<T: Topology + HelmCommand + TenantManager> Installable<T> for Prometheus {
impl<T: Topology + HelmCommand + TenantManager> Installable<T> for KubePrometheus {
async fn configure(&self, _inventory: &Inventory, topology: &T) -> Result<(), InterpretError> {
self.configure_with_topology(topology).await;
Ok(())
@@ -51,11 +51,17 @@ impl<T: Topology + HelmCommand + TenantManager> Installable<T> for Prometheus {
}
#[derive(Debug)]
pub struct Prometheus {
pub struct KubePrometheus {
pub config: Arc<Mutex<KubePrometheusConfig>>,
}
impl Prometheus {
impl Default for KubePrometheus {
fn default() -> Self {
Self::new()
}
}
impl KubePrometheus {
pub fn new() -> Self {
Self {
config: Arc::new(Mutex::new(KubePrometheusConfig::new())),
@@ -75,7 +81,7 @@ impl Prometheus {
pub async fn install_receiver(
&self,
prometheus_receiver: &dyn PrometheusReceiver,
prometheus_receiver: &dyn KubePrometheusReceiver,
) -> Result<Outcome, InterpretError> {
let prom_receiver = prometheus_receiver.configure_receiver().await;
debug!(
@@ -113,19 +119,18 @@ impl Prometheus {
topology: &T,
) -> Result<Outcome, InterpretError> {
kube_prometheus_helm_chart_score(self.config.clone())
.create_interpret()
.execute(inventory, topology)
.interpret(inventory, topology)
.await
}
}
#[async_trait]
pub trait PrometheusReceiver: Send + Sync + std::fmt::Debug {
pub trait KubePrometheusReceiver: Send + Sync + std::fmt::Debug {
fn name(&self) -> String;
async fn configure_receiver(&self) -> AlertManagerChannelConfig;
}
impl Serialize for Box<dyn AlertReceiver<Prometheus>> {
impl Serialize for Box<dyn AlertReceiver<KubePrometheus>> {
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
@@ -134,19 +139,19 @@ impl Serialize for Box<dyn AlertReceiver<Prometheus>> {
}
}
impl Clone for Box<dyn AlertReceiver<Prometheus>> {
impl Clone for Box<dyn AlertReceiver<KubePrometheus>> {
fn clone(&self) -> Self {
self.clone_box()
}
}
#[async_trait]
pub trait PrometheusRule: Send + Sync + std::fmt::Debug {
pub trait KubePrometheusRule: Send + Sync + std::fmt::Debug {
fn name(&self) -> String;
async fn configure_rule(&self) -> AlertManagerAdditionalPromRules;
}
impl Serialize for Box<dyn AlertRule<Prometheus>> {
impl Serialize for Box<dyn AlertRule<KubePrometheus>> {
fn serialize<S>(&self, _serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
@@ -155,7 +160,7 @@ impl Serialize for Box<dyn AlertRule<Prometheus>> {
}
}
impl Clone for Box<dyn AlertRule<Prometheus>> {
impl Clone for Box<dyn AlertRule<KubePrometheus>> {
fn clone(&self) -> Self {
self.clone_box()
}

View File

@@ -1,7 +1,8 @@
use std::collections::{BTreeMap, HashMap};
use async_trait::async_trait;
use serde::Serialize;
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use serde_yaml::{Mapping, Sequence, Value};
use crate::modules::monitoring::alert_rule::prometheus_alert_rule::AlertManagerRuleGroup;
@@ -55,6 +56,14 @@ pub struct AlertManagerChannelConfig {
#[serde(rename_all = "camelCase")]
pub struct AlertManagerSpec {
pub(crate) resources: Resources,
pub replicas: u32,
pub alert_manager_config_selector: AlertManagerConfigSelector,
}
#[derive(Debug, Clone, Serialize)]
#[serde(rename_all = "camelCase")]
pub struct AlertManagerConfigSelector {
pub match_labels: BTreeMap<String, String>,
}
#[derive(Debug, Clone, Serialize)]
@@ -86,7 +95,7 @@ pub struct AlertGroup {
pub groups: Vec<AlertManagerRuleGroup>,
}
#[derive(Debug, Clone, Serialize)]
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
pub enum HTTPScheme {
#[serde(rename = "http")]
HTTP,
@@ -94,7 +103,7 @@ pub enum HTTPScheme {
HTTPS,
}
#[derive(Debug, Clone, Serialize)]
#[derive(Debug, Clone, Serialize, Deserialize, JsonSchema)]
pub enum Operator {
In,
NotIn,
@@ -139,80 +148,89 @@ pub struct ServiceMonitorTLSConfig {
pub server_name: Option<String>,
}
#[derive(Debug, Clone, Serialize)]
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema, Default)]
#[serde(rename_all = "camelCase")]
pub struct ServiceMonitorEndpoint {
// ## Name of the endpoint's service port
// ## Mutually exclusive with targetPort
/// Name of the service port this endpoint refers to.
pub port: Option<String>,
// ## Name or number of the endpoint's target port
// ## Mutually exclusive with port
pub target_port: Option<String>,
// ## File containing bearer token to be used when scraping targets
// ##
pub bearer_token_file: Option<String>,
// ## Interval at which metrics should be scraped
// ##
/// Interval at which metrics should be scraped.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub interval: Option<String>,
// ## HTTP path to scrape for metrics
// ##
pub path: String,
/// The HTTP path to scrape for metrics.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub path: Option<String>,
// ## HTTP scheme to use for scraping
// ##
pub scheme: HTTPScheme,
/// HTTP scheme to use for scraping.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub scheme: Option<HTTPScheme>,
// ## TLS configuration to use when scraping the endpoint
// ##
pub tls_config: Option<ServiceMonitorTLSConfig>,
/// Relabelings to apply to samples before scraping.
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub relabelings: Vec<RelabelConfig>,
// ## MetricRelabelConfigs to apply to samples after scraping, but before ingestion.
// ## ref: https://github.com/prometheus-operator/prometheus-operator/blob/main/Documentation/api-reference/api.md#relabelconfig
// ##
// # - action: keep
// # regex: 'kube_(daemonset|deployment|pod|namespace|node|statefulset).+'
// # sourceLabels: [__name__]
pub metric_relabelings: Vec<Mapping>,
// ## RelabelConfigs to apply to samples before scraping
// ## ref: https://github.com/prometheus-operator/prometheus-operator/blob/main/Documentation/api-reference/api.md#relabelconfig
// ##
// # - sourceLabels: [__meta_kubernetes_pod_node_name]
// # separator: ;
// # regex: ^(.*)$
// # targetLabel: nodename
// # replacement: $1
// # action: replace
pub relabelings: Vec<Mapping>,
/// MetricRelabelings to apply to samples after scraping, but before ingestion.
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub metric_relabelings: Vec<RelabelConfig>,
}
#[derive(Debug, Clone, Serialize)]
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct RelabelConfig {
/// The action to perform based on the regex matching.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub action: Option<String>,
/// A list of labels from which to extract values.
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub source_labels: Vec<String>,
/// Separator to be used when concatenating source_labels.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub separator: Option<String>,
/// The label to which the resulting value is written.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub target_label: Option<String>,
/// A regular expression to match against the concatenated source label values.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub regex: Option<String>,
/// The replacement value to use.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub replacement: Option<String>,
}
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema)]
#[serde(rename_all = "camelCase")]
pub struct MatchExpression {
pub key: String,
pub operator: Operator,
pub operator: Operator, // "In", "NotIn", "Exists", "DoesNotExist"
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub values: Vec<String>,
}
#[derive(Debug, Clone, Serialize)]
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema, Default)]
#[serde(rename_all = "camelCase")]
pub struct Selector {
// # label selector for services
/// A map of key-value pairs to match.
#[serde(default, skip_serializing_if = "HashMap::is_empty")]
pub match_labels: HashMap<String, String>,
/// A list of label selector requirements.
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub match_expressions: Vec<MatchExpression>,
}
#[derive(Debug, Clone, Serialize)]
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(rename_all = "camelCase")]
pub struct ServiceMonitor {
pub name: String,
// # Additional labels to set used for the ServiceMonitorSelector. Together with standard labels from the chart
pub additional_labels: Option<Mapping>,
pub additional_labels: Option<HashMap<String, String>>,
// # Service label for use in assembling a job name of the form <label value>-<port>
// # If no label is specified, the service name is used.
@@ -240,7 +258,7 @@ pub struct ServiceMonitor {
// any: bool,
// # Explicit list of namespace names to select
// matchNames: Vec,
pub namespace_selector: Option<Mapping>,
pub namespace_selector: Option<NamespaceSelector>,
// # Endpoints of the selected service to be monitored
pub endpoints: Vec<ServiceMonitorEndpoint>,
@@ -250,6 +268,18 @@ pub struct ServiceMonitor {
pub fallback_scrape_protocol: Option<String>,
}
#[derive(Serialize, Deserialize, Debug, Clone, JsonSchema, Default)]
#[serde(rename_all = "camelCase")]
pub struct NamespaceSelector {
/// Select all namespaces.
#[serde(default, skip_serializing_if = "std::ops::Not::not")]
pub any: bool,
/// List of namespace names to select from.
#[serde(default, skip_serializing_if = "Vec::is_empty")]
pub match_names: Vec<String>,
}
impl Default for ServiceMonitor {
fn default() -> Self {
Self {
@@ -268,19 +298,3 @@ impl Default for ServiceMonitor {
}
}
}
impl Default for ServiceMonitorEndpoint {
fn default() -> Self {
Self {
port: Some("80".to_string()),
target_port: Default::default(),
bearer_token_file: Default::default(),
interval: Default::default(),
path: "/metrics".to_string(),
scheme: HTTPScheme::HTTP,
tls_config: Default::default(),
metric_relabelings: Default::default(),
relabelings: Default::default(),
}
}
}

View File

@@ -1,4 +1,7 @@
pub mod alert_channel;
pub mod alert_rule;
pub mod application_monitoring;
pub mod grafana;
pub mod kube_prometheus;
pub mod ntfy;
pub mod prometheus;

View File

@@ -3,7 +3,7 @@ use std::str::FromStr;
use crate::modules::helm::chart::{HelmChartScore, HelmRepository};
pub fn ntfy_helm_chart_score(namespace: String) -> HelmChartScore {
pub fn ntfy_helm_chart_score(namespace: String, host: String) -> HelmChartScore {
let values = format!(
r#"
replicaCount: 1
@@ -28,12 +28,12 @@ service:
port: 80
ingress:
enabled: false
enabled: true
# annotations:
# kubernetes.io/ingress.class: nginx
# kubernetes.io/tls-acme: "true"
hosts:
- host: ntfy.host.com
- host: {host}
paths:
- path: /
pathType: ImplementationSpecific
@@ -58,6 +58,7 @@ config:
# web-root: "disable"
enable-signup: false
enable-login: "true"
enable-metrics: "true"
persistence:
enabled: true

View File

@@ -1,2 +1,3 @@
pub mod helm;
#[allow(clippy::module_inception)]
pub mod ntfy;

View File

@@ -17,6 +17,7 @@ use crate::{
#[derive(Debug, Clone, Serialize)]
pub struct NtfyScore {
pub namespace: String,
pub host: String,
}
impl<T: Topology + HelmCommand + K8sclient> Score<T> for NtfyScore {
@@ -27,7 +28,7 @@ impl<T: Topology + HelmCommand + K8sclient> Score<T> for NtfyScore {
}
fn name(&self) -> String {
format!("Ntfy")
"alert receiver [NtfyScore]".into()
}
}
@@ -38,31 +39,21 @@ pub struct NtfyInterpret {
#[derive(Debug, EnumString, Display)]
enum NtfyAccessMode {
#[strum(serialize = "read-write", serialize = "rw", to_string = "read-write")]
#[strum(serialize = "read-write", serialize = "rw")]
ReadWrite,
#[strum(
serialize = "read-only",
serialize = "ro",
serialize = "read",
to_string = "read-only"
)]
#[strum(serialize = "read-only", serialize = "ro", serialize = "read")]
ReadOnly,
#[strum(
serialize = "write-only",
serialize = "wo",
serialize = "write",
to_string = "write-only"
)]
#[strum(serialize = "write-only", serialize = "wo", serialize = "write")]
WriteOnly,
#[strum(serialize = "none", to_string = "deny")]
#[strum(serialize = "deny", serialize = "none")]
Deny,
}
#[derive(Debug, EnumString, Display)]
enum NtfyRole {
#[strum(serialize = "user", to_string = "user")]
#[strum(serialize = "user")]
User,
#[strum(serialize = "admin", to_string = "admin")]
#[strum(serialize = "admin")]
Admin,
}
@@ -94,28 +85,6 @@ impl NtfyInterpret {
Ok(())
}
async fn set_access(
&self,
k8s_client: Arc<K8sClient>,
username: &str,
topic: &str,
mode: NtfyAccessMode,
) -> Result<(), String> {
k8s_client
.exec_app(
"ntfy".to_string(),
Some(&self.score.namespace),
vec![
"sh",
"-c",
format!("ntfy access {username} {topic} {mode}").as_str(),
],
)
.await?;
Ok(())
}
}
/// We need a ntfy interpret to wrap the HelmChartScore in order to run the score, and then bootstrap the config inside ntfy
@@ -126,9 +95,8 @@ impl<T: Topology + HelmCommand + K8sclient> Interpret<T> for NtfyInterpret {
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
ntfy_helm_chart_score(self.score.namespace.clone())
.create_interpret()
.execute(inventory, topology)
ntfy_helm_chart_score(self.score.namespace.clone(), self.score.host.clone())
.interpret(inventory, topology)
.await?;
debug!("installed ntfy helm chart");
@@ -140,7 +108,7 @@ impl<T: Topology + HelmCommand + K8sclient> Interpret<T> for NtfyInterpret {
client
.wait_until_deployment_ready(
"ntfy".to_string(),
Some(&self.score.namespace.as_str()),
Some(self.score.namespace.as_str()),
None,
)
.await?;
@@ -151,12 +119,13 @@ impl<T: Topology + HelmCommand + K8sclient> Interpret<T> for NtfyInterpret {
debug!("exec into pod done");
Ok(Outcome::success("installed ntfy".to_string()))
Ok(Outcome::success("Ntfy installed".to_string()))
}
fn get_name(&self) -> InterpretName {
todo!()
InterpretName::Ntfy
}
fn get_version(&self) -> Version {
todo!()
}

View File

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

View File

@@ -0,0 +1,47 @@
use std::str::FromStr;
use std::sync::{Arc, Mutex};
use non_blank_string_rs::NonBlankString;
use crate::modules::{
helm::chart::HelmChartScore, monitoring::prometheus::prometheus_config::PrometheusConfig,
};
pub fn prometheus_helm_chart_score(config: Arc<Mutex<PrometheusConfig>>) -> HelmChartScore {
let config = config.lock().unwrap();
let ns = config.namespace.clone().unwrap();
let values = format!(
r#"
rbac:
create: true
kube-state-metrics:
enabled: false
nodeExporter:
enabled: false
alertmanager:
enabled: false
pushgateway:
enabled: false
server:
serviceAccount:
create: false
rbac:
create: true
fullnameOverride: prometheus-{ns}
"#
);
HelmChartScore {
namespace: Some(NonBlankString::from_str(&config.namespace.clone().unwrap()).unwrap()),
release_name: NonBlankString::from_str("prometheus").unwrap(),
chart_name: NonBlankString::from_str(
"oci://ghcr.io/prometheus-community/charts/prometheus",
)
.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 +1,4 @@
pub mod helm;
#[allow(clippy::module_inception)]
pub mod prometheus;
pub mod prometheus_config;

View File

@@ -0,0 +1,194 @@
use std::sync::{Arc, Mutex};
use async_trait::async_trait;
use log::{debug, error};
use serde::Serialize;
use crate::{
interpret::{InterpretError, Outcome},
inventory::Inventory,
modules::monitoring::{
alert_rule::prometheus_alert_rule::AlertManagerRuleGroup,
grafana::helm::helm_grafana::grafana_helm_chart_score,
kube_prometheus::types::{AlertManagerAdditionalPromRules, AlertManagerChannelConfig},
},
score::Score,
topology::{
HelmCommand, Topology,
installable::Installable,
oberservability::monitoring::{AlertReceiver, AlertRule, AlertSender},
tenant::TenantManager,
},
};
use super::{
helm::prometheus_helm::prometheus_helm_chart_score, prometheus_config::PrometheusConfig,
};
#[derive(Debug)]
pub struct Prometheus {
pub config: Arc<Mutex<PrometheusConfig>>,
}
#[async_trait]
impl AlertSender for Prometheus {
fn name(&self) -> String {
"Prometheus".to_string()
}
}
impl Default for Prometheus {
fn default() -> Self {
Self::new()
}
}
impl Prometheus {
pub fn new() -> Self {
Self {
config: Arc::new(Mutex::new(PrometheusConfig::new())),
}
}
pub async fn configure_with_topology<T: TenantManager>(&self, topology: &T) {
let ns = topology
.get_tenant_config()
.await
.map(|cfg| cfg.name.clone())
.unwrap_or_else(|| "monitoring".to_string());
error!("This must be refactored, see comments in pr #74");
debug!("NS: {}", ns);
self.config.lock().unwrap().namespace = Some(ns);
}
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> {
prometheus_helm_chart_score(self.config.clone())
.interpret(inventory, topology)
.await
}
pub async fn install_grafana<T: Topology + HelmCommand + Send + Sync>(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<Outcome, InterpretError> {
let namespace = {
let config = self.config.lock().unwrap();
config.namespace.clone()
};
if let Some(ns) = namespace.as_deref() {
grafana_helm_chart_score(ns)
.interpret(inventory, topology)
.await
} else {
Err(InterpretError::new(
"could not install grafana, missing namespace".to_string(),
))
}
}
}
#[async_trait]
impl<T: Topology + HelmCommand + TenantManager> Installable<T> for Prometheus {
async fn configure(&self, _inventory: &Inventory, topology: &T) -> Result<(), InterpretError> {
self.configure_with_topology(topology).await;
Ok(())
}
async fn ensure_installed(
&self,
inventory: &Inventory,
topology: &T,
) -> Result<(), InterpretError> {
self.install_prometheus(inventory, topology).await?;
let install_grafana = {
let config = self.config.lock().unwrap();
config.grafana
};
if install_grafana {
self.install_grafana(inventory, topology).await?;
}
Ok(())
}
}
#[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

@@ -0,0 +1,40 @@
use crate::modules::monitoring::kube_prometheus::types::{
AlertManagerAdditionalPromRules, AlertManagerChannelConfig, ServiceMonitor,
};
#[derive(Debug)]
pub struct PrometheusConfig {
pub namespace: Option<String>,
pub default_rules: bool,
pub alert_manager: bool,
pub node_exporter: bool,
pub kube_state_metrics: bool,
pub grafana: bool,
pub prometheus_pushgateway: bool,
pub alert_receiver_configs: Vec<AlertManagerChannelConfig>,
pub alert_rules: Vec<AlertManagerAdditionalPromRules>,
pub additional_service_monitors: Vec<ServiceMonitor>,
}
impl Default for PrometheusConfig {
fn default() -> Self {
Self::new()
}
}
impl PrometheusConfig {
pub fn new() -> Self {
Self {
namespace: None,
default_rules: true,
alert_manager: true,
node_exporter: false,
kube_state_metrics: false,
grafana: true,
prometheus_pushgateway: false,
alert_receiver_configs: vec![],
alert_rules: vec![],
additional_service_monitors: vec![],
}
}
}

View File

@@ -32,7 +32,7 @@ impl OKDBootstrapDhcpScore {
logical_host: topology.bootstrap_host.clone(),
physical_host: inventory
.worker_host
.get(0)
.first()
.expect("Should have at least one worker to be used as bootstrap node")
.clone(),
});

View File

@@ -6,6 +6,12 @@ pub struct OKDUpgradeScore {
_target_version: Version,
}
impl Default for OKDUpgradeScore {
fn default() -> Self {
Self::new()
}
}
impl OKDUpgradeScore {
pub fn new() -> Self {
Self {

View File

@@ -0,0 +1,23 @@
use std::collections::HashMap;
use crate::modules::monitoring::alert_rule::prometheus_alert_rule::PrometheusAlertRule;
pub fn alert_deployment_unavailable() -> PrometheusAlertRule {
PrometheusAlertRule {
alert: "DeploymentUnavailable".into(),
expr: "kube_deployment_status_replicas_unavailable > 0".into(),
r#for: Some("2m".into()),
labels: HashMap::from([("severity".into(), "warning".into())]),
annotations: HashMap::from([
(
"summary".into(),
"Deployment has unavailable replicas".into(),
),
(
"description".into(),
"A deployment in this namespace has unavailable replicas for over 2 minutes."
.into(),
),
]),
}
}

Some files were not shown because too many files have changed in this diff Show More