You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
minio/cmd/admin-handlers.go

892 lines
24 KiB

/*
* Minio Cloud Storage, (C) 2016, 2017, 2018 Minio, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package cmd
import (
"bytes"
"context"
"encoding/base64"
"encoding/json"
"errors"
"io"
"net/http"
"strconv"
"strings"
"sync"
"time"
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
"github.com/gorilla/mux"
"github.com/minio/minio/cmd/logger"
"github.com/minio/minio/pkg/auth"
"github.com/minio/minio/pkg/handlers"
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
"github.com/minio/minio/pkg/madmin"
"github.com/minio/minio/pkg/quick"
"github.com/tidwall/gjson"
"github.com/tidwall/sjson"
)
const (
maxConfigJSONSize = 256 * 1024 // 256KiB
)
// Type-safe query params.
type mgmtQueryKey string
// Only valid query params for mgmt admin APIs.
const (
mgmtBucket mgmtQueryKey = "bucket"
mgmtPrefix mgmtQueryKey = "prefix"
mgmtClientToken mgmtQueryKey = "clientToken"
mgmtForceStart mgmtQueryKey = "forceStart"
)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
var (
// This struct literal represents the Admin API version that
// the server uses.
adminAPIVersionInfo = madmin.AdminAPIVersionInfo{
Version: "1",
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
)
// VersionHandler - GET /minio/admin/version
// -----------
// Returns Administration API version
func (a adminAPIHandlers) VersionHandler(w http.ResponseWriter, r *http.Request) {
adminAPIErr := checkAdminRequestAuthType(r, "")
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
if adminAPIErr != ErrNone {
writeErrorResponseJSON(w, adminAPIErr, r.URL)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
return
}
jsonBytes, err := json.Marshal(adminAPIVersionInfo)
if err != nil {
writeErrorResponseJSON(w, ErrInternalError, r.URL)
logger.LogIf(context.Background(), err)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeSuccessResponseJSON(w, jsonBytes)
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// ServiceStatusHandler - GET /minio/admin/v1/service
// ----------
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// Returns server version and uptime.
func (a adminAPIHandlers) ServiceStatusHandler(w http.ResponseWriter, r *http.Request) {
adminAPIErr := checkAdminRequestAuthType(r, "")
if adminAPIErr != ErrNone {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, adminAPIErr, r.URL)
return
}
// Fetch server version
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
serverVersion := madmin.ServerVersion{
Version: Version,
CommitID: CommitID,
}
// Fetch uptimes from all peers. This may fail to due to lack
// of read-quorum availability.
uptime, err := getPeerUptimes(globalAdminPeers)
if err != nil {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, toAPIErrorCode(err), r.URL)
logger.LogIf(context.Background(), err)
return
}
// Create API response
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
serverStatus := madmin.ServiceStatus{
ServerVersion: serverVersion,
Uptime: uptime,
}
// Marshal API response
jsonBytes, err := json.Marshal(serverStatus)
if err != nil {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, ErrInternalError, r.URL)
logger.LogIf(context.Background(), err)
return
}
// Reply with storage information (across nodes in a
// distributed setup) as json.
writeSuccessResponseJSON(w, jsonBytes)
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// ServiceStopNRestartHandler - POST /minio/admin/v1/service
// Body: {"action": <restart-action>}
// ----------
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// Restarts/Stops minio server gracefully. In a distributed setup,
// restarts all the servers in the cluster.
func (a adminAPIHandlers) ServiceStopNRestartHandler(w http.ResponseWriter, r *http.Request) {
adminAPIErr := checkAdminRequestAuthType(r, "")
if adminAPIErr != ErrNone {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, adminAPIErr, r.URL)
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
var sa madmin.ServiceAction
err := json.NewDecoder(r.Body).Decode(&sa)
if err != nil {
logger.LogIf(context.Background(), err)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, ErrRequestBodyParse, r.URL)
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
var serviceSig serviceSignal
switch sa.Action {
case madmin.ServiceActionValueRestart:
serviceSig = serviceRestart
case madmin.ServiceActionValueStop:
serviceSig = serviceStop
default:
writeErrorResponseJSON(w, ErrMalformedPOSTRequest, r.URL)
logger.LogIf(context.Background(), errors.New("Invalid service action received"))
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// Reply to the client before restarting minio server.
writeSuccessResponseHeadersOnly(w)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
sendServiceCmd(globalAdminPeers, serviceSig)
}
// ServerProperties holds some server information such as, version, region
// uptime, etc..
type ServerProperties struct {
Uptime time.Duration `json:"uptime"`
Version string `json:"version"`
CommitID string `json:"commitID"`
Region string `json:"region"`
SQSARN []string `json:"sqsARN"`
}
// ServerConnStats holds transferred bytes from/to the server
type ServerConnStats struct {
TotalInputBytes uint64 `json:"transferred"`
TotalOutputBytes uint64 `json:"received"`
Throughput uint64 `json:"throughput,omitempty"`
}
// ServerHTTPMethodStats holds total number of HTTP operations from/to the server,
// including the average duration the call was spent.
type ServerHTTPMethodStats struct {
Count uint64 `json:"count"`
AvgDuration string `json:"avgDuration"`
}
// ServerHTTPStats holds all type of http operations performed to/from the server
// including their average execution time.
type ServerHTTPStats struct {
TotalHEADStats ServerHTTPMethodStats `json:"totalHEADs"`
SuccessHEADStats ServerHTTPMethodStats `json:"successHEADs"`
TotalGETStats ServerHTTPMethodStats `json:"totalGETs"`
SuccessGETStats ServerHTTPMethodStats `json:"successGETs"`
TotalPUTStats ServerHTTPMethodStats `json:"totalPUTs"`
SuccessPUTStats ServerHTTPMethodStats `json:"successPUTs"`
TotalPOSTStats ServerHTTPMethodStats `json:"totalPOSTs"`
SuccessPOSTStats ServerHTTPMethodStats `json:"successPOSTs"`
TotalDELETEStats ServerHTTPMethodStats `json:"totalDELETEs"`
SuccessDELETEStats ServerHTTPMethodStats `json:"successDELETEs"`
}
// ServerInfoData holds storage, connections and other
// information of a given server.
type ServerInfoData struct {
StorageInfo StorageInfo `json:"storage"`
ConnStats ServerConnStats `json:"network"`
HTTPStats ServerHTTPStats `json:"http"`
Properties ServerProperties `json:"server"`
}
// ServerInfo holds server information result of one node
type ServerInfo struct {
Error string `json:"error"`
Addr string `json:"addr"`
Data *ServerInfoData `json:"data"`
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// ServerInfoHandler - GET /minio/admin/v1/info
// ----------
// Get server information
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
func (a adminAPIHandlers) ServerInfoHandler(w http.ResponseWriter, r *http.Request) {
// Authenticate request
// Setting the region as empty so as the mc server info command is irrespective to the region.
adminAPIErr := checkAdminRequestAuthType(r, "")
if adminAPIErr != ErrNone {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, adminAPIErr, r.URL)
return
}
// Web service response
reply := make([]ServerInfo, len(globalAdminPeers))
var wg sync.WaitGroup
// Gather server information for all nodes
for i, p := range globalAdminPeers {
wg.Add(1)
// Gather information from a peer in a goroutine
go func(idx int, peer adminPeer) {
defer wg.Done()
// Initialize server info at index
reply[idx] = ServerInfo{Addr: peer.addr}
serverInfoData, err := peer.cmdRunner.ServerInfo()
if err != nil {
reqInfo := (&logger.ReqInfo{}).AppendTags("peerAddress", peer.addr)
ctx := logger.SetReqInfo(context.Background(), reqInfo)
logger.LogIf(ctx, err)
reply[idx].Error = err.Error()
return
}
reply[idx].Data = &serverInfoData
}(i, p)
}
wg.Wait()
// Marshal API response
jsonBytes, err := json.Marshal(reply)
if err != nil {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, ErrInternalError, r.URL)
logger.LogIf(context.Background(), err)
return
}
// Reply with storage information (across nodes in a
// distributed setup) as json.
writeSuccessResponseJSON(w, jsonBytes)
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// extractHealInitParams - Validates params for heal init API.
func extractHealInitParams(r *http.Request) (bucket, objPrefix string,
hs madmin.HealOpts, clientToken string, forceStart bool,
err APIErrorCode) {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
vars := mux.Vars(r)
bucket = vars[string(mgmtBucket)]
objPrefix = vars[string(mgmtPrefix)]
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
if bucket == "" {
if objPrefix != "" {
// Bucket is required if object-prefix is given
err = ErrHealMissingBucket
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
} else if !IsValidBucketName(bucket) {
err = ErrInvalidBucketName
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// empty prefix is valid.
if !IsValidObjectPrefix(objPrefix) {
err = ErrInvalidObjectName
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
qParms := r.URL.Query()
if len(qParms[string(mgmtClientToken)]) > 0 {
clientToken = qParms[string(mgmtClientToken)][0]
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
if _, ok := qParms[string(mgmtForceStart)]; ok {
forceStart = true
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// ignore body if clientToken is provided
if clientToken == "" {
jerr := json.NewDecoder(r.Body).Decode(&hs)
if jerr != nil {
logger.LogIf(context.Background(), jerr)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
err = ErrRequestBodyParse
return
}
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
err = ErrNone
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// HealHandler - POST /minio/admin/v1/heal/
// -----------
// Start heal processing and return heal status items.
//
// On a successful heal sequence start, a unique client token is
// returned. Subsequent requests to this endpoint providing the client
// token will receive heal status records from the running heal
// sequence.
//
// If no client token is provided, and a heal sequence is in progress
// an error is returned with information about the running heal
// sequence. However, if the force-start flag is provided, the server
// aborts the running heal sequence and starts a new one.
func (a adminAPIHandlers) HealHandler(w http.ResponseWriter, r *http.Request) {
ctx := newContext(r, w, "Heal")
// Get object layer instance.
objLayer := newObjectLayerFn()
if objLayer == nil {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, ErrServerNotInitialized, r.URL)
return
}
// Validate request signature.
adminAPIErr := checkAdminRequestAuthType(r, "")
if adminAPIErr != ErrNone {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, adminAPIErr, r.URL)
return
}
// Check if this setup has an erasure coded backend.
if !globalIsXL {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, ErrHealNotImplemented, r.URL)
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
bucket, objPrefix, hs, clientToken, forceStart, apiErr := extractHealInitParams(r)
if apiErr != ErrNone {
writeErrorResponseJSON(w, apiErr, r.URL)
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
type healResp struct {
respBytes []byte
errCode APIErrorCode
errBody string
}
// Define a closure to start sending whitespace to client
// after 10s unless a response item comes in
keepConnLive := func(w http.ResponseWriter, respCh chan healResp) {
ticker := time.NewTicker(time.Second * 10)
defer ticker.Stop()
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
started := false
forLoop:
for {
select {
case <-ticker.C:
if !started {
// Start writing response to client
started = true
setCommonHeaders(w)
w.Header().Set("Content-Type", string(mimeJSON))
// Set 200 OK status
w.WriteHeader(200)
}
// Send whitespace and keep connection open
w.Write([]byte("\n\r"))
w.(http.Flusher).Flush()
case hr := <-respCh:
switch {
case hr.errCode == ErrNone:
writeSuccessResponseJSON(w, hr.respBytes)
case hr.errBody == "":
writeErrorResponseJSON(w, hr.errCode, r.URL)
default:
writeCustomErrorResponseJSON(w, hr.errCode, hr.errBody, r.URL)
}
break forLoop
}
}
}
// find number of disks in the setup
info := objLayer.StorageInfo(ctx)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
numDisks := info.Backend.OfflineDisks + info.Backend.OnlineDisks
if clientToken == "" {
// Not a status request
nh := newHealSequence(bucket, objPrefix, handlers.GetSourceIP(r),
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
numDisks, hs, forceStart)
respCh := make(chan healResp)
go func() {
respBytes, errCode, errMsg := globalAllHealState.LaunchNewHealSequence(nh)
hr := healResp{respBytes, errCode, errMsg}
respCh <- hr
}()
// Due to the force-starting functionality, the Launch
// call above can take a long time - to keep the
// connection alive, we start sending whitespace
keepConnLive(w, respCh)
} else {
// Since clientToken is given, fetch heal status from running
// heal sequence.
path := bucket + "/" + objPrefix
respBytes, errCode := globalAllHealState.PopHealStatusJSON(
path, clientToken)
if errCode != ErrNone {
writeErrorResponseJSON(w, errCode, r.URL)
} else {
writeSuccessResponseJSON(w, respBytes)
}
}
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// GetConfigHandler - GET /minio/admin/v1/config
// Get config.json of this minio setup.
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
func (a adminAPIHandlers) GetConfigHandler(w http.ResponseWriter, r *http.Request) {
ctx := newContext(r, w, "GetConfigHandler")
// Get current object layer instance.
objectAPI := newObjectLayerFn()
if objectAPI == nil {
writeErrorResponseJSON(w, ErrServerNotInitialized, r.URL)
return
}
// Validate request signature.
adminAPIErr := checkAdminRequestAuthType(r, "")
if adminAPIErr != ErrNone {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, adminAPIErr, r.URL)
return
}
config, err := readServerConfig(ctx, objectAPI)
if err != nil {
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
configData, err := json.MarshalIndent(config, "", "\t")
if err != nil {
logger.LogIf(ctx, err)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
password := config.GetCredential().SecretKey
econfigData, err := madmin.EncryptServerConfigData(password, configData)
if err != nil {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
writeSuccessResponseJSON(w, econfigData)
}
// Disable tidwall json array notation in JSON key path so
// users can set json with a key as a number.
// In tidwall json, notify.webhook.0 = val means { "notify" : { "webhook" : [val] }}
// In Minio, notify.webhook.0 = val means { "notify" : { "webhook" : {"0" : val}}}
func normalizeJSONKey(input string) (key string) {
subKeys := strings.Split(input, ".")
for i, k := range subKeys {
if i > 0 {
key += "."
}
if _, err := strconv.Atoi(k); err == nil {
key += ":" + k
} else {
key += k
}
}
return
}
// GetConfigHandler - GET /minio/admin/v1/config-keys
// Get some keys in config.json of this minio setup.
func (a adminAPIHandlers) GetConfigKeysHandler(w http.ResponseWriter, r *http.Request) {
ctx := newContext(r, w, "GetConfigKeysHandler")
// Get current object layer instance.
objectAPI := newObjectLayerFn()
if objectAPI == nil {
writeErrorResponseJSON(w, ErrServerNotInitialized, r.URL)
return
}
// Validate request signature.
adminAPIErr := checkAdminRequestAuthType(r, "")
if adminAPIErr != ErrNone {
writeErrorResponseJSON(w, adminAPIErr, r.URL)
return
}
var keys []string
queries := r.URL.Query()
for k := range queries {
keys = append(keys, k)
}
config, err := readServerConfig(ctx, objectAPI)
if err != nil {
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
configData, err := json.Marshal(config)
if err != nil {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
configStr := string(configData)
newConfigStr := `{}`
for _, key := range keys {
// sjson.Set does not return an error if key is empty
// we should check by ourselves here
if key == "" {
continue
}
val := gjson.Get(configStr, key)
if j, err := sjson.Set(newConfigStr, normalizeJSONKey(key), val.Value()); err == nil {
newConfigStr = j
}
}
password := config.GetCredential().SecretKey
econfigData, err := madmin.EncryptServerConfigData(password, []byte(newConfigStr))
if err != nil {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
writeSuccessResponseJSON(w, []byte(econfigData))
}
// toAdminAPIErrCode - converts errXLWriteQuorum error to admin API
// specific error.
func toAdminAPIErrCode(err error) APIErrorCode {
switch err {
case errXLWriteQuorum:
return ErrAdminConfigNoQuorum
default:
return toAPIErrorCode(err)
}
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// SetConfigHandler - PUT /minio/admin/v1/config
func (a adminAPIHandlers) SetConfigHandler(w http.ResponseWriter, r *http.Request) {
ctx := newContext(r, w, "SetConfigHandler")
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// Get current object layer instance.
objectAPI := newObjectLayerFn()
if objectAPI == nil {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, ErrServerNotInitialized, r.URL)
return
}
// Validate request signature.
adminAPIErr := checkAdminRequestAuthType(r, "")
if adminAPIErr != ErrNone {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, adminAPIErr, r.URL)
return
}
// Deny if WORM is enabled
if globalWORMEnabled {
writeErrorResponseJSON(w, ErrMethodNotAllowed, r.URL)
return
}
// Read configuration bytes from request body.
configBuf := make([]byte, maxConfigJSONSize+1)
n, err := io.ReadFull(r.Body, configBuf)
if err == nil {
// More than maxConfigSize bytes were available
writeErrorResponseJSON(w, ErrAdminConfigTooLarge, r.URL)
return
}
if err != io.ErrUnexpectedEOF {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, toAPIErrorCode(err), r.URL)
return
}
password := globalServerConfig.GetCredential().SecretKey
configBytes, err := madmin.DecryptServerConfigData(password, bytes.NewReader(configBuf[:n]))
if err != nil {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, ErrAdminConfigBadJSON, r.URL)
return
}
// Validate JSON provided in the request body: check the
// client has not sent JSON objects with duplicate keys.
if err = quick.CheckDuplicateKeys(string(configBytes)); err != nil {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, ErrAdminConfigBadJSON, r.URL)
return
}
var config serverConfig
err = json.Unmarshal(configBytes, &config)
if err != nil {
logger.LogIf(ctx, err)
writeCustomErrorResponseJSON(w, ErrAdminConfigBadJSON, err.Error(), r.URL)
return
}
// If credentials for the server are provided via environment,
// then credentials in the provided configuration must match.
if globalIsEnvCreds {
creds := globalServerConfig.GetCredential()
if config.Credential.AccessKey != creds.AccessKey ||
config.Credential.SecretKey != creds.SecretKey {
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, ErrAdminCredentialsMismatch, r.URL)
return
}
}
if err = config.Validate(); err != nil {
writeCustomErrorResponseJSON(w, ErrAdminConfigBadJSON, err.Error(), r.URL)
return
}
if err = saveServerConfig(ctx, objectAPI, &config); err != nil {
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
// Reply to the client before restarting minio server.
writeSuccessResponseHeadersOnly(w)
sendServiceCmd(globalAdminPeers, serviceRestart)
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
func convertValueType(elem []byte, jsonType gjson.Type) (interface{}, error) {
str := string(elem)
switch jsonType {
case gjson.False, gjson.True:
return strconv.ParseBool(str)
case gjson.JSON:
return gjson.Parse(str).Value(), nil
case gjson.String:
return str, nil
case gjson.Number:
return strconv.ParseFloat(str, 64)
default:
return nil, nil
}
}
// SetConfigKeysHandler - PUT /minio/admin/v1/config-keys
func (a adminAPIHandlers) SetConfigKeysHandler(w http.ResponseWriter, r *http.Request) {
ctx := newContext(r, w, "SetConfigKeysHandler")
// Get current object layer instance.
objectAPI := newObjectLayerFn()
if objectAPI == nil {
writeErrorResponseJSON(w, ErrServerNotInitialized, r.URL)
return
}
// Deny if WORM is enabled
if globalWORMEnabled {
writeErrorResponseJSON(w, ErrMethodNotAllowed, r.URL)
return
}
// Validate request signature.
adminAPIErr := checkAdminRequestAuthType(r, "")
if adminAPIErr != ErrNone {
writeErrorResponseJSON(w, adminAPIErr, r.URL)
return
}
// Load config
configStruct, err := readServerConfig(ctx, objectAPI)
if err != nil {
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
// Convert config to json bytes
configBytes, err := json.Marshal(configStruct)
if err != nil {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
configStr := string(configBytes)
queries := r.URL.Query()
password := globalServerConfig.GetCredential().SecretKey
// Set key values in the JSON config
for k := range queries {
// Decode encrypted data associated to the current key
encryptedElem, dErr := base64.StdEncoding.DecodeString(queries.Get(k))
if dErr != nil {
reqInfo := (&logger.ReqInfo{}).AppendTags("key", k)
ctx = logger.SetReqInfo(ctx, reqInfo)
logger.LogIf(ctx, dErr)
writeErrorResponseJSON(w, ErrAdminConfigBadJSON, r.URL)
return
}
elem, dErr := madmin.DecryptServerConfigData(password, bytes.NewBuffer([]byte(encryptedElem)))
if dErr != nil {
logger.LogIf(ctx, dErr)
writeErrorResponseJSON(w, ErrAdminConfigBadJSON, r.URL)
return
}
// Calculate the type of the current key from the
// original config json
jsonFieldType := gjson.Get(configStr, k).Type
// Convert passed value to json filed type
val, cErr := convertValueType(elem, jsonFieldType)
if cErr != nil {
writeCustomErrorResponseJSON(w, ErrAdminConfigBadJSON, cErr.Error(), r.URL)
return
}
// Set the key/value in the new json document
if s, sErr := sjson.Set(configStr, normalizeJSONKey(k), val); sErr == nil {
configStr = s
}
}
configBytes = []byte(configStr)
// Validate config
var config serverConfig
if err = json.Unmarshal(configBytes, &config); err != nil {
writeCustomErrorResponseJSON(w, ErrAdminConfigBadJSON, err.Error(), r.URL)
return
}
if err = config.Validate(); err != nil {
writeCustomErrorResponseJSON(w, ErrAdminConfigBadJSON, err.Error(), r.URL)
return
}
if err = config.TestNotificationTargets(); err != nil {
writeCustomErrorResponseJSON(w, ErrAdminConfigBadJSON, err.Error(), r.URL)
return
}
// If credentials for the server are provided via environment,
// then credentials in the provided configuration must match.
if globalIsEnvCreds {
creds := globalServerConfig.GetCredential()
if config.Credential.AccessKey != creds.AccessKey ||
config.Credential.SecretKey != creds.SecretKey {
writeErrorResponseJSON(w, ErrAdminCredentialsMismatch, r.URL)
return
}
}
if err = saveServerConfig(ctx, objectAPI, &config); err != nil {
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
// Send success response
writeSuccessResponseHeadersOnly(w)
sendServiceCmd(globalAdminPeers, serviceRestart)
}
// UpdateCredsHandler - POST /minio/admin/v1/config/credential
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// ----------
// Update credentials in a minio server. In a distributed setup,
// update all the servers in the cluster.
func (a adminAPIHandlers) UpdateCredentialsHandler(w http.ResponseWriter,
r *http.Request) {
ctx := newContext(r, w, "UpdateCredentialsHandler")
// Get current object layer instance.
objectAPI := newObjectLayerFn()
if objectAPI == nil {
writeErrorResponseJSON(w, ErrServerNotInitialized, r.URL)
return
}
// Avoid setting new credentials when they are already passed
// by the environment. Deny if WORM is enabled.
if globalIsEnvCreds {
writeErrorResponseJSON(w, ErrMethodNotAllowed, r.URL)
return
}
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// Authenticate request
adminAPIErr := checkAdminRequestAuthType(r, "")
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
if adminAPIErr != ErrNone {
writeErrorResponseJSON(w, adminAPIErr, r.URL)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
return
}
// Read configuration bytes from request body.
configBuf := make([]byte, maxConfigJSONSize+1)
n, err := io.ReadFull(r.Body, configBuf)
if err == nil {
// More than maxConfigSize bytes were available
writeErrorResponseJSON(w, ErrAdminConfigTooLarge, r.URL)
return
}
if err != io.ErrUnexpectedEOF {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, toAPIErrorCode(err), r.URL)
return
}
password := globalServerConfig.GetCredential().SecretKey
configBytes, err := madmin.DecryptServerConfigData(password, bytes.NewReader(configBuf[:n]))
if err != nil {
logger.LogIf(ctx, err)
writeErrorResponseJSON(w, ErrAdminConfigBadJSON, r.URL)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
return
}
// Decode request body
var req madmin.SetCredsReq
if err = json.Unmarshal(configBytes, &req); err != nil {
logger.LogIf(ctx, err)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
writeErrorResponseJSON(w, ErrRequestBodyParse, r.URL)
return
}
creds, err := auth.CreateCredentials(req.AccessKey, req.SecretKey)
if err != nil {
writeErrorResponseJSON(w, toAPIErrorCode(err), r.URL)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
return
}
// Acquire lock before updating global configuration.
globalServerConfigMu.Lock()
defer globalServerConfigMu.Unlock()
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
// Update local credentials in memory.
globalServerConfig.SetCredential(creds)
if err = saveServerConfig(ctx, objectAPI, globalServerConfig); err != nil {
writeErrorResponseJSON(w, toAdminAPIErrCode(err), r.URL)
return
}
// Notify all other Minio peers to update credentials
for host, err := range globalNotificationSys.LoadCredentials() {
if err != nil {
logger.GetReqInfo(ctx).SetTags("peerAddress", host.String())
logger.LogIf(ctx, err)
}
}
// Reply to the client before restarting minio server.
writeSuccessResponseHeadersOnly(w)
Move admin APIs to new path and add redesigned heal APIs (#5351) - Changes related to moving admin APIs - admin APIs now have an endpoint under /minio/admin - admin APIs are now versioned - a new API to server the version is added at "GET /minio/admin/version" and all API operations have the path prefix /minio/admin/v1/<operation> - new service stop API added - credentials change API is moved to /minio/admin/v1/config/credential - credentials change API and configuration get/set API now require TLS so that credentials are protected - all API requests now receive JSON - heal APIs are disabled as they will be changed substantially - Heal API changes Heal API is now provided at a single endpoint with the ability for a client to start a heal sequence on all the data in the server, a single bucket, or under a prefix within a bucket. When a heal sequence is started, the server returns a unique token that needs to be used for subsequent 'status' requests to fetch heal results. On each status request from the client, the server returns heal result records that it has accumulated since the previous status request. The server accumulates upto 1000 records and pauses healing further objects until the client requests for status. If the client does not request any further records for a long time, the server aborts the heal sequence automatically. A heal result record is returned for each entity healed on the server, such as system metadata, object metadata, buckets and objects, and has information about the before and after states on each disk. A client may request to force restart a heal sequence - this causes the running heal sequence to be aborted at the next safe spot and starts a new heal sequence.
7 years ago
}