steve/llama-swap
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Changes and fixes before the release (docs/small tweaks) (#750)

Browse Source 
- update README.md with new docker instructions
- update docs/configuration.md
- update .github/workflows to have pinned action versions
- gofmt events package
- fix small bugs in CI scripts
- reduce config options for internal/perf/monitor and config. A ring buffer is used to keep 1hr of entries at max 5s granularity. For long term stats use prometheus monitoring on /metrics

Fixes #744
This commit is contained in:
Benson Wong
2026-05-13 21:18:19 -07:00
committed by GitHub
parent 3e3646f9f9
commit a4b91e08cf
23 changed files with 499 additions and 569 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.github/workflows/closeinactive.yml
+1 -1
View File
@@ -11,7 +11,7 @@ jobs:
issues: write issues: write
pull-requests: write pull-requests: write
steps: steps:
- uses: actions/stale@v9 - uses: actions/stale@b5d41d4e1d5dceea10e7104786b73624c18a190f #v10.2.0
with: with:
days-before-issue-stale: 14 days-before-issue-stale: 14
days-before-issue-close: 14 days-before-issue-close: 14
.github/workflows/config-schema.yml
+2 -2
View File
@@ -21,7 +21,7 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # 6.0.2
- name: Validate JSON Schema - name: Validate JSON Schema
run: | run: |
@@ -45,7 +45,7 @@ jobs:
echo "✓ config-schema.json is valid" echo "✓ config-schema.json is valid"
- name: Set up Python - name: Set up Python
uses: actions/setup-python@v5 uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 #v6.2.0
with: with:
python-version: "3.x" python-version: "3.x"
.github/workflows/containers.yml
+4 -4
View File
@@ -38,7 +38,7 @@ jobs:
fail-fast: false fail-fast: false
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v6 uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # 6.0.2
- name: Free up disk space - name: Free up disk space
if: matrix.platform == 'rocm' if: matrix.platform == 'rocm'
@@ -58,13 +58,13 @@ jobs:
# no-op for amd64-only builds, so leaving it on for every matrix # no-op for amd64-only builds, so leaving it on for every matrix
# entry keeps the workflow simple. # entry keeps the workflow simple.
- name: Set up QEMU - name: Set up QEMU
uses: docker/setup-qemu-action@v4 uses: docker/setup-qemu-action@ce360397dd3f832beb865e1373c09c0e9f86d70a #v4.0.0
- name: Set up Docker Buildx - name: Set up Docker Buildx
uses: docker/setup-buildx-action@v4 uses: docker/setup-buildx-action@4d04d5d9486b7bd6fa91e7baf45bbb4f8b9deedd #v4.0.0
- name: Log in to GitHub Container Registry - name: Log in to GitHub Container Registry
uses: docker/login-action@v4 uses: docker/login-action@4907a6ddec9925e35a0a9e82d7399ccc52663121 #v4.1.0
with: with:
registry: ghcr.io registry: ghcr.io
username: ${{ github.actor }} username: ${{ github.actor }}
.github/workflows/go-ci-windows.yml
+6 -6
View File
@@ -31,17 +31,17 @@ jobs:
run-tests: run-tests:
runs-on: windows-latest runs-on: windows-latest
steps: steps:
- uses: actions/checkout@v4 - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # 6.0.2
- name: Set up Go - name: Set up Go
uses: actions/setup-go@v4 uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c #6.4.0
with: with:
go-version: '1.23' go-version-file: go.mod
# cache simple-responder to save the build time # cache simple-responder to save the build time
- name: Restore Simple Responder - name: Restore Simple Responder
id: restore-simple-responder id: restore-simple-responder
uses: actions/cache/restore@v4 uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
with: with:
path: ./build path: ./build
key: ${{ runner.os }}-simple-responder-${{ hashFiles('cmd/simple-responder/simple-responder.go') }} key: ${{ runner.os }}-simple-responder-${{ hashFiles('cmd/simple-responder/simple-responder.go') }}
@@ -56,11 +56,11 @@ jobs:
# nothing new to save ... skip this step # nothing new to save ... skip this step
if: steps.restore-simple-responder.outputs.cache-hit != 'true' if: steps.restore-simple-responder.outputs.cache-hit != 'true'
id: save-simple-responder id: save-simple-responder
uses: actions/cache/save@v4 uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
with: with:
path: ./build path: ./build
key: ${{ runner.os }}-simple-responder-${{ hashFiles('cmd/simple-responder/simple-responder.go') }} key: ${{ runner.os }}-simple-responder-${{ hashFiles('cmd/simple-responder/simple-responder.go') }}
- name: Test all - name: Test all
shell: bash shell: bash
run: make test-all run: make test-all
.github/workflows/go-ci.yml
+6 -6
View File
@@ -30,24 +30,24 @@ jobs:
run-tests: run-tests:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- uses: actions/checkout@v4 - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # 6.0.2
- name: Set up Go - name: Set up Go
uses: actions/setup-go@v4 uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c #6.4.0
with: with:
go-version-file: go.mod go-version-file: go.mod
# Only run in this linux based runner # Only run in this linux based runner
- name: Check Formatting - name: Check Formatting
run: | run: |
if [ "$(gofmt -l . | grep -v 'event/.*_test.go' | wc -l)" -gt 0 ]; then if [ "$(gofmt -l . | wc -l)" -gt 0 ]; then
gofmt -l . | grep -v 'event/.*_test.go' gofmt -l .
exit 1 exit 1
fi fi
# cache simple-responder to save the build time # cache simple-responder to save the build time
- name: Restore Simple Responder - name: Restore Simple Responder
id: restore-simple-responder id: restore-simple-responder
uses: actions/cache/restore@v4 uses: actions/cache/restore@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
with: with:
path: ./build path: ./build
key: ${{ runner.os }}-simple-responder-${{ hashFiles('cmd/simple-responder/simple-responder.go') }} key: ${{ runner.os }}-simple-responder-${{ hashFiles('cmd/simple-responder/simple-responder.go') }}
@@ -60,7 +60,7 @@ jobs:
# nothing new to save ... skip this step # nothing new to save ... skip this step
if: steps.restore-simple-responder.outputs.cache-hit != 'true' if: steps.restore-simple-responder.outputs.cache-hit != 'true'
id: save-simple-responder id: save-simple-responder
uses: actions/cache/save@v4 uses: actions/cache/save@27d5ce7f107fe9357f9df03efb73ab90386fccae #v5.0.5
with: with:
path: ./build path: ./build
key: ${{ runner.os }}-simple-responder-${{ hashFiles('cmd/simple-responder/simple-responder.go') }} key: ${{ runner.os }}-simple-responder-${{ hashFiles('cmd/simple-responder/simple-responder.go') }}
.github/workflows/release.yml
+5 -5
View File
@@ -20,14 +20,14 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Checkout - name: Checkout
uses: actions/checkout@v4 - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # 6.0.2
with: with:
fetch-depth: 0 fetch-depth: 0
ref: ${{ github.event.inputs.tag || github.ref }} ref: ${{ github.event.inputs.tag || github.ref }}
- name: Set up Go - name: Set up Go
uses: actions/setup-go@v5 uses: actions/setup-go@4a3601121dd01d1626a1e23e37211e3254c1c06c #6.4.0
- name: Set up Node.js - name: Set up Node.js
uses: actions/setup-node@v4 uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # 6.4.0
with: with:
node-version: "24" node-version: "24"
- name: Install dependencies and build UI - name: Install dependencies and build UI
@@ -37,7 +37,7 @@ jobs:
npm run build npm run build
- name: Run GoReleaser - name: Run GoReleaser
uses: goreleaser/goreleaser-action@v6 uses: goreleaser/goreleaser-action@1a80836c5c9d9e5755a25cb59ec6f45a3b5f41a8 #7.2.1
with: with:
# either 'goreleaser' (default) or 'goreleaser-pro' # either 'goreleaser' (default) or 'goreleaser-pro'
distribution: goreleaser distribution: goreleaser
@@ -61,7 +61,7 @@ jobs:
fi fi
- name: "Trigger tap repository update" - name: "Trigger tap repository update"
uses: peter-evans/repository-dispatch@v2 uses: peter-evans/repository-dispatch@28959ce8df70de7be546dd1250a005dd32156697 #4.0.1
with: with:
token: ${{ secrets.TAP_REPO_PAT }} token: ${{ secrets.TAP_REPO_PAT }}
repository: mostlygeek/homebrew-llama-swap repository: mostlygeek/homebrew-llama-swap
.github/workflows/ui-tests.yml
+2 -2
View File
@@ -20,10 +20,10 @@ jobs:
run-tests: run-tests:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- uses: actions/checkout@v4 - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # 6.0.2
- name: Set up Node.js - name: Set up Node.js
uses: actions/setup-node@v4 uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # 6.4.0
with: with:
node-version: '24' node-version: '24'
cache: 'npm' cache: 'npm'
.github/workflows/unified-docker.yml
+3 -3
View File
@@ -75,7 +75,7 @@ jobs:
backend: ${{ fromJSON(needs.setup.outputs.matrix) }} backend: ${{ fromJSON(needs.setup.outputs.matrix) }}
steps: steps:
- name: Checkout code - name: Checkout code
uses: actions/checkout@v4 uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # 6.0.2
- name: Free up disk space - name: Free up disk space
run: | run: |
@@ -94,11 +94,11 @@ jobs:
# llama-swap-builder (which has ccache warm) to avoid exhausting disk. # llama-swap-builder (which has ccache warm) to avoid exhausting disk.
- name: Set up Docker Buildx - name: Set up Docker Buildx
if: ${{ !env.ACT }} if: ${{ !env.ACT }}
uses: docker/setup-buildx-action@v3 uses: docker/setup-buildx-action@4d04d5d9486b7bd6fa91e7baf45bbb4f8b9deedd #v4.0.0
- name: Log in to GitHub Container Registry - name: Log in to GitHub Container Registry
if: ${{ !env.ACT }} if: ${{ !env.ACT }}
uses: docker/login-action@v3 uses: docker/login-action@4907a6ddec9925e35a0a9e82d7399ccc52663121 #v4.1.0
with: with:
registry: ghcr.io registry: ghcr.io
username: ${{ github.actor }} username: ${{ github.actor }}
README.md
+21 -12
View File
@@ -57,8 +57,9 @@ Built in Go for performance and simplicity, llama-swap has zero dependencies and
- ✅ Customizable - ✅ Customizable
- Run concurrent models with a custom DSL swap matrix ([#643](https://github.com/mostlygeek/llama-swap/issues/643)) - Run concurrent models with a custom DSL swap matrix ([#643](https://github.com/mostlygeek/llama-swap/issues/643))
- Automatic unloading of models after timeout by setting a `ttl` - Automatic unloading of models after timeout by setting a `ttl`
- Reliable Docker and Podman support using `cmd` and `cmdStop` together - Docker and Podman support using `cmd` and `cmdStop` together
- Preload models on startup with `hooks` ([#235](https://github.com/mostlygeek/llama-swap/pull/235)) - Preload models on startup with `hooks` ([#235](https://github.com/mostlygeek/llama-swap/pull/235))
- Apply filters to requests to control inference with `stripParams`, `setParams` and `setParamsByID`
### Web UI ### Web UI
@@ -94,8 +95,24 @@ llama-swap can be installed in multiple ways
### Docker Install ([download images](https://github.com/mostlygeek/llama-swap/pkgs/container/llama-swap)) ### Docker Install ([download images](https://github.com/mostlygeek/llama-swap/pkgs/container/llama-swap))
Nightly container images with llama-swap and llama-server are built for multiple platforms (cuda, vulkan, intel, etc.) including [non-root variants with improved security](docs/container-security.md). Two types of container images are built nightly for llama-swap:
The stable-diffusion.cpp server is also included for the musa and vulkan platforms.
1. A unified container with llama-server, ik-llama-server, stable-diffusion.cpp, whisper.cpp and llama-swap built from source. This is only available for cuda and vulkan but has more capabilities. This one is recommended for use.
2. A legacy image that is based on llama.cpp's images and llama-swap copied into the container. Use this one if you prefer to stay close to llama.cpp's container images.
#### Unified container (Recommended)
```shell
$ docker pull ghcr.io/mostlygeek/llama-swap:unified-cuda
# run with a custom configuration and models directory
$ docker run -it --rm --runtime nvidia -p 9292:8080 \
-v /path/to/models:/models \
-v /path/to/custom/config.yaml:/etc/llama-swap/config/config.yaml \
ghcr.io/mostlygeek/llama-swap:unified-cuda
```
#### Legacy container
```shell ```shell
$ docker pull ghcr.io/mostlygeek/llama-swap:cuda $ docker pull ghcr.io/mostlygeek/llama-swap:cuda
@@ -105,14 +122,6 @@ $ docker run -it --rm --runtime nvidia -p 9292:8080 \
-v /path/to/models:/models \ -v /path/to/models:/models \
-v /path/to/custom/config.yaml:/app/config.yaml \ -v /path/to/custom/config.yaml:/app/config.yaml \
ghcr.io/mostlygeek/llama-swap:cuda ghcr.io/mostlygeek/llama-swap:cuda
# configuration hot reload supported with a
# directory volume mount
$ docker run -it --rm --runtime nvidia -p 9292:8080 \
-v /path/to/models:/models \
-v /path/to/custom/config.yaml:/app/config.yaml \
-v /path/to/config:/config \
ghcr.io/mostlygeek/llama-swap:cuda -config /config/config.yaml -watch-config
``` ```
<details> <details>
@@ -268,6 +277,6 @@ For Python based inference servers like vllm or tabbyAPI it is recommended to ru
## Star History ## Star History
> [!NOTE] > [!NOTE]
> ⭐️ Star this project to help others discover it! > Thank you to everyone who has given this project a ⭐️!
[![Star History Chart](https://api.star-history.com/svg?repos=mostlygeek/llama-swap&type=Date)](https://www.star-history.com/#mostlygeek/llama-swap&Date) [![Star History Chart](https://api.star-history.com/svg?repos=mostlygeek/llama-swap&type=Date)](https://www.star-history.com/#mostlygeek/llama-swap&Date)
cmd/monitor-test/main.go
+1 -1
View File
@@ -75,7 +75,7 @@ func main() {
} }
if *stream { if *stream {
m, _ := perf.New(config.PerformanceConfig{Enable: true, Every: every}, l) m, _ := perf.New(config.PerformanceConfig{Every: every}, l)
m.Start() m.Start()
defer m.Stop() defer m.Stop()
sysCh, gpuCh, unsub := m.Subscribe() sysCh, gpuCh, unsub := m.Subscribe()
config-schema.json
+5 -17
View File
@@ -145,33 +145,21 @@
"performance": { "performance": {
"type": "object", "type": "object",
"properties": { "properties": {
"enable": { "disabled": {
"type": "boolean", "type": "boolean",
"default": true, "default": false,
"description": "Enable or disable system performance monitoring." "description": "Disable system performance monitoring."
}, },
"every": { "every": {
"type": "string", "type": "string",
"pattern": "^[-+]?(\\d+(\\.\\d+)?(ns|us|ms|s|m|h))+$", "pattern": "^[-+]?(\\d+(\\.\\d+)?(ns|us|ms|s|m|h))+$",
"default": "15s", "default": "15s",
"description": "Delay between polling for new performance statistics. Minimum duration is 1s. Lower values use more RAM as stats are kept in memory." "description": "Delay between polling for new performance statistics. Minimum duration is 1s. Lower values use more RAM as stats are kept in memory."
},
"maxAge": {
"type": "string",
"pattern": "^[-+]?(\\d+(\\.\\d+)?(ns|us|ms|s|m|h))+$",
"default": "1h",
"description": "Maximum age of performance statistics before they are eligible for garbage collection."
},
"gc": {
"type": "string",
"pattern": "^[-+]?(\\d+(\\.\\d+)?(ns|us|ms|s|m|h))+$",
"default": "5m",
"description": "Garbage collection frequency for clearing old performance statistics."
} }
}, },
"additionalProperties": false, "additionalProperties": false,
"default": {}, "default": {},
"description": "Configuration for system monitoring statistics. Timing values are duration strings like 1s, 1h30m, 90m, 2h10s." "description": "Configuration for CPU, RAM and GPU monitoring statistics."
}, },
"startPort": { "startPort": {
"type": "integer", "type": "integer",
@@ -548,4 +536,4 @@
} }
} }
] ]
} }
config.example.yaml
+9 -19
View File
@@ -58,25 +58,15 @@ captureBuffer: 15
# performance: configuration for system monitoring statistics # performance: configuration for system monitoring statistics
# - timing values are duration strings like 1s, 1h30m, 90m, 2h10s, etc. # - timing values are duration strings like 1s, 1h30m, 90m, 2h10s, etc.
performance: performance:
# enabled: boolean # disabled: boolean
# - default: true # - default: false
enable: true disabled: false
# every: delay between polling for new performance statistics # every: delay between polling for new performance statistics
# - default: 15s # - default: 5s
# - minimum duration 1s # - minimum duration 5s
# - note: setting this very low will use up more RAM as stats are kept in memory.
every: 15s every: 15s
# maxAge: maximum age of a performance statistics before it is eligible for garbage collection
# - default: 1h
maxAge: 12h
# gc: garbage collection frequency in seconds
# - how many seconds the garbage collector runs to clear old stats
# - default 5m
gc: 5m
# startPort: sets the starting port number for the automatic ${PORT} macro. # startPort: sets the starting port number for the automatic ${PORT} macro.
# - optional, default: 5800 # - optional, default: 5800
# - the ${PORT} macro can be used in model.cmd and model.proxy settings # - the ${PORT} macro can be used in model.cmd and model.proxy settings
@@ -118,8 +108,7 @@ globalTTL: 0
macros: macros:
# Example of a multi-line macro # Example of a multi-line macro
"latest-llama": > "latest-llama": >
/path/to/llama-server/llama-server-ec9e0301 /path/to/llama-server/llama-server-ec9e0301 --port ${PORT}
--port ${PORT}
"default_ctx": 4096 "default_ctx": 4096
@@ -279,7 +268,8 @@ models:
# the ${temp} macro will remain a float # the ${temp} macro will remain a float
temperature: ${temp} temperature: ${temp}
note: "The ${MODEL_ID} is running on port ${PORT} temp=${temp}, context=${default_ctx}" note: "The ${MODEL_ID} is running on port ${PORT} temp=${temp},
context=${default_ctx}"
a_list: a_list:
- 1 - 1
@@ -291,7 +281,7 @@ models:
b: 2 b: 2
# objects can contain complex types with macro substitution # objects can contain complex types with macro substitution
# becomes: c: [0.7, false, "model: llama"] # becomes: c: [0.7, false, "model: llama"]
c: ["${temp}", false, "model: ${MODEL_ID}"] c: [ "${temp}", false, "model: ${MODEL_ID}" ]
# concurrencyLimit: overrides the allowed number of active parallel requests to a model # concurrencyLimit: overrides the allowed number of active parallel requests to a model
# - optional, default: 0 # - optional, default: 0
docs/configuration.md
+15 -3
View File
@@ -146,6 +146,18 @@ metricsMaxInMemory: 1000
# - set to 0 to disable # - set to 0 to disable
captureBuffer: 15 captureBuffer: 15
# performance: configuration for system monitoring statistics
# - timing values are duration strings like 1s, 1h30m, 90m, 2h10s, etc.
performance:
# disabled: boolean
# - default: false
enable: true
# every: delay between polling for new performance statistics
# - default: 5s
# - minimum duration 5s
every: 5s
# startPort: sets the starting port number for the automatic ${PORT} macro. # startPort: sets the starting port number for the automatic ${PORT} macro.
# - optional, default: 5800 # - optional, default: 5800
# - the ${PORT} macro can be used in model.cmd and model.proxy settings # - the ${PORT} macro can be used in model.cmd and model.proxy settings
@@ -187,8 +199,7 @@ globalTTL: 0
macros: macros:
# Example of a multi-line macro # Example of a multi-line macro
"latest-llama": > "latest-llama": >
/path/to/llama-server/llama-server-ec9e0301 /path/to/llama-server/llama-server-ec9e0301 --port ${PORT}
--port ${PORT}
"default_ctx": 4096 "default_ctx": 4096
@@ -348,7 +359,8 @@ models:
# the ${temp} macro will remain a float # the ${temp} macro will remain a float
temperature: ${temp} temperature: ${temp}
note: "The ${MODEL_ID} is running on port ${PORT} temp=${temp}, context=${default_ctx}" note: "The ${MODEL_ID} is running on port ${PORT} temp=${temp},
context=${default_ctx}"
a_list: a_list:
- 1 - 1
event/default_test.go
+54 -54
View File
@@ -1,54 +1,54 @@
// Copyright (c) Roman Atachiants and contributore. All rights reserved. // Copyright (c) Roman Atachiants and contributore. All rights reserved.
// Licensed under the MIT license. See LICENSE file in the project root for detaile. // Licensed under the MIT license. See LICENSE file in the project root for detaile.
package event package event
import ( import (
"sync" "sync"
"sync/atomic" "sync/atomic"
"testing" "testing"
"github.com/stretchr/testify/assert" "github.com/stretchr/testify/assert"
) )
/* /*
cpu: 13th Gen Intel(R) Core(TM) i7-13700K cpu: 13th Gen Intel(R) Core(TM) i7-13700K
BenchmarkSubcribeConcurrent-24 1826686 606.3 ns/op 1648 B/op 5 allocs/op BenchmarkSubcribeConcurrent-24 1826686 606.3 ns/op 1648 B/op 5 allocs/op
*/ */
func BenchmarkSubscribeConcurrent(b *testing.B) { func BenchmarkSubscribeConcurrent(b *testing.B) {
d := NewDispatcher() d := NewDispatcher()
b.ReportAllocs() b.ReportAllocs()
b.ResetTimer() b.ResetTimer()
b.RunParallel(func(pb *testing.PB) { b.RunParallel(func(pb *testing.PB) {
for pb.Next() { for pb.Next() {
unsub := Subscribe(d, func(ev MyEvent1) {}) unsub := Subscribe(d, func(ev MyEvent1) {})
unsub() unsub()
} }
}) })
} }
func TestDefaultPublish(t *testing.T) { func TestDefaultPublish(t *testing.T) {
var wg sync.WaitGroup var wg sync.WaitGroup
// Subscribe // Subscribe
var count int64 var count int64
defer On(func(ev MyEvent1) { defer On(func(ev MyEvent1) {
atomic.AddInt64(&count, 1) atomic.AddInt64(&count, 1)
wg.Done() wg.Done()
})() })()
defer OnType(TypeEvent1, func(ev MyEvent1) { defer OnType(TypeEvent1, func(ev MyEvent1) {
atomic.AddInt64(&count, 1) atomic.AddInt64(&count, 1)
wg.Done() wg.Done()
})() })()
// Publish // Publish
wg.Add(4) wg.Add(4)
Emit(MyEvent1{}) Emit(MyEvent1{})
Emit(MyEvent1{}) Emit(MyEvent1{})
// Wait and check // Wait and check
wg.Wait() wg.Wait()
assert.Equal(t, int64(4), count) assert.Equal(t, int64(4), count)
} }
event/event_test.go
+324 -324
View File
@@ -1,324 +1,324 @@
// Copyright (c) Roman Atachiants and contributore. All rights reserved. // Copyright (c) Roman Atachiants and contributore. All rights reserved.
// Licensed under the MIT license. See LICENSE file in the project root for detaile. // Licensed under the MIT license. See LICENSE file in the project root for detaile.
package event package event
import ( import (
"fmt" "fmt"
"sync" "sync"
"sync/atomic" "sync/atomic"
"testing" "testing"
"time" "time"
"github.com/stretchr/testify/assert" "github.com/stretchr/testify/assert"
) )
func TestPublish(t *testing.T) { func TestPublish(t *testing.T) {
d := NewDispatcher() d := NewDispatcher()
var wg sync.WaitGroup var wg sync.WaitGroup
// Subscribe, must be received in order // Subscribe, must be received in order
var count int64 var count int64
defer Subscribe(d, func(ev MyEvent1) { defer Subscribe(d, func(ev MyEvent1) {
assert.Equal(t, int(atomic.AddInt64(&count, 1)), ev.Number) assert.Equal(t, int(atomic.AddInt64(&count, 1)), ev.Number)
wg.Done() wg.Done()
})() })()
// Publish // Publish
wg.Add(3) wg.Add(3)
Publish(d, MyEvent1{Number: 1}) Publish(d, MyEvent1{Number: 1})
Publish(d, MyEvent1{Number: 2}) Publish(d, MyEvent1{Number: 2})
Publish(d, MyEvent1{Number: 3}) Publish(d, MyEvent1{Number: 3})
// Wait and check // Wait and check
wg.Wait() wg.Wait()
assert.Equal(t, int64(3), count) assert.Equal(t, int64(3), count)
} }
func TestUnsubscribe(t *testing.T) { func TestUnsubscribe(t *testing.T) {
d := NewDispatcher() d := NewDispatcher()
assert.Equal(t, 0, d.count(TypeEvent1)) assert.Equal(t, 0, d.count(TypeEvent1))
unsubscribe := Subscribe(d, func(ev MyEvent1) { unsubscribe := Subscribe(d, func(ev MyEvent1) {
// Nothing // Nothing
}) })
assert.Equal(t, 1, d.count(TypeEvent1)) assert.Equal(t, 1, d.count(TypeEvent1))
unsubscribe() unsubscribe()
assert.Equal(t, 0, d.count(TypeEvent1)) assert.Equal(t, 0, d.count(TypeEvent1))
} }
func TestConcurrent(t *testing.T) { func TestConcurrent(t *testing.T) {
const max = 1000000 const max = 1000000
var count int64 var count int64
var wg sync.WaitGroup var wg sync.WaitGroup
wg.Add(1) wg.Add(1)
d := NewDispatcher() d := NewDispatcher()
defer Subscribe(d, func(ev MyEvent1) { defer Subscribe(d, func(ev MyEvent1) {
if current := atomic.AddInt64(&count, 1); current == max { if current := atomic.AddInt64(&count, 1); current == max {
wg.Done() wg.Done()
} }
})() })()
// Asynchronously publish // Asynchronously publish
go func() { go func() {
for i := 0; i < max; i++ { for i := 0; i < max; i++ {
Publish(d, MyEvent1{}) Publish(d, MyEvent1{})
} }
}() }()
defer Subscribe(d, func(ev MyEvent1) { defer Subscribe(d, func(ev MyEvent1) {
// Subscriber that does nothing // Subscriber that does nothing
})() })()
wg.Wait() wg.Wait()
assert.Equal(t, max, int(count)) assert.Equal(t, max, int(count))
} }
func TestSubscribeDifferentType(t *testing.T) { func TestSubscribeDifferentType(t *testing.T) {
d := NewDispatcher() d := NewDispatcher()
assert.Panics(t, func() { assert.Panics(t, func() {
SubscribeTo(d, TypeEvent1, func(ev MyEvent1) {}) SubscribeTo(d, TypeEvent1, func(ev MyEvent1) {})
SubscribeTo(d, TypeEvent1, func(ev MyEvent2) {}) SubscribeTo(d, TypeEvent1, func(ev MyEvent2) {})
}) })
} }
func TestPublishDifferentType(t *testing.T) { func TestPublishDifferentType(t *testing.T) {
d := NewDispatcher() d := NewDispatcher()
assert.Panics(t, func() { assert.Panics(t, func() {
SubscribeTo(d, TypeEvent1, func(ev MyEvent2) {}) SubscribeTo(d, TypeEvent1, func(ev MyEvent2) {})
Publish(d, MyEvent1{}) Publish(d, MyEvent1{})
}) })
} }
func TestCloseDispatcher(t *testing.T) { func TestCloseDispatcher(t *testing.T) {
d := NewDispatcher() d := NewDispatcher()
defer SubscribeTo(d, TypeEvent1, func(ev MyEvent2) {})() defer SubscribeTo(d, TypeEvent1, func(ev MyEvent2) {})()
assert.NoError(t, d.Close()) assert.NoError(t, d.Close())
assert.Panics(t, func() { assert.Panics(t, func() {
SubscribeTo(d, TypeEvent1, func(ev MyEvent2) {}) SubscribeTo(d, TypeEvent1, func(ev MyEvent2) {})
}) })
} }
func TestMatrix(t *testing.T) { func TestMatrix(t *testing.T) {
const amount = 1000 const amount = 1000
for _, subs := range []int{1, 10, 100} { for _, subs := range []int{1, 10, 100} {
for _, topics := range []int{1, 10} { for _, topics := range []int{1, 10} {
expected := subs * topics * amount expected := subs * topics * amount
t.Run(fmt.Sprintf("%dx%d", topics, subs), func(t *testing.T) { t.Run(fmt.Sprintf("%dx%d", topics, subs), func(t *testing.T) {
var count atomic.Int64 var count atomic.Int64
var wg sync.WaitGroup var wg sync.WaitGroup
wg.Add(expected) wg.Add(expected)
d := NewDispatcher() d := NewDispatcher()
for i := 0; i < subs; i++ { for i := 0; i < subs; i++ {
for id := 0; id < topics; id++ { for id := 0; id < topics; id++ {
defer SubscribeTo(d, uint32(id), func(ev MyEvent3) { defer SubscribeTo(d, uint32(id), func(ev MyEvent3) {
count.Add(1) count.Add(1)
wg.Done() wg.Done()
})() })()
} }
} }
for n := 0; n < amount; n++ { for n := 0; n < amount; n++ {
for id := 0; id < topics; id++ { for id := 0; id < topics; id++ {
go Publish(d, MyEvent3{ID: id}) go Publish(d, MyEvent3{ID: id})
} }
} }
wg.Wait() wg.Wait()
assert.Equal(t, expected, int(count.Load())) assert.Equal(t, expected, int(count.Load()))
}) })
} }
} }
} }
func TestConcurrentSubscriptionRace(t *testing.T) { func TestConcurrentSubscriptionRace(t *testing.T) {
// This test specifically targets the race condition that occurs when multiple // This test specifically targets the race condition that occurs when multiple
// goroutines try to subscribe to different event types simultaneously. // goroutines try to subscribe to different event types simultaneously.
// Without the CAS loop, subscriptions could be lost due to registry corruption. // Without the CAS loop, subscriptions could be lost due to registry corruption.
const numGoroutines = 100 const numGoroutines = 100
const numEventTypes = 50 const numEventTypes = 50
d := NewDispatcher() d := NewDispatcher()
defer d.Close() defer d.Close()
var wg sync.WaitGroup var wg sync.WaitGroup
var receivedCount int64 var receivedCount int64
var subscribedTypes sync.Map // Thread-safe map var subscribedTypes sync.Map // Thread-safe map
wg.Add(numGoroutines) wg.Add(numGoroutines)
// Start multiple goroutines that subscribe to different event types concurrently // Start multiple goroutines that subscribe to different event types concurrently
for i := 0; i < numGoroutines; i++ { for i := 0; i < numGoroutines; i++ {
go func(goroutineID int) { go func(goroutineID int) {
defer wg.Done() defer wg.Done()
// Each goroutine subscribes to a unique event type // Each goroutine subscribes to a unique event type
eventType := uint32(goroutineID%numEventTypes + 1000) // Offset to avoid collision with other tests eventType := uint32(goroutineID%numEventTypes + 1000) // Offset to avoid collision with other tests
// Subscribe to the event type // Subscribe to the event type
SubscribeTo(d, eventType, func(ev MyEvent3) { SubscribeTo(d, eventType, func(ev MyEvent3) {
atomic.AddInt64(&receivedCount, 1) atomic.AddInt64(&receivedCount, 1)
}) })
// Record that this type was subscribed // Record that this type was subscribed
subscribedTypes.Store(eventType, true) subscribedTypes.Store(eventType, true)
}(i) }(i)
} }
// Wait for all subscriptions to complete // Wait for all subscriptions to complete
wg.Wait() wg.Wait()
// Count the number of unique event types subscribed // Count the number of unique event types subscribed
expectedTypes := 0 expectedTypes := 0
subscribedTypes.Range(func(key, value interface{}) bool { subscribedTypes.Range(func(key, value interface{}) bool {
expectedTypes++ expectedTypes++
return true return true
}) })
// Small delay to ensure all subscriptions are fully processed // Small delay to ensure all subscriptions are fully processed
time.Sleep(10 * time.Millisecond) time.Sleep(10 * time.Millisecond)
// Publish events to each subscribed type // Publish events to each subscribed type
subscribedTypes.Range(func(key, value interface{}) bool { subscribedTypes.Range(func(key, value interface{}) bool {
eventType := key.(uint32) eventType := key.(uint32)
Publish(d, MyEvent3{ID: int(eventType)}) Publish(d, MyEvent3{ID: int(eventType)})
return true return true
}) })
// Wait for all events to be processed // Wait for all events to be processed
time.Sleep(50 * time.Millisecond) time.Sleep(50 * time.Millisecond)
// Verify that we received at least the expected number of events // Verify that we received at least the expected number of events
// (there might be more if multiple goroutines subscribed to the same event type) // (there might be more if multiple goroutines subscribed to the same event type)
received := atomic.LoadInt64(&receivedCount) received := atomic.LoadInt64(&receivedCount)
assert.GreaterOrEqual(t, int(received), expectedTypes, assert.GreaterOrEqual(t, int(received), expectedTypes,
"Should have received at least %d events, got %d", expectedTypes, received) "Should have received at least %d events, got %d", expectedTypes, received)
// Verify that we have the expected number of unique event types // Verify that we have the expected number of unique event types
assert.Equal(t, numEventTypes, expectedTypes, assert.Equal(t, numEventTypes, expectedTypes,
"Should have exactly %d unique event types", numEventTypes) "Should have exactly %d unique event types", numEventTypes)
} }
func TestConcurrentHandlerRegistration(t *testing.T) { func TestConcurrentHandlerRegistration(t *testing.T) {
const numGoroutines = 100 const numGoroutines = 100
// Test concurrent subscriptions to the same event type // Test concurrent subscriptions to the same event type
t.Run("SameEventType", func(t *testing.T) { t.Run("SameEventType", func(t *testing.T) {
d := NewDispatcher() d := NewDispatcher()
var handlerCount int64 var handlerCount int64
var wg sync.WaitGroup var wg sync.WaitGroup
// Start multiple goroutines subscribing to the same event type (0x1) // Start multiple goroutines subscribing to the same event type (0x1)
for i := 0; i < numGoroutines; i++ { for i := 0; i < numGoroutines; i++ {
wg.Add(1) wg.Add(1)
go func() { go func() {
defer wg.Done() defer wg.Done()
SubscribeTo(d, uint32(0x1), func(ev MyEvent1) { SubscribeTo(d, uint32(0x1), func(ev MyEvent1) {
atomic.AddInt64(&handlerCount, 1) atomic.AddInt64(&handlerCount, 1)
}) })
}() }()
} }
wg.Wait() wg.Wait()
// Verify all handlers were registered by publishing an event // Verify all handlers were registered by publishing an event
atomic.StoreInt64(&handlerCount, 0) atomic.StoreInt64(&handlerCount, 0)
Publish(d, MyEvent1{}) Publish(d, MyEvent1{})
// Small delay to ensure all handlers have executed // Small delay to ensure all handlers have executed
time.Sleep(10 * time.Millisecond) time.Sleep(10 * time.Millisecond)
assert.Equal(t, int64(numGoroutines), atomic.LoadInt64(&handlerCount), assert.Equal(t, int64(numGoroutines), atomic.LoadInt64(&handlerCount),
"Not all handlers were registered due to race condition") "Not all handlers were registered due to race condition")
}) })
// Test concurrent subscriptions to different event types // Test concurrent subscriptions to different event types
t.Run("DifferentEventTypes", func(t *testing.T) { t.Run("DifferentEventTypes", func(t *testing.T) {
d := NewDispatcher() d := NewDispatcher()
var wg sync.WaitGroup var wg sync.WaitGroup
receivedEvents := make(map[uint32]*int64) receivedEvents := make(map[uint32]*int64)
// Create multiple event types and subscribe concurrently // Create multiple event types and subscribe concurrently
for i := 0; i < numGoroutines; i++ { for i := 0; i < numGoroutines; i++ {
eventType := uint32(100 + i) eventType := uint32(100 + i)
counter := new(int64) counter := new(int64)
receivedEvents[eventType] = counter receivedEvents[eventType] = counter
wg.Add(1) wg.Add(1)
go func(et uint32, cnt *int64) { go func(et uint32, cnt *int64) {
defer wg.Done() defer wg.Done()
SubscribeTo(d, et, func(ev MyEvent3) { SubscribeTo(d, et, func(ev MyEvent3) {
atomic.AddInt64(cnt, 1) atomic.AddInt64(cnt, 1)
}) })
}(eventType, counter) }(eventType, counter)
} }
wg.Wait() wg.Wait()
// Publish events to all types // Publish events to all types
for eventType := uint32(100); eventType < uint32(100+numGoroutines); eventType++ { for eventType := uint32(100); eventType < uint32(100+numGoroutines); eventType++ {
Publish(d, MyEvent3{ID: int(eventType)}) Publish(d, MyEvent3{ID: int(eventType)})
} }
// Small delay to ensure all handlers have executed // Small delay to ensure all handlers have executed
time.Sleep(10 * time.Millisecond) time.Sleep(10 * time.Millisecond)
// Verify all event types received their events // Verify all event types received their events
for eventType, counter := range receivedEvents { for eventType, counter := range receivedEvents {
assert.Equal(t, int64(1), atomic.LoadInt64(counter), assert.Equal(t, int64(1), atomic.LoadInt64(counter),
"Event type %d did not receive its event", eventType) "Event type %d did not receive its event", eventType)
} }
}) })
} }
func TestBackpressure(t *testing.T) { func TestBackpressure(t *testing.T) {
d := NewDispatcher() d := NewDispatcher()
d.maxQueue = 10 d.maxQueue = 10
var processedCount int64 var processedCount int64
unsub := SubscribeTo(d, uint32(0x200), func(ev MyEvent3) { unsub := SubscribeTo(d, uint32(0x200), func(ev MyEvent3) {
atomic.AddInt64(&processedCount, 1) atomic.AddInt64(&processedCount, 1)
}) })
defer unsub() defer unsub()
const eventsToPublish = 1000 const eventsToPublish = 1000
for i := 0; i < eventsToPublish; i++ { for i := 0; i < eventsToPublish; i++ {
Publish(d, MyEvent3{ID: 0x200}) Publish(d, MyEvent3{ID: 0x200})
} }
time.Sleep(100 * time.Millisecond) time.Sleep(100 * time.Millisecond)
// Verify all events were eventually processed // Verify all events were eventually processed
finalProcessed := atomic.LoadInt64(&processedCount) finalProcessed := atomic.LoadInt64(&processedCount)
assert.Equal(t, int64(eventsToPublish), finalProcessed) assert.Equal(t, int64(eventsToPublish), finalProcessed)
t.Logf("Events processed: %d/%d", finalProcessed, eventsToPublish) t.Logf("Events processed: %d/%d", finalProcessed, eventsToPublish)
} }
// ------------------------------------- Test Events ------------------------------------- // ------------------------------------- Test Events -------------------------------------
const ( const (
TypeEvent1 = 0x1 TypeEvent1 = 0x1
TypeEvent2 = 0x2 TypeEvent2 = 0x2
) )
type MyEvent1 struct { type MyEvent1 struct {
Number int Number int
} }
func (t MyEvent1) Type() uint32 { return TypeEvent1 } func (t MyEvent1) Type() uint32 { return TypeEvent1 }
type MyEvent2 struct { type MyEvent2 struct {
Text string Text string
} }
func (t MyEvent2) Type() uint32 { return TypeEvent2 } func (t MyEvent2) Type() uint32 { return TypeEvent2 }
type MyEvent3 struct { type MyEvent3 struct {
ID int ID int
} }
func (t MyEvent3) Type() uint32 { return uint32(t.ID) } func (t MyEvent3) Type() uint32 { return uint32(t.ID) }
internal/perf/monitor.go
+4 -8
View File
@@ -31,7 +31,7 @@ type Monitor struct {
} }
func ringCapacity(c config.PerformanceConfig) int { func ringCapacity(c config.PerformanceConfig) int {
n := int(c.MaxAge / c.Every) n := int(time.Hour / c.Every)
if n < 1 { if n < 1 {
n = 1 n = 1
} }
@@ -43,12 +43,6 @@ func New(c config.PerformanceConfig, logger *logmon.Monitor) (*Monitor, error) {
if c.Every < 100*time.Millisecond { if c.Every < 100*time.Millisecond {
c.Every = 100 * time.Millisecond c.Every = 100 * time.Millisecond
} }
if c.GC < 1*time.Second {
c.GC = 1 * time.Second
}
if c.MaxAge < 1*time.Minute {
c.MaxAge = 1 * time.Minute
}
if logger == nil { if logger == nil {
return nil, errors.New("logger is required") return nil, errors.New("logger is required")
@@ -92,7 +86,9 @@ func (m *Monitor) UpdateConfig(newConf config.PerformanceConfig) {
m.sysRing = ring.NewBuffer[SysStat](capacity) m.sysRing = ring.NewBuffer[SysStat](capacity)
m.gpuRing = ring.NewBuffer[[]GpuStat](capacity) m.gpuRing = ring.NewBuffer[[]GpuStat](capacity)
m.mutex.Unlock() m.mutex.Unlock()
m.Start() if !newConf.Disabled {
m.Start()
}
} }
// Subscribe returns channels to listen to system and GPU stats. // Subscribe returns channels to listen to system and GPU stats.
internal/perf/monitor_test.go
+2 -14
View File
@@ -24,26 +24,19 @@ func TestNew_DefaultConfig(t *testing.T) {
require.NotNil(t, m) require.NotNil(t, m)
assert.Equal(t, 100*time.Millisecond, m.conf.Every) assert.Equal(t, 100*time.Millisecond, m.conf.Every)
assert.Equal(t, 1*time.Second, m.conf.GC)
assert.Equal(t, 1*time.Minute, m.conf.MaxAge)
} }
func TestNew_CustomConfig(t *testing.T) { func TestNew_CustomConfig(t *testing.T) {
logger := newTestLogger() logger := newTestLogger()
cfg := config.PerformanceConfig{ cfg := config.PerformanceConfig{
Enable: true, Every: 500 * time.Millisecond,
Every: 500 * time.Millisecond,
GC: 5 * time.Second,
MaxAge: 10 * time.Minute,
} }
m, err := New(cfg, logger) m, err := New(cfg, logger)
require.NoError(t, err) require.NoError(t, err)
assert.Equal(t, 500*time.Millisecond, m.conf.Every) assert.Equal(t, 500*time.Millisecond, m.conf.Every)
assert.Equal(t, 5*time.Second, m.conf.GC)
assert.Equal(t, 10*time.Minute, m.conf.MaxAge)
} }
func TestNew_NilLogger(t *testing.T) { func TestNew_NilLogger(t *testing.T) {
@@ -56,18 +49,13 @@ func TestNew_BelowMinimumConfig(t *testing.T) {
logger := newTestLogger() logger := newTestLogger()
cfg := config.PerformanceConfig{ cfg := config.PerformanceConfig{
Enable: true, Every: 1 * time.Millisecond,
Every: 1 * time.Millisecond,
GC: 100 * time.Millisecond,
MaxAge: 1 * time.Second,
} }
m, err := New(cfg, logger) m, err := New(cfg, logger)
require.NoError(t, err) require.NoError(t, err)
assert.Equal(t, 100*time.Millisecond, m.conf.Every) assert.Equal(t, 100*time.Millisecond, m.conf.Every)
assert.Equal(t, 1*time.Second, m.conf.GC)
assert.Equal(t, 1*time.Minute, m.conf.MaxAge)
} }
func TestSubscribe_ReturnsChannels(t *testing.T) { func TestSubscribe_ReturnsChannels(t *testing.T) {
llama-swap.go
+17 -8
View File
@@ -93,12 +93,17 @@ func main() {
listenStr = &defaultPort listenStr = &defaultPort
} }
mon, err := perf.New(conf.Performance, mainLogger) var mon *perf.Monitor
if err != nil { if !conf.Performance.Disabled {
mainLogger.Errorf("failed to create monitor: %s", err.Error()) mon, err = perf.New(conf.Performance, mainLogger)
os.Exit(1) if err != nil {
mainLogger.Errorf("failed to create monitor: %s", err.Error())
os.Exit(1)
}
mon.Start()
} else {
mainLogger.Info("performance monitoring is disabled")
} }
mon.Start()
// Setup channels for server management // Setup channels for server management
exitChan := make(chan struct{}) exitChan := make(chan struct{})
@@ -108,7 +113,7 @@ func main() {
// Context that bounds the lifetime of background watcher goroutines. // Context that bounds the lifetime of background watcher goroutines.
watcherCtx, watcherCancel := context.WithCancel(context.Background()) watcherCtx, watcherCancel := context.WithCancel(context.Background())
// Create server with initial handler // Create server with initial handlergit
srv := &http.Server{ srv := &http.Server{
Addr: *listenStr, Addr: *listenStr,
} }
@@ -140,7 +145,9 @@ func main() {
mainLogger.Debug("Configuration Changed") mainLogger.Debug("Configuration Changed")
currentPM.Shutdown() currentPM.Shutdown()
mon.UpdateConfig(conf.Performance) if mon != nil {
mon.UpdateConfig(conf.Performance)
}
newPM := proxy.New(conf) newPM := proxy.New(conf)
newPM.SetVersion(date, commit, version) newPM.SetVersion(date, commit, version)
newPM.SetPerfMonitor(mon) newPM.SetPerfMonitor(mon)
@@ -197,7 +204,9 @@ func main() {
reloadProxyManager() reloadProxyManager()
case syscall.SIGINT, syscall.SIGTERM: case syscall.SIGINT, syscall.SIGTERM:
mainLogger.Debugf("Received signal %v, shutting down...", sig) mainLogger.Debugf("Received signal %v, shutting down...", sig)
mon.Stop() if mon != nil {
mon.Stop()
}
watcherCancel() watcherCancel()
ctx, cancel := context.WithTimeout(context.Background(), time.Second*5) ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
defer cancel() defer cancel()
proxy/config/config.go
+2 -5
View File
@@ -223,11 +223,8 @@ func LoadConfigFromReader(r io.Reader) (Config, error) {
} }
// Apply defaults for performance config when section is missing // Apply defaults for performance config when section is missing
if !config.Performance.Enable && config.Performance.Every == 0 && config.Performance.MaxAge == 0 && config.Performance.GC == 0 { if config.Performance.Every == 0 {
config.Performance.Enable = true config.Performance.Every = 5 * time.Second
config.Performance.Every = 15 * time.Second
config.Performance.MaxAge = 1 * time.Hour
config.Performance.GC = 5 * time.Minute
} }
if err = config.Performance.Validate(); err != nil { if err = config.Performance.Validate(); err != nil {
return Config{}, fmt.Errorf("performance: %w", err) return Config{}, fmt.Errorf("performance: %w", err)
proxy/config/config_posix_test.go
+1 -4
View File
@@ -231,10 +231,7 @@ groups:
MetricsMaxInMemory: 1000, MetricsMaxInMemory: 1000,
CaptureBuffer: 5, CaptureBuffer: 5,
Performance: PerformanceConfig{ Performance: PerformanceConfig{
Enable: true, Every: 5 * time.Second,
Every: 15 * time.Second,
MaxAge: 1 * time.Hour,
GC: 5 * time.Minute,
}, },
Profiles: map[string][]string{ Profiles: map[string][]string{
"test": {"model1", "model2"}, "test": {"model1", "model2"},
proxy/config/config_windows_test.go
+1 -4
View File
@@ -220,10 +220,7 @@ groups:
MetricsMaxInMemory: 1000, MetricsMaxInMemory: 1000,
CaptureBuffer: 5, CaptureBuffer: 5,
Performance: PerformanceConfig{ Performance: PerformanceConfig{
Enable: true, Every: 5 * time.Second,
Every: 15 * time.Second,
MaxAge: 1 * time.Hour,
GC: 5 * time.Minute,
}, },
Profiles: map[string][]string{ Profiles: map[string][]string{
"test": {"model1", "model2"}, "test": {"model1", "model2"},
proxy/config/performance.go
+5 -16
View File
@@ -7,19 +7,14 @@ import (
// PerformanceConfig holds configuration for system performance monitoring // PerformanceConfig holds configuration for system performance monitoring
type PerformanceConfig struct { type PerformanceConfig struct {
Enable bool `yaml:"enable"` Disabled bool `yaml:"disabled"`
Every time.Duration `yaml:"every"` Every time.Duration `yaml:"every"`
MaxAge time.Duration `yaml:"maxAge"`
GC time.Duration `yaml:"gc"`
} }
func (p *PerformanceConfig) UnmarshalYAML(unmarshal func(interface{}) error) error { func (p *PerformanceConfig) UnmarshalYAML(unmarshal func(interface{}) error) error {
type rawPerformanceConfig PerformanceConfig type rawPerformanceConfig PerformanceConfig
defaults := rawPerformanceConfig{ defaults := rawPerformanceConfig{
Enable: true, Every: 5 * time.Second,
Every: 15 * time.Second,
MaxAge: 1 * time.Hour,
GC: 5 * time.Minute,
} }
if err := unmarshal(&defaults); err != nil { if err := unmarshal(&defaults); err != nil {
@@ -32,14 +27,8 @@ func (p *PerformanceConfig) UnmarshalYAML(unmarshal func(interface{}) error) err
// Validate checks the PerformanceConfig values and returns an error if invalid // Validate checks the PerformanceConfig values and returns an error if invalid
func (p *PerformanceConfig) Validate() error { func (p *PerformanceConfig) Validate() error {
if p.Every < time.Second { if p.Every < 5*time.Second {
return fmt.Errorf("every must be at least 1s, got %v", p.Every) return fmt.Errorf("every must be at least 5s, got %v", p.Every)
}
if p.MaxAge <= 0 {
return fmt.Errorf("maxAge must be greater than 0, got %v", p.MaxAge)
}
if p.GC <= 0 {
return fmt.Errorf("gc must be greater than 0, got %v", p.GC)
} }
return nil return nil
} }
proxy/config/performance_test.go
+9 -51
View File
@@ -18,10 +18,8 @@ models:
assert.NoError(t, err) assert.NoError(t, err)
// When performance section is missing, defaults should be applied // When performance section is missing, defaults should be applied
assert.True(t, config.Performance.Enable) assert.False(t, config.Performance.Disabled)
assert.Equal(t, 15*time.Second, config.Performance.Every) assert.Equal(t, 5*time.Second, config.Performance.Every)
assert.Equal(t, 1*time.Hour, config.Performance.MaxAge)
assert.Equal(t, 5*time.Minute, config.Performance.GC)
} }
func TestPerformanceConfig_CustomValues(t *testing.T) { func TestPerformanceConfig_CustomValues(t *testing.T) {
@@ -29,8 +27,6 @@ func TestPerformanceConfig_CustomValues(t *testing.T) {
performance: performance:
enable: true enable: true
every: 30s every: 30s
maxAge: 12h
gc: 10m
models: models:
model1: model1:
cmd: path/to/cmd --port ${PORT} cmd: path/to/cmd --port ${PORT}
@@ -38,16 +34,14 @@ models:
config, err := LoadConfigFromReader(strings.NewReader(content)) config, err := LoadConfigFromReader(strings.NewReader(content))
assert.NoError(t, err) assert.NoError(t, err)
assert.True(t, config.Performance.Enable) assert.False(t, config.Performance.Disabled)
assert.Equal(t, 30*time.Second, config.Performance.Every) assert.Equal(t, 30*time.Second, config.Performance.Every)
assert.Equal(t, 12*time.Hour, config.Performance.MaxAge)
assert.Equal(t, 10*time.Minute, config.Performance.GC)
} }
func TestPerformanceConfig_Disabled(t *testing.T) { func TestPerformanceConfig_Disabled(t *testing.T) {
content := ` content := `
performance: performance:
enable: false disabled: true
models: models:
model1: model1:
cmd: path/to/cmd --port ${PORT} cmd: path/to/cmd --port ${PORT}
@@ -55,18 +49,15 @@ models:
config, err := LoadConfigFromReader(strings.NewReader(content)) config, err := LoadConfigFromReader(strings.NewReader(content))
assert.NoError(t, err) assert.NoError(t, err)
assert.False(t, config.Performance.Enable) assert.True(t, config.Performance.Disabled)
// Duration defaults should still apply // Duration defaults should still apply
assert.Equal(t, 15*time.Second, config.Performance.Every) assert.Equal(t, 5*time.Second, config.Performance.Every)
assert.Equal(t, 1*time.Hour, config.Performance.MaxAge)
assert.Equal(t, 5*time.Minute, config.Performance.GC)
} }
func TestPerformanceConfig_PartialValues(t *testing.T) { func TestPerformanceConfig_PartialValues(t *testing.T) {
content := ` content := `
performance: performance:
every: 10s every: 10s
maxAge: 6h
models: models:
model1: model1:
cmd: path/to/cmd --port ${PORT} cmd: path/to/cmd --port ${PORT}
@@ -75,58 +66,27 @@ models:
assert.NoError(t, err) assert.NoError(t, err)
// enable should default to true // enable should default to true
assert.True(t, config.Performance.Enable) assert.False(t, config.Performance.Disabled)
assert.Equal(t, 10*time.Second, config.Performance.Every) assert.Equal(t, 10*time.Second, config.Performance.Every)
assert.Equal(t, 6*time.Hour, config.Performance.MaxAge)
// gc should use default
assert.Equal(t, 5*time.Minute, config.Performance.GC)
} }
func TestPerformanceConfig_InvalidEvery(t *testing.T) { func TestPerformanceConfig_InvalidEvery(t *testing.T) {
content := ` content := `
performance: performance:
every: 500ms every: 4s
models: models:
model1: model1:
cmd: path/to/cmd --port ${PORT} cmd: path/to/cmd --port ${PORT}
` `
_, err := LoadConfigFromReader(strings.NewReader(content)) _, err := LoadConfigFromReader(strings.NewReader(content))
assert.Error(t, err) assert.Error(t, err)
assert.Contains(t, err.Error(), "every must be at least 1s") assert.Contains(t, err.Error(), "every must be at least 5s")
}
func TestPerformanceConfig_InvalidMaxAge(t *testing.T) {
content := `
performance:
maxAge: 0s
models:
model1:
cmd: path/to/cmd --port ${PORT}
`
_, err := LoadConfigFromReader(strings.NewReader(content))
assert.Error(t, err)
assert.Contains(t, err.Error(), "maxAge must be greater than 0")
}
func TestPerformanceConfig_InvalidGC(t *testing.T) {
content := `
performance:
gc: 0s
models:
model1:
cmd: path/to/cmd --port ${PORT}
`
_, err := LoadConfigFromReader(strings.NewReader(content))
assert.Error(t, err)
assert.Contains(t, err.Error(), "gc must be greater than 0")
} }
func TestPerformanceConfig_ComplexDurations(t *testing.T) { func TestPerformanceConfig_ComplexDurations(t *testing.T) {
content := ` content := `
performance: performance:
every: 1m30s every: 1m30s
maxAge: 2h10m
gc: 1m
models: models:
model1: model1:
cmd: path/to/cmd --port ${PORT} cmd: path/to/cmd --port ${PORT}
@@ -135,6 +95,4 @@ models:
assert.NoError(t, err) assert.NoError(t, err)
assert.Equal(t, 90*time.Second, config.Performance.Every) assert.Equal(t, 90*time.Second, config.Performance.Every)
assert.Equal(t, (2*time.Hour)+(10*time.Minute), config.Performance.MaxAge)
assert.Equal(t, 1*time.Minute, config.Performance.GC)
} }