Files
261ab374ce enhance(cache): scope cache entries by ref like GitHub (#1361)
Cache entries were only scoped by repository, so an entry a pull request saved could be restored on the default branch, for example through a `restore-keys` prefix.

Entries now carry the ref of the job that saved them. As on GitHub ([Restrictions for accessing a cache](https://docs.github.com/en/actions/reference/workflows-and-actions/dependency-caching#restrictions-for-accessing-a-cache)), a job restores entries of its own ref first, then of its pull request's base branch, then of the default branch. Pull request events are scoped by their pull request number.

Entries saved before this change are no longer restored and expire through `cache.retention`, so caches start cold once after the upgrade. Runners sharing a cache, and an external `cache-server`, should be upgraded together.

Co-authored-by: silverwind <[email protected]>
Co-authored-by: silverwind <[email protected]>
Reviewed-on: https://gitea.com/gitea/runner/pulls/1361
Reviewed-by: silverwind <[email protected]>
Co-authored-by: JorgeDLMV <[email protected]>
2026-10-30 22:34:45 +00:00

345 lines
14 KiB
Go

// Copyright 2026 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
package artifactcache
import (
"bytes"
"encoding/base64"
"encoding/json/v2"
"fmt"
"io"
"net/http"
"net/url"
"strconv"
"strings"
"testing"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func v2Call(t *testing.T, handler *Handler, client *http.Client, method string, request any) map[string]any {
t.Helper()
body, err := json.Marshal(request)
require.NoError(t, err)
resp, err := client.Post(handler.ExternalURL()+cacheServiceV2Path+"/"+method, "application/json", bytes.NewReader(body))
require.NoError(t, err)
defer resp.Body.Close()
require.Equal(t, http.StatusOK, resp.StatusCode)
got := map[string]any{}
require.NoError(t, json.UnmarshalRead(resp.Body, &got))
return got
}
func putBlob(t *testing.T, url string, content []byte) int {
t.Helper()
req, err := http.NewRequestWithContext(t.Context(), http.MethodPut, url, bytes.NewReader(content))
require.NoError(t, err)
resp, err := http.DefaultClient.Do(req)
require.NoError(t, err)
defer resp.Body.Close()
if resp.StatusCode == http.StatusCreated {
require.NotEmpty(t, resp.Header.Get("x-ms-request-id"))
}
return resp.StatusCode
}
func getURL(t *testing.T, url string) []byte {
t.Helper()
req, err := http.NewRequestWithContext(t.Context(), http.MethodGet, url, nil)
require.NoError(t, err)
resp, err := http.DefaultClient.Do(req)
require.NoError(t, err)
defer resp.Body.Close()
body, err := io.ReadAll(resp.Body)
require.NoError(t, err)
return body
}
func saveV2(t *testing.T, handler *Handler, key, version string, content []byte) (finalized map[string]any, uploadURL string) {
t.Helper()
created := v2Call(t, handler, testClient, "CreateCacheEntry", map[string]any{"key": key, "version": version})
require.Equal(t, true, created["ok"])
uploadURL, _ = created["signed_upload_url"].(string)
require.NotEmpty(t, uploadURL)
require.Equal(t, http.StatusCreated, putBlob(t, uploadURL, content))
return v2Call(t, handler, testClient, "FinalizeCacheEntryUpload", map[string]any{
"key": key, "version": version,
"size_bytes": strconv.Itoa(len(content)),
}), uploadURL
}
func TestCacheServiceV2RoundTrip(t *testing.T) {
t.Parallel()
handler := newTestHandler(t, Policy{})
content := []byte("the cached archive")
unsigned := fmt.Sprintf("%s%s/1", handler.ExternalURL(), blobPath)
assert.Equal(t, http.StatusUnauthorized, putBlob(t, unsigned, content))
finalized, uploadURL := saveV2(t, handler, "deps-v1", "abc123", content)
require.Equal(t, true, finalized["ok"])
assert.NotEmpty(t, finalized["entry_id"])
assert.Equal(t, http.StatusBadRequest, putBlob(t, uploadURL, []byte("poisoned")))
resp, err := http.Get(uploadURL)
require.NoError(t, err)
defer resp.Body.Close()
assert.Equal(t, http.StatusMethodNotAllowed, resp.StatusCode)
got := v2Call(t, handler, testClient, "GetCacheEntryDownloadURL", map[string]any{"key": "deps-v1", "version": "abc123"})
require.Equal(t, true, got["ok"])
assert.Equal(t, "deps-v1", got["matched_key"])
downloadURL, _ := got["signed_download_url"].(string)
require.NotEmpty(t, downloadURL)
assert.Equal(t, content, getURL(t, downloadURL))
}
func TestCacheServiceV2BlockUpload(t *testing.T) {
t.Parallel()
handler := newTestHandler(t, Policy{})
created := v2Call(t, handler, testClient, "CreateCacheEntry", map[string]any{"key": "blocks", "version": "v1"})
uploadURL, _ := created["signed_upload_url"].(string)
require.NotEmpty(t, uploadURL)
blocks := map[string][]byte{}
var order []string
for i, part := range []string{"hello ", "world", "!"} {
blockID := base64.StdEncoding.EncodeToString(fmt.Appendf(nil, "block-%d", i))
blocks[blockID] = []byte(part)
order = append(order, blockID)
}
for _, blockID := range []string{order[2], order[0], order[1]} {
require.Equal(t, http.StatusCreated, putBlob(t, uploadURL+"&comp=block&blockid="+blockID, blocks[blockID]))
}
var list bytes.Buffer
list.WriteString(`<?xml version="1.0" encoding="utf-8"?><BlockList>`)
for _, blockID := range order {
fmt.Fprintf(&list, "<Latest>%s</Latest>", blockID)
}
list.WriteString(`</BlockList>`)
require.Equal(t, http.StatusCreated, putBlob(t, uploadURL+"&comp=blocklist", list.Bytes()))
require.Equal(t, http.StatusCreated, putBlob(t, uploadURL+"&comp=blocklist", list.Bytes()),
"a client that lost the first answer retries the list it already sent")
var reordered bytes.Buffer
reordered.WriteString(`<?xml version="1.0" encoding="utf-8"?><BlockList>`)
for _, blockID := range []string{order[1], order[0], order[2]} {
fmt.Fprintf(&reordered, "<Latest>%s</Latest>", blockID)
}
reordered.WriteString(`</BlockList>`)
assert.Equal(t, http.StatusInternalServerError, putBlob(t, uploadURL+"&comp=blocklist", reordered.Bytes()),
"the blocks are already assembled, so a different order would silently disagree with them")
finalized := v2Call(t, handler, testClient, "FinalizeCacheEntryUpload", map[string]any{
"key": "blocks", "version": "v1", "size_bytes": len("hello world!"),
})
require.Equal(t, true, finalized["ok"])
got := v2Call(t, handler, testClient, "GetCacheEntryDownloadURL", map[string]any{"key": "blocks", "version": "v1"})
require.Equal(t, true, got["ok"])
assert.Equal(t, "hello world!", string(getURL(t, got["signed_download_url"].(string))))
}
func TestCacheServiceV2Lookups(t *testing.T) {
t.Parallel()
handler := newTestHandler(t, Policy{})
saved, _ := saveV2(t, handler, "deps-abc", "v1", []byte("x"))
require.Equal(t, true, saved["ok"])
t.Run("reports a miss for an unknown key", func(t *testing.T) {
got := v2Call(t, handler, testClient, "GetCacheEntryDownloadURL", map[string]any{"key": "nothing", "version": "v1"})
assert.Equal(t, false, got["ok"])
})
for _, field := range []string{"restore_keys", "restoreKeys"} {
t.Run("restore keys match by prefix, spelled "+field, func(t *testing.T) {
got := v2Call(t, handler, testClient, "GetCacheEntryDownloadURL", map[string]any{
"key": "deps-zzz", field: []string{"deps-"}, "version": "v1",
})
require.Equal(t, true, got["ok"])
assert.Equal(t, "deps-abc", got["matched_key"])
})
}
t.Run("an existing entry is not reserved twice", func(t *testing.T) {
again := v2Call(t, handler, testClient, "CreateCacheEntry", map[string]any{"key": "deps-abc", "version": "v1"})
assert.Equal(t, false, again["ok"])
})
t.Run("a prefix of an existing key is still reserved", func(t *testing.T) {
reserved := v2Call(t, handler, testClient, "CreateCacheEntry", map[string]any{"key": "deps", "version": "v1"})
require.Equal(t, true, reserved["ok"])
assert.NotEmpty(t, reserved["signed_upload_url"])
})
t.Run("a proxied job is handed the address its runner registered", func(t *testing.T) {
const proxy = "https://cache.example.invalid"
handler.RegisterJob("proxied", JobCredential{Repo: testRepo, PublicURL: proxy + "/"})
client := &http.Client{Transport: &bearerTransport{token: "proxied"}}
created := v2Call(t, handler, client, "CreateCacheEntry", map[string]any{"key": "proxied-key", "version": "v1"})
assert.True(t, strings.HasPrefix(created["signed_upload_url"].(string), proxy+blobPath+"/"))
got := v2Call(t, handler, client, "GetCacheEntryDownloadURL", map[string]any{"key": "deps-abc", "version": "v1"})
assert.True(t, strings.HasPrefix(got["signed_download_url"].(string), proxy+apiPath+"/artifacts/"))
})
t.Run("finalizing without a reservation is not ok", func(t *testing.T) {
got := v2Call(t, handler, testClient, "FinalizeCacheEntryUpload", map[string]any{
"key": "never-reserved", "version": "v1", "size_bytes": 1,
})
assert.Equal(t, false, got["ok"])
})
t.Run("finalizing with the wrong size is not ok", func(t *testing.T) {
created := v2Call(t, handler, testClient, "CreateCacheEntry", map[string]any{"key": "wrong-size", "version": "v1"})
require.Equal(t, http.StatusCreated, putBlob(t, created["signed_upload_url"].(string), []byte("four")))
got := v2Call(t, handler, testClient, "FinalizeCacheEntryUpload", map[string]any{
"key": "wrong-size", "version": "v1", "size_bytes": 99,
})
assert.Equal(t, false, got["ok"])
})
t.Run("a v1 client sees an entry written through v2", func(t *testing.T) {
resp, err := testClient.Get(fmt.Sprintf("%s%s/cache?keys=deps-abc&version=v1", handler.ExternalURL(), apiPath))
require.NoError(t, err)
defer resp.Body.Close()
require.Equal(t, http.StatusOK, resp.StatusCode)
got := map[string]any{}
require.NoError(t, json.UnmarshalRead(resp.Body, &got))
assert.Equal(t, "deps-abc", got["cacheKey"])
assert.NotEmpty(t, got["archiveLocation"])
})
t.Run("another repository sees nothing", func(t *testing.T) {
handler.RegisterJob("other-runtime-token", JobCredential{Repo: "other/repo"})
otherClient := &http.Client{Transport: &bearerTransport{token: "other-runtime-token"}}
got := v2Call(t, handler, otherClient, "GetCacheEntryDownloadURL", map[string]any{"key": "deps-abc", "version": "v1"})
assert.Equal(t, false, got["ok"])
})
t.Run("each ref saves and restores its own entry for the same key", func(t *testing.T) {
clients := map[string]*http.Client{"x": testClient}
for _, ref := range []string{"y", "z"} {
handler.RegisterJob(ref, JobCredential{Repo: testRepo, Ref: "refs/pull/" + ref + "/head"})
clients[ref] = &http.Client{Transport: &bearerTransport{token: ref}}
created := v2Call(t, handler, clients[ref], "CreateCacheEntry", map[string]any{"key": "deps-abc", "version": "v1"})
require.Equal(t, true, created["ok"])
require.Equal(t, http.StatusCreated, putBlob(t, created["signed_upload_url"].(string), []byte(ref)))
}
for _, ref := range []string{"y", "z"} {
require.Equal(t, true, v2Call(t, handler, clients[ref], "FinalizeCacheEntryUpload", map[string]any{"key": "deps-abc", "version": "v1", "size_bytes": 1})["ok"])
}
for content, client := range clients {
got := v2Call(t, handler, client, "GetCacheEntryDownloadURL", map[string]any{"key": "deps-abc", "version": "v1"})
assert.Equal(t, content, string(getURL(t, got["signed_download_url"].(string))))
}
})
}
func TestCacheServiceV2RefusalsAreTwirp(t *testing.T) {
t.Parallel()
handler := newTestHandler(t, Policy{})
tests := []struct {
name string
token string
body string
status int
code string
}{
{"no bearer at all", "", `{"key":"k","version":"v"}`, http.StatusUnauthorized, twirpUnauthenticated},
{"a bearer nobody registered", "not-a-job", `{"key":"k","version":"v"}`, http.StatusUnauthorized, twirpUnauthenticated},
{"a body that is not the request", testToken, `{`, http.StatusBadRequest, "malformed"},
{"a request missing its key", testToken, `{"version":"v"}`, http.StatusBadRequest, "invalid_argument"},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
req, err := http.NewRequestWithContext(t.Context(), http.MethodPost,
handler.ExternalURL()+cacheServiceV2Path+"/CreateCacheEntry", strings.NewReader(tt.body))
require.NoError(t, err)
if tt.token != "" {
req.Header.Set("Authorization", "Bearer "+tt.token)
}
resp, err := http.DefaultClient.Do(req)
require.NoError(t, err)
defer resp.Body.Close()
require.Equal(t, tt.status, resp.StatusCode)
got := map[string]string{}
require.NoError(t, json.UnmarshalRead(resp.Body, &got))
assert.Equal(t, tt.code, got["code"])
assert.NotEmpty(t, got["msg"])
})
}
}
func TestCacheServiceV2ConcurrentSaveAndRetries(t *testing.T) {
t.Parallel()
handler := newTestHandler(t, Policy{})
content := []byte("the cached archive")
const otherJob = "another-jobs-token"
defer handler.RegisterJob(otherJob, JobCredential{Repo: testRepo})()
reserve := func(client *http.Client) map[string]any {
return v2Call(t, handler, client, "CreateCacheEntry", map[string]any{"key": "shared", "version": "v1"})
}
first := reserve(testClient)
require.Equal(t, true, first["ok"])
retry := reserve(testClient)
assert.Equal(t, true, retry["ok"], "a job retrying its own reservation is not a conflict")
entryOf := func(reserved map[string]any) string {
parsed, err := url.Parse(reserved["signed_upload_url"].(string))
require.NoError(t, err)
return parsed.Path
}
assert.Equal(t, entryOf(first), entryOf(retry), "the same reservation, however freshly its URL is signed")
assert.Equal(t, false, reserve(&http.Client{Transport: &bearerTransport{token: otherJob}})["ok"],
"another job's reservation would be finalized against the first upload")
otherClient := &http.Client{Transport: &bearerTransport{token: otherJob}}
reserved := v2Call(t, handler, otherClient, "CreateCacheEntry", map[string]any{"key": "owned", "version": "v1"})
require.Equal(t, true, reserved["ok"])
require.Equal(t, http.StatusCreated, putBlob(t, reserved["signed_upload_url"].(string), content))
assert.Equal(t, false, v2Call(t, handler, testClient, "FinalizeCacheEntryUpload",
map[string]any{"key": "owned", "version": "v1", "size_bytes": strconv.Itoa(len(content))})["ok"],
"another job's upload is not this job's to finalize")
base := handler.ExternalURL() + apiPath
madeByV1 := reserveCache(t, testClient, base, "v1-made", "v1", int64(len(content)))
upload, err := http.NewRequestWithContext(t.Context(), http.MethodPatch,
fmt.Sprintf("%s/caches/%d", base, madeByV1), bytes.NewReader(content))
require.NoError(t, err)
upload.Header.Set("Content-Range", fmt.Sprintf("bytes 0-%d/*", len(content)-1))
resp, err := testClient.Do(upload)
require.NoError(t, err)
resp.Body.Close()
require.Equal(t, http.StatusOK, resp.StatusCode)
assert.Equal(t, false, v2Call(t, handler, testClient, "FinalizeCacheEntryUpload",
map[string]any{"key": "v1-made", "version": "v1", "size_bytes": strconv.Itoa(len(content))})["ok"],
"a reservation this job did not make through v2 is not its to commit")
finalized, _ := saveV2(t, handler, "deps", "v1", content)
require.Equal(t, true, finalized["ok"])
again := v2Call(t, handler, testClient, "FinalizeCacheEntryUpload", map[string]any{
"key": "deps", "version": "v1", "size_bytes": strconv.Itoa(len(content)),
})
assert.Equal(t, true, again["ok"], "a retry whose first answer was lost must not report a failed save")
assert.Equal(t, finalized["entry_id"], again["entry_id"])
}