Add health check endpoint (#18465)
* chore: add health check endpoint docs: update document about health check fix: fix up Sqlite3 ping. current ping will success even if the db file is missing fix: do not expose privacy information in output field * refactor: remove HealthChecker struct * Added `/api/healthz` to install routes. This was needed for using /api/healthz endpoint in Docker healthchecks, otherwise, Docker would never become healthy if using healthz endpoint and users would not be able to complete the installation of Gitea. * Update modules/cache/cache.go * fine tune * Remove unnecessary test code. Now there are 2 routes for installation (and maybe more in future) Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com> Co-authored-by: wxiaoguang <wxiaoguang@gmail.com> Co-authored-by: 6543 <6543@obermui.de> Co-authored-by: Marcos de Oliveira <marcossantos@furb.br>tokarchuk/v1.17
parent
3114cd30b8
commit
e933f31426
@ -0,0 +1,143 @@ |
|||||||
|
// Copyright 2022 The Gitea Authors. All rights reserved.
|
||||||
|
// Use of this source code is governed by a MIT-style
|
||||||
|
// license that can be found in the LICENSE file.
|
||||||
|
|
||||||
|
package healthcheck |
||||||
|
|
||||||
|
import ( |
||||||
|
"net/http" |
||||||
|
"os" |
||||||
|
"time" |
||||||
|
|
||||||
|
"code.gitea.io/gitea/models/db" |
||||||
|
"code.gitea.io/gitea/modules/cache" |
||||||
|
"code.gitea.io/gitea/modules/json" |
||||||
|
"code.gitea.io/gitea/modules/log" |
||||||
|
"code.gitea.io/gitea/modules/setting" |
||||||
|
) |
||||||
|
|
||||||
|
type status string |
||||||
|
|
||||||
|
const ( |
||||||
|
// pass healthy (acceptable aliases: "ok" to support Node's Terminus and "up" for Java's SpringBoot)
|
||||||
|
// fail unhealthy (acceptable aliases: "error" to support Node's Terminus and "down" for Java's SpringBoot), and
|
||||||
|
// warn healthy, with some concerns.
|
||||||
|
//
|
||||||
|
// ref https://datatracker.ietf.org/doc/html/draft-inadarei-api-health-check#section-3.1
|
||||||
|
// status: (required) indicates whether the service status is acceptable
|
||||||
|
// or not. API publishers SHOULD use following values for the field:
|
||||||
|
// The value of the status field is case-insensitive and is tightly
|
||||||
|
// related with the HTTP response code returned by the health endpoint.
|
||||||
|
// For "pass" status, HTTP response code in the 2xx-3xx range MUST be
|
||||||
|
// used. For "fail" status, HTTP response code in the 4xx-5xx range
|
||||||
|
// MUST be used. In case of the "warn" status, endpoints MUST return
|
||||||
|
// HTTP status in the 2xx-3xx range, and additional information SHOULD
|
||||||
|
// be provided, utilizing optional fields of the response.
|
||||||
|
pass status = "pass" |
||||||
|
fail status = "fail" |
||||||
|
warn status = "warn" |
||||||
|
) |
||||||
|
|
||||||
|
func (s status) ToHTTPStatus() int { |
||||||
|
if s == pass || s == warn { |
||||||
|
return http.StatusOK |
||||||
|
} |
||||||
|
return http.StatusFailedDependency |
||||||
|
} |
||||||
|
|
||||||
|
type checks map[string][]componentStatus |
||||||
|
|
||||||
|
// response is the data returned by the health endpoint, which will be marshaled to JSON format
|
||||||
|
type response struct { |
||||||
|
Status status `json:"status"` |
||||||
|
Description string `json:"description"` // a human-friendly description of the service
|
||||||
|
Checks checks `json:"checks,omitempty"` // The Checks Object, should be omitted on installation route
|
||||||
|
} |
||||||
|
|
||||||
|
// componentStatus presents one status of a single check object
|
||||||
|
// an object that provides detailed health statuses of additional downstream systems and endpoints
|
||||||
|
// which can affect the overall health of the main API.
|
||||||
|
type componentStatus struct { |
||||||
|
Status status `json:"status"` |
||||||
|
Time string `json:"time"` // the date-time, in ISO8601 format
|
||||||
|
Output string `json:"output,omitempty"` // this field SHOULD be omitted for "pass" state.
|
||||||
|
} |
||||||
|
|
||||||
|
// Check is the health check API handler
|
||||||
|
func Check(w http.ResponseWriter, r *http.Request) { |
||||||
|
rsp := response{ |
||||||
|
Status: pass, |
||||||
|
Description: setting.AppName, |
||||||
|
Checks: make(checks), |
||||||
|
} |
||||||
|
|
||||||
|
statuses := make([]status, 0) |
||||||
|
if setting.InstallLock { |
||||||
|
statuses = append(statuses, checkDatabase(rsp.Checks)) |
||||||
|
statuses = append(statuses, checkCache(rsp.Checks)) |
||||||
|
} |
||||||
|
for _, s := range statuses { |
||||||
|
if s != pass { |
||||||
|
rsp.Status = fail |
||||||
|
break |
||||||
|
} |
||||||
|
} |
||||||
|
|
||||||
|
data, _ := json.MarshalIndent(rsp, "", " ") |
||||||
|
w.Header().Set("Content-Type", "application/json") |
||||||
|
w.WriteHeader(rsp.Status.ToHTTPStatus()) |
||||||
|
_, _ = w.Write(data) |
||||||
|
} |
||||||
|
|
||||||
|
// database checks gitea database status
|
||||||
|
func checkDatabase(checks checks) status { |
||||||
|
st := componentStatus{} |
||||||
|
if err := db.GetEngine(db.DefaultContext).Ping(); err != nil { |
||||||
|
st.Status = fail |
||||||
|
st.Time = getCheckTime() |
||||||
|
log.Error("database ping failed with error: %v", err) |
||||||
|
} else { |
||||||
|
st.Status = pass |
||||||
|
st.Time = getCheckTime() |
||||||
|
} |
||||||
|
|
||||||
|
if setting.Database.UseSQLite3 && st.Status == pass { |
||||||
|
if !setting.EnableSQLite3 { |
||||||
|
st.Status = fail |
||||||
|
st.Time = getCheckTime() |
||||||
|
log.Error("SQLite3 health check failed with error: %v", "this Gitea binary is built without SQLite3 enabled") |
||||||
|
} else { |
||||||
|
if _, err := os.Stat(setting.Database.Path); err != nil { |
||||||
|
st.Status = fail |
||||||
|
st.Time = getCheckTime() |
||||||
|
log.Error("SQLite3 file exists check failed with error: %v", err) |
||||||
|
} |
||||||
|
} |
||||||
|
} |
||||||
|
|
||||||
|
checks["database:ping"] = []componentStatus{st} |
||||||
|
return st.Status |
||||||
|
} |
||||||
|
|
||||||
|
// cache checks gitea cache status
|
||||||
|
func checkCache(checks checks) status { |
||||||
|
if !setting.CacheService.Enabled { |
||||||
|
return pass |
||||||
|
} |
||||||
|
|
||||||
|
st := componentStatus{} |
||||||
|
if err := cache.Ping(); err != nil { |
||||||
|
st.Status = fail |
||||||
|
st.Time = getCheckTime() |
||||||
|
log.Error("cache ping failed with error: %v", err) |
||||||
|
} else { |
||||||
|
st.Status = pass |
||||||
|
st.Time = getCheckTime() |
||||||
|
} |
||||||
|
checks["cache:ping"] = []componentStatus{st} |
||||||
|
return st.Status |
||||||
|
} |
||||||
|
|
||||||
|
func getCheckTime() string { |
||||||
|
return time.Now().UTC().Format(time.RFC3339) |
||||||
|
} |
Loading…
Reference in new issue