4 Commits

Author SHA1 Message Date
e874c3483d fix(rpm): explicitly Provides user(name) to satisfy systemd unit Requires
Some checks failed
CI / Build cortex SRPM (push) Has been cancelled
CI / Build neuron SRPM (push) Has been cancelled
CI / Publish cortex to COPR (push) Has been cancelled
CI / Publish neuron to COPR (push) Has been cancelled
CI / Bump version in source (push) Has been cancelled
CI / Format, lint, build, test (push) Has been cancelled
Diagnosing the persistent "Nothing to do" on v0.1.10 surfaced that
removing %attr(,,name) from %files wasn't enough. systemd-rpm-macros
ships its own rpm dep generator (/usr/lib/rpm/systemd.req) that parses
User=/Group= directives from every .service file the package ships
and emits Requires: user(NAME)/group(NAME) accordingly.

Rpmbuild log from v0.1.10 shows these Requires are still emitted even
after the %attr removal. Meanwhile the sysusers provides-generator
emits group(NAME) in both unversioned and versioned forms, but only
a versioned user(NAME) = <base64> when the u-line has GECOS/home/shell
fields. The asymmetry leaves Requires: user(NAME) unresolvable.

Add explicit Provides: user(NAME) back to both specs, with a comment
documenting the actual cause (systemd unit parsing, not file attrs)
so the next person touching these specs doesn't repeat the mistake.

Why monsoon didn't hit this: it creates its user in %pre via
groupadd/useradd (not sysusers.d), so no Provides are generated at
all — matching the Requires: user(monsoon) by luck of the rpm solver
treating unknown symbols as soft-fails for that path. Ours went through
the sysusers Provides code path and hit the asymmetry instead.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-16 15:30:55 +03:00
2caaae018a ci: migrate rpm changelog generation to reusable action
Replace the local .gitea/scripts/generate-rpm-changelog.sh with the
shared composite action at https://git.lair.cafe/actions/rpm-changelog@v1.
Behaviour is identical — collect commits since the previous v* tag,
filter bump-version and merge noise, prepend a dated entry to the
spec — but the logic now lives in one place that other projects can
consume.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-16 15:23:45 +03:00
18d00001cf ci: auto-generate rpm changelog entry per release
On every tag push, build a %changelog entry from the git log since
the previous v* tag and prepend it to each spec. Stops the initial
entry from drifting further and catches bogus-date / stale-version
warnings automatically since the generated date always matches the
day the CI runs.

The generator drops "chore: bump version" commits (bot-authored,
noisy in user-facing changelogs) and merge commits. Author defaults
to the gitea-actions identity but can be overridden via
CHANGELOG_AUTHOR env var if a human release is desired.

Requires fetch-depth: 0 on checkout so git describe can see prior
tags and git log can reach them.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-16 15:04:36 +03:00
ad1442c096 fix(rpm): correct weekday in changelog entry
April 15 2026 was a Wednesday, not Tuesday. rpmbuild validates the
day-of-week against the date and warns on mismatch.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-16 14:58:40 +03:00
14 changed files with 68 additions and 100 deletions

View File

@@ -24,6 +24,19 @@ jobs:
steps: steps:
- uses: actions/checkout@v4 - uses: actions/checkout@v4
- name: Cache cargo registry and target
uses: actions/cache@v4
with:
path: |
~/.cargo/bin
~/.cargo/registry/index
~/.cargo/registry/cache
~/.cargo/git/db
target
key: ${{ runner.os }}-cargo-${{ hashFiles('**/Cargo.lock') }}
restore-keys: |
${{ runner.os }}-cargo-
- name: Ensure sccache with S3 support - name: Ensure sccache with S3 support
env: env:
RUSTC_WRAPPER: "" RUSTC_WRAPPER: ""
@@ -126,37 +139,37 @@ jobs:
run: | run: |
VERSION="${{ steps.version.outputs.VERSION }}" VERSION="${{ steps.version.outputs.VERSION }}"
sed -i '/\[workspace\.package\]/,/\[/{ s/^version = ".*"/version = "'"${VERSION}"'"/ }' Cargo.toml sed -i '/\[workspace\.package\]/,/\[/{ s/^version = ".*"/version = "'"${VERSION}"'"/ }' Cargo.toml
sed -i "s/^Version:.*/Version: ${VERSION}/" helexa-neuron.spec sed -i "s/^Version:.*/Version: ${VERSION}/" neuron.spec
- name: Generate changelog entry - name: Generate changelog entry
uses: https://git.lair.cafe/actions/rpm-changelog@v1 uses: https://git.lair.cafe/actions/rpm-changelog@v1
with: with:
spec: helexa-neuron.spec spec: neuron.spec
version: ${{ steps.version.outputs.VERSION }} version: ${{ steps.version.outputs.VERSION }}
- name: Generate source tarball - name: Generate source tarball
run: | run: |
set -ex set -ex
VERSION="${{ steps.version.outputs.VERSION }}" VERSION="${{ steps.version.outputs.VERSION }}"
tar czf /tmp/helexa-neuron-${VERSION}.tar.gz \ tar czf /tmp/neuron-${VERSION}.tar.gz \
--transform "s,^\.,helexa-neuron-${VERSION}," \ --transform "s,^\.,neuron-${VERSION}," \
--exclude='./target' \ --exclude='./target' \
--exclude='./.git' \ --exclude='./.git' \
--exclude='*.tar.gz' \ --exclude='*.tar.gz' \
--exclude='*.src.rpm' \ --exclude='*.src.rpm' \
. .
mv /tmp/helexa-neuron-${VERSION}.tar.gz . mv /tmp/neuron-${VERSION}.tar.gz .
- name: Vendor Rust dependencies - name: Vendor Rust dependencies
run: | run: |
VERSION="${{ steps.version.outputs.VERSION }}" VERSION="${{ steps.version.outputs.VERSION }}"
cargo vendor vendor/ cargo vendor vendor/
tar czf helexa-neuron-${VERSION}-vendor.tar.gz vendor/ tar czf neuron-${VERSION}-vendor.tar.gz vendor/
rm -rf vendor/ rm -rf vendor/
- name: Build SRPM - name: Build SRPM
run: | run: |
rpmbuild -bs helexa-neuron.spec \ rpmbuild -bs neuron.spec \
--define "_sourcedir $(pwd)" \ --define "_sourcedir $(pwd)" \
--define "_srcrpmdir $(pwd)" --define "_srcrpmdir $(pwd)"
@@ -179,7 +192,7 @@ jobs:
- name: Publish to COPR - name: Publish to COPR
uses: https://git.lair.cafe/actions/copr-publish@v1 uses: https://git.lair.cafe/actions/copr-publish@v1
with: with:
project: helexa/helexa project: helexa/cortex
srpm: "*.src.rpm" srpm: "*.src.rpm"
copr-config: ${{ secrets.COPR_CONFIG }} copr-config: ${{ secrets.COPR_CONFIG }}
@@ -196,7 +209,7 @@ jobs:
- name: Publish to COPR - name: Publish to COPR
uses: https://git.lair.cafe/actions/copr-publish@v1 uses: https://git.lair.cafe/actions/copr-publish@v1
with: with:
project: helexa/helexa project: helexa/neuron
srpm: "*.src.rpm" srpm: "*.src.rpm"
copr-config: ${{ secrets.COPR_CONFIG }} copr-config: ${{ secrets.COPR_CONFIG }}
@@ -206,43 +219,21 @@ jobs:
needs: [copr-cortex, copr-neuron] needs: [copr-cortex, copr-neuron]
steps: steps:
- uses: actions/checkout@v4 - uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Determine version - name: Stamp version and push
id: version
run: echo "VERSION=${GITHUB_REF#refs/tags/v}" >> "$GITHUB_OUTPUT"
- name: Stamp version
run: |
VERSION="${{ steps.version.outputs.VERSION }}"
sed -i '/\[workspace\.package\]/,/\[/{ s/^version = ".*"/version = "'"${VERSION}"'"/ }' Cargo.toml
sed -i "s/^Version:.*/Version: ${VERSION}/" cortex.spec
sed -i "s/^Version:.*/Version: ${VERSION}/" helexa-neuron.spec
cargo check --workspace 2>/dev/null || true
- name: Generate cortex changelog entry
uses: https://git.lair.cafe/actions/rpm-changelog@v1
with:
spec: cortex.spec
version: ${{ steps.version.outputs.VERSION }}
- name: Generate helexa-neuron changelog entry
uses: https://git.lair.cafe/actions/rpm-changelog@v1
with:
spec: helexa-neuron.spec
version: ${{ steps.version.outputs.VERSION }}
- name: Commit and push
env: env:
GITEA_TOKEN: ${{ secrets.GITEA_TOKEN }} GITEA_TOKEN: ${{ secrets.GITEA_TOKEN }}
run: | run: |
VERSION="${{ steps.version.outputs.VERSION }}" VERSION="${GITHUB_REF#refs/tags/v}"
sed -i '/\[workspace\.package\]/,/\[/{ s/^version = ".*"/version = "'"${VERSION}"'"/ }' Cargo.toml
sed -i "s/^Version:.*/Version: ${VERSION}/" cortex.spec
sed -i "s/^Version:.*/Version: ${VERSION}/" neuron.spec
cargo check --workspace 2>/dev/null || true
git config user.name "Gitea Actions" git config user.name "Gitea Actions"
git config user.email "actions@git.lair.cafe" git config user.email "actions@git.lair.cafe"
git add Cargo.toml Cargo.lock cortex.spec helexa-neuron.spec git add Cargo.toml Cargo.lock cortex.spec neuron.spec
if git diff --cached --quiet; then if git diff --cached --quiet; then
echo "Nothing to commit for ${VERSION}" echo "Version already at ${VERSION}"
else else
git commit -m "chore: bump version to ${VERSION}" git commit -m "chore: bump version to ${VERSION}"
git remote set-url origin "https://gitea-actions:${GITEA_TOKEN}@git.lair.cafe/helexa/cortex.git" git remote set-url origin "https://gitea-actions:${GITEA_TOKEN}@git.lair.cafe/helexa/cortex.git"

1
.gitignore vendored
View File

@@ -5,4 +5,3 @@
.vscode/ .vscode/
cortex.toml cortex.toml
doc/plan/* doc/plan/*
script/deploy.sh

View File

@@ -125,8 +125,7 @@ automatically. Clippy warnings must be resolved, not suppressed with
- One or more GPU nodes running mistral.rs on port 8080 - One or more GPU nodes running mistral.rs on port 8080
- Optionally a metrics-only node (no GPU) for Prometheus/Grafana - Optionally a metrics-only node (no GPU) for Prometheus/Grafana
- Each node runs `mistralrs serve` on port 8080 - Each node runs `mistralrs serve` on port 8080
- Gateway listens on port 31313 (API) and 31314 (metrics) - Gateway listens on port 8000 (API) and 9100 (metrics)
- neuron listens on port 13131 on each GPU host
- TLS terminated at gateway or via nginx; internal traffic is plaintext over WireGuard - TLS terminated at gateway or via nginx; internal traffic is plaintext over WireGuard
## Conventions ## Conventions
@@ -381,7 +380,7 @@ processes (one process per loaded model, each on its own port).
## neuron API ## neuron API
neuron exposes an HTTP API on port 13131 that cortex polls and calls. neuron exposes an HTTP API on port 9090 that cortex polls and calls.
``` ```
GET /discovery GET /discovery
@@ -425,8 +424,8 @@ endpoint. cortex.toml shrinks to:
```toml ```toml
[gateway] [gateway]
listen = "0.0.0.0:31313" listen = "0.0.0.0:8000"
metrics_listen = "0.0.0.0:31314" metrics_listen = "0.0.0.0:9100"
[eviction] [eviction]
strategy = "lru" strategy = "lru"
@@ -434,15 +433,15 @@ defrag_after_cycles = 50
[[neurons]] [[neurons]]
name = "beast" name = "beast"
endpoint = "http://beast.hanzalova.internal:13131" endpoint = "http://beast.hanzalova.internal:9090"
[[neurons]] [[neurons]]
name = "benjy" name = "benjy"
endpoint = "http://benjy.hanzalova.internal:13131" endpoint = "http://benjy.kosherinata.internal:9090"
[[neurons]] [[neurons]]
name = "quadbrat" name = "quadbrat"
endpoint = "http://quadbrat.hanzalova.internal:13131" endpoint = "http://quadbrat.hanzalova.internal:9090"
``` ```
On startup and periodically, cortex calls `GET /discovery` and On startup and periodically, cortex calls `GET /discovery` and
@@ -522,7 +521,7 @@ cortex/
│ │ └── metrics.rs # prometheus exporter (unchanged) │ │ └── metrics.rs # prometheus exporter (unchanged)
│ ├── neuron/ # node plane (replaces cortex-agent) │ ├── neuron/ # node plane (replaces cortex-agent)
│ │ └── src/ │ │ └── src/
│ │ ├── main.rs # binary entrypoint, axum server on :13131 │ │ ├── main.rs # binary entrypoint, axum server on :9090
│ │ ├── discovery.rs # nvidia-smi, device enumeration │ │ ├── discovery.rs # nvidia-smi, device enumeration
│ │ ├── health.rs # runtime GPU polling │ │ ├── health.rs # runtime GPU polling
│ │ ├── api.rs # HTTP handlers for /discovery, /models, etc. │ │ ├── api.rs # HTTP handlers for /discovery, /models, etc.
@@ -596,24 +595,16 @@ placement matching can be added incrementally.
Completed. Both packages have RPM specs, systemd units, and example configs. Completed. Both packages have RPM specs, systemd units, and example configs.
CI builds parallel SRPMs on tag push and publishes to separate COPR repos. CI builds parallel SRPMs on tag push and publishes to separate COPR repos.
- `cortex.spec` — installs the `cortex` binary. Package name keeps the - `cortex.spec` `helexa/cortex` COPR: binary, systemd unit, config files
short `cortex` because no Fedora package collides with it. - `neuron.spec``helexa/neuron` COPR: binary, systemd unit, config
- `helexa-neuron.spec` — installs the `neuron` binary under package name
`helexa-neuron`. Renamed from bare `neuron` to avoid collision with
Fedora's NEURON neural-simulation package
(https://src.fedoraproject.org/rpms/neuron); binary, systemd unit,
system user, and config dir all stay named `neuron` since those are
project-local contexts.
- `data/cortex.service`, `data/neuron.service` — systemd units - `data/cortex.service`, `data/neuron.service` — systemd units
- `cortex.example.toml`, `neuron.example.toml`, `models.example.toml` - `cortex.example.toml`, `neuron.example.toml`, `models.example.toml`
- CI: parallel `srpm-cortex` + `srpm-neuron` jobs, then parallel COPR - CI: parallel `srpm-cortex` + `srpm-neuron` jobs, then parallel COPR publish
publish to a single project `helexa/helexa` hosting both packages.
Install: Install:
```sh ```sh
dnf copr enable helexa/helexa dnf copr enable helexa/cortex && dnf install cortex # gateway host
dnf install cortex # gateway host dnf copr enable helexa/neuron && dnf install neuron # GPU nodes
dnf install helexa-neuron # GPU nodes
``` ```
### Phase 11: llama.cpp harness stub ### Phase 11: llama.cpp harness stub

8
Cargo.lock generated
View File

@@ -351,7 +351,7 @@ checksum = "773648b94d0e5d620f64f280777445740e61fe701025087ec8b57f45c791888b"
[[package]] [[package]]
name = "cortex-cli" name = "cortex-cli"
version = "0.1.14" version = "0.1.8"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"clap", "clap",
@@ -366,7 +366,7 @@ dependencies = [
[[package]] [[package]]
name = "cortex-core" name = "cortex-core"
version = "0.1.14" version = "0.1.8"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"async-trait", "async-trait",
@@ -381,7 +381,7 @@ dependencies = [
[[package]] [[package]]
name = "cortex-gateway" name = "cortex-gateway"
version = "0.1.14" version = "0.1.8"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"axum", "axum",
@@ -1184,7 +1184,7 @@ dependencies = [
[[package]] [[package]]
name = "neuron" name = "neuron"
version = "0.1.14" version = "0.1.8"
dependencies = [ dependencies = [
"anyhow", "anyhow",
"async-trait", "async-trait",

View File

@@ -8,7 +8,7 @@ members = [
] ]
[workspace.package] [workspace.package]
version = "0.1.14" version = "0.1.8"
edition = "2024" edition = "2024"
license = "GPL-3.0-or-later" license = "GPL-3.0-or-later"
repository = "https://git.lair.cafe/helexa/cortex" repository = "https://git.lair.cafe/helexa/cortex"

View File

@@ -88,8 +88,8 @@ WantedBy=multi-user.target
```toml ```toml
# cortex.toml # cortex.toml
[gateway] [gateway]
listen = "0.0.0.0:31313" listen = "0.0.0.0:8000"
metrics_listen = "0.0.0.0:31314" metrics_listen = "0.0.0.0:9100"
[eviction] [eviction]
strategy = "lru" # lru | priority strategy = "lru" # lru | priority
@@ -143,7 +143,7 @@ cortex serve --config cortex.toml
cortex status cortex status
# list all models across nodes # list all models across nodes
curl http://localhost:31313/v1/models curl http://localhost:8000/v1/models
``` ```
## License ## License

View File

@@ -3,11 +3,11 @@
# Copy to cortex.toml and adjust for your environment. # Copy to cortex.toml and adjust for your environment.
# #
# Environment variable overrides use CORTEX_ prefix with __ separators: # Environment variable overrides use CORTEX_ prefix with __ separators:
# CORTEX_GATEWAY__LISTEN=0.0.0.0:31313 # CORTEX_GATEWAY__LISTEN=0.0.0.0:9000
[gateway] [gateway]
listen = "0.0.0.0:31313" listen = "0.0.0.0:8000"
metrics_listen = "0.0.0.0:31314" metrics_listen = "0.0.0.0:9100"
[eviction] [eviction]
strategy = "lru" strategy = "lru"

View File

@@ -1,5 +1,5 @@
Name: cortex Name: cortex
Version: 0.1.14 Version: 0.1.8
Release: 1%{?dist} Release: 1%{?dist}
Summary: Inference gateway for multi-node GPU clusters Summary: Inference gateway for multi-node GPU clusters
@@ -83,10 +83,5 @@ install -Dm644 models.example.toml %{buildroot}%{_sysconfdir}/cortex/models.toml
%config(noreplace) %{_sysconfdir}/cortex/models.toml %config(noreplace) %{_sysconfdir}/cortex/models.toml
%changelog %changelog
* Thu Apr 16 2026 Gitea Actions <actions@git.lair.cafe> - 0.1.14-1
- ci: publish both packages to a single helexa/helexa COPR project
- fix(rpm): rename neuron package to helexa-neuron
- ci: commit generated %changelog entries back to main
* Wed Apr 15 2026 Rob Thijssen <grenade@rob.tn> - 0.1.0-1 * Wed Apr 15 2026 Rob Thijssen <grenade@rob.tn> - 0.1.0-1
- Initial package - Initial package

View File

@@ -23,7 +23,7 @@ enum Commands {
/// Print the fleet status (models, nodes, health). /// Print the fleet status (models, nodes, health).
Status { Status {
/// Gateway API endpoint to query. /// Gateway API endpoint to query.
#[arg(short, long, default_value = "http://localhost:31313")] #[arg(short, long, default_value = "http://localhost:8000")]
endpoint: String, endpoint: String,
}, },
} }

View File

@@ -22,9 +22,9 @@ fn default_models_path() -> String {
#[derive(Debug, Clone, Serialize, Deserialize)] #[derive(Debug, Clone, Serialize, Deserialize)]
pub struct GatewaySettings { pub struct GatewaySettings {
/// Address to listen on for API requests (e.g. "0.0.0.0:31313") /// Address to listen on for API requests (e.g. "0.0.0.0:8000")
pub listen: String, pub listen: String,
/// Address to listen on for Prometheus metrics (e.g. "0.0.0.0:31314") /// Address to listen on for Prometheus metrics (e.g. "0.0.0.0:9100")
pub metrics_listen: String, pub metrics_listen: String,
} }
@@ -50,7 +50,7 @@ pub enum EvictionStrategy {
pub struct NeuronEndpoint { pub struct NeuronEndpoint {
/// Human-readable node name (e.g. "beast") /// Human-readable node name (e.g. "beast")
pub name: String, pub name: String,
/// Base URL of the neuron daemon (e.g. "http://beast.internal:13131") /// Base URL of the neuron daemon (e.g. "http://beast.internal:9090")
pub endpoint: String, pub endpoint: String,
} }
@@ -70,8 +70,8 @@ impl Default for GatewayConfig {
fn default() -> Self { fn default() -> Self {
Self { Self {
gateway: GatewaySettings { gateway: GatewaySettings {
listen: "0.0.0.0:31313".into(), listen: "0.0.0.0:8000".into(),
metrics_listen: "0.0.0.0:31314".into(), metrics_listen: "0.0.0.0:9100".into(),
}, },
eviction: EvictionSettings { eviction: EvictionSettings {
strategy: EvictionStrategy::Lru, strategy: EvictionStrategy::Lru,

View File

@@ -6,7 +6,7 @@ use std::collections::HashMap;
#[derive(Debug, Clone)] #[derive(Debug, Clone)]
pub struct NodeState { pub struct NodeState {
pub name: String, pub name: String,
/// Base URL of the neuron daemon (e.g. "http://beast.internal:13131"). /// Base URL of the neuron daemon (e.g. "http://beast.internal:9090").
pub endpoint: String, pub endpoint: String,
pub healthy: bool, pub healthy: bool,
pub models: HashMap<String, ModelEntry>, pub models: HashMap<String, ModelEntry>,

View File

@@ -17,7 +17,7 @@ pub struct NeuronConfig {
} }
fn default_port() -> u16 { fn default_port() -> u16 {
13131 9090
} }
impl NeuronConfig { impl NeuronConfig {
@@ -33,7 +33,7 @@ impl NeuronConfig {
impl Default for NeuronConfig { impl Default for NeuronConfig {
fn default() -> Self { fn default() -> Self {
Self { Self {
port: 13131, port: 9090,
harnesses: vec![], harnesses: vec![],
} }
} }

View File

@@ -3,9 +3,9 @@
# Copy to /etc/neuron/neuron.toml and adjust for your environment. # Copy to /etc/neuron/neuron.toml and adjust for your environment.
# #
# Environment variable overrides use NEURON_ prefix with __ separators: # Environment variable overrides use NEURON_ prefix with __ separators:
# NEURON_PORT=13131 # NEURON_PORT=9090
port = 13131 port = 9090
# -- Harnesses --------------------------------------------------------------- # -- Harnesses ---------------------------------------------------------------
# Each [[harnesses]] entry declares an inference engine managed by neuron. # Each [[harnesses]] entry declares an inference engine managed by neuron.

View File

@@ -1,10 +1,7 @@
Name: helexa-neuron Name: neuron
Version: 0.1.14 Version: 0.1.8
Release: 1%{?dist} Release: 1%{?dist}
Summary: Per-node GPU discovery and harness management daemon for cortex Summary: Per-node GPU discovery and harness management daemon for cortex
# Package name disambiguates from Fedora's existing "neuron" package
# (NEURON neural simulation environment from Yale). Binary, systemd
# unit, and system user are still called "neuron" for brevity.
License: GPL-3.0-or-later License: GPL-3.0-or-later
URL: https://git.lair.cafe/helexa/cortex URL: https://git.lair.cafe/helexa/cortex
@@ -83,10 +80,5 @@ install -Dm644 neuron.example.toml %{buildroot}%{_sysconfdir}/neuron/neuron.toml
%config(noreplace) %{_sysconfdir}/neuron/neuron.toml %config(noreplace) %{_sysconfdir}/neuron/neuron.toml
%changelog %changelog
* Thu Apr 16 2026 Gitea Actions <actions@git.lair.cafe> - 0.1.14-1
- ci: publish both packages to a single helexa/helexa COPR project
- fix(rpm): rename neuron package to helexa-neuron
- ci: commit generated %changelog entries back to main
* Wed Apr 15 2026 Rob Thijssen <grenade@rob.tn> - 0.1.0-1 * Wed Apr 15 2026 Rob Thijssen <grenade@rob.tn> - 0.1.0-1
- Initial package - Initial package