minio/cmd/storage-rpc-client.go
Harshavardhana fb96779a8a Add large bucket support for erasure coded backend (#5160)
This PR implements an object layer which
combines input erasure sets of XL layers
into a unified namespace.

This object layer extends the existing
erasure coded implementation, it is assumed
in this design that providing > 16 disks is
a static configuration as well i.e if you started
the setup with 32 disks with 4 sets 8 disks per
pack then you would need to provide 4 sets always.

Some design details and restrictions:

- Objects are distributed using consistent ordering
  to a unique erasure coded layer.
- Each pack has its own dsync so locks are synchronized
  properly at pack (erasure layer).
- Each pack still has a maximum of 16 disks
  requirement, you can start with multiple
  such sets statically.
- Static sets set of disks and cannot be
  changed, there is no elastic expansion allowed.
- Static sets set of disks and cannot be
  changed, there is no elastic removal allowed.
- ListObjects() across sets can be noticeably
  slower since List happens on all servers,
  and is merged at this sets layer.

Fixes #5465
Fixes #5464
Fixes #5461
Fixes #5460
Fixes #5459
Fixes #5458
Fixes #5460
Fixes #5488
Fixes #5489
Fixes #5497
Fixes #5496
2018-02-15 17:45:57 -08:00

316 lines
8.5 KiB
Go

/*
* Minio Cloud Storage, (C) 2016 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"
"io"
"net"
"net/rpc"
"path"
"strings"
"github.com/minio/minio/pkg/disk"
)
type networkStorage struct {
rpcClient *AuthRPCClient
connected bool
}
const (
storageRPCPath = "/storage"
)
func isErrorNetworkDisconnect(err error) bool {
if err == nil {
return false
}
if _, ok := err.(*net.OpError); ok {
return true
}
if err == rpc.ErrShutdown {
return true
}
return false
}
// Converts rpc.ServerError to underlying error. This function is
// written so that the storageAPI errors are consistent across network
// disks as well.
func toStorageErr(err error) error {
if err == nil {
return nil
}
if isErrorNetworkDisconnect(err) {
return errDiskNotFound
}
switch err.Error() {
case io.EOF.Error():
return io.EOF
case io.ErrUnexpectedEOF.Error():
return io.ErrUnexpectedEOF
case errUnexpected.Error():
return errUnexpected
case errDiskFull.Error():
return errDiskFull
case errVolumeNotFound.Error():
return errVolumeNotFound
case errVolumeExists.Error():
return errVolumeExists
case errFileNotFound.Error():
return errFileNotFound
case errFileNameTooLong.Error():
return errFileNameTooLong
case errFileAccessDenied.Error():
return errFileAccessDenied
case errIsNotRegular.Error():
return errIsNotRegular
case errVolumeNotEmpty.Error():
return errVolumeNotEmpty
case errVolumeAccessDenied.Error():
return errVolumeAccessDenied
case errCorruptedFormat.Error():
return errCorruptedFormat
case errUnformattedDisk.Error():
return errUnformattedDisk
case errInvalidAccessKeyID.Error():
return errInvalidAccessKeyID
case errAuthentication.Error():
return errAuthentication
case errRPCAPIVersionUnsupported.Error():
return errRPCAPIVersionUnsupported
case errServerTimeMismatch.Error():
return errServerTimeMismatch
}
return err
}
// Initialize new storage rpc client.
func newStorageRPC(endpoint Endpoint) StorageAPI {
// Dial minio rpc storage http path.
rpcPath := path.Join(minioReservedBucketPath, storageRPCPath, endpoint.Path)
serverCred := globalServerConfig.GetCredential()
disk := &networkStorage{
rpcClient: newAuthRPCClient(authConfig{
accessKey: serverCred.AccessKey,
secretKey: serverCred.SecretKey,
serverAddr: endpoint.Host,
serviceEndpoint: rpcPath,
secureConn: globalIsSSL,
serviceName: "Storage",
disableReconnect: true,
}),
}
// Attempt a remote login.
disk.connected = disk.rpcClient.Login() == nil
return disk
}
// Stringer provides a canonicalized representation of network device.
func (n *networkStorage) String() string {
// Remove the storage RPC path prefix, internal paths are meaningless.
serviceEndpoint := strings.TrimPrefix(n.rpcClient.ServiceEndpoint(),
path.Join(minioReservedBucketPath, storageRPCPath))
// Check for the transport layer being used.
scheme := "http"
if n.rpcClient.config.secureConn {
scheme = "https"
}
// Finally construct the disk endpoint in http://<server>/<path> form.
return scheme + "://" + n.rpcClient.ServerAddr() + path.Join("/", serviceEndpoint)
}
func (n *networkStorage) Close() error {
n.connected = false
return toStorageErr(n.rpcClient.Close())
}
func (n *networkStorage) IsOnline() bool {
return n.connected
}
func (n *networkStorage) call(handler string, args interface {
SetAuthToken(string)
SetRPCAPIVersion(semVersion)
}, reply interface{}) error {
if !n.connected {
return errDiskNotFound
}
if err := n.rpcClient.Call(handler, args, reply); err != nil {
if isErrorNetworkDisconnect(err) {
n.connected = false
}
return toStorageErr(err)
}
return nil
}
// DiskInfo - fetch disk information for a remote disk.
func (n *networkStorage) DiskInfo() (info disk.Info, err error) {
args := AuthRPCArgs{}
if err = n.call("Storage.DiskInfoHandler", &args, &info); err != nil {
return disk.Info{}, err
}
return info, nil
}
// MakeVol - create a volume on a remote disk.
func (n *networkStorage) MakeVol(volume string) (err error) {
reply := AuthRPCReply{}
args := GenericVolArgs{Vol: volume}
return n.call("Storage.MakeVolHandler", &args, &reply)
}
// ListVols - List all volumes on a remote disk.
func (n *networkStorage) ListVols() (vols []VolInfo, err error) {
ListVols := ListVolsReply{}
if err = n.call("Storage.ListVolsHandler", &AuthRPCArgs{}, &ListVols); err != nil {
return nil, err
}
return ListVols.Vols, nil
}
// StatVol - get volume info over the network.
func (n *networkStorage) StatVol(volume string) (volInfo VolInfo, err error) {
args := GenericVolArgs{Vol: volume}
if err = n.call("Storage.StatVolHandler", &args, &volInfo); err != nil {
return VolInfo{}, err
}
return volInfo, nil
}
// DeleteVol - Deletes a volume over the network.
func (n *networkStorage) DeleteVol(volume string) (err error) {
reply := AuthRPCReply{}
args := GenericVolArgs{Vol: volume}
return n.call("Storage.DeleteVolHandler", &args, &reply)
}
// File operations.
func (n *networkStorage) PrepareFile(volume, path string, length int64) (err error) {
reply := AuthRPCReply{}
return n.call("Storage.PrepareFileHandler", &PrepareFileArgs{
Vol: volume,
Path: path,
Size: length,
}, &reply)
}
// AppendFile - append file writes buffer to a remote network path.
func (n *networkStorage) AppendFile(volume, path string, buffer []byte) (err error) {
reply := AuthRPCReply{}
return n.call("Storage.AppendFileHandler", &AppendFileArgs{
Vol: volume,
Path: path,
Buffer: buffer,
}, &reply)
}
// StatFile - get latest Stat information for a file at path.
func (n *networkStorage) StatFile(volume, path string) (fileInfo FileInfo, err error) {
if err = n.call("Storage.StatFileHandler", &StatFileArgs{
Vol: volume,
Path: path,
}, &fileInfo); err != nil {
return FileInfo{}, err
}
return fileInfo, nil
}
// ReadAll - reads entire contents of the file at path until EOF, returns the
// contents in a byte slice. Returns buf == nil if err != nil.
// This API is meant to be used on files which have small memory footprint, do
// not use this on large files as it would cause server to crash.
func (n *networkStorage) ReadAll(volume, path string) (buf []byte, err error) {
if err = n.call("Storage.ReadAllHandler", &ReadAllArgs{
Vol: volume,
Path: path,
}, &buf); err != nil {
return nil, err
}
return buf, nil
}
// ReadFile - reads a file at remote path and fills the buffer.
func (n *networkStorage) ReadFile(volume string, path string, offset int64, buffer []byte, verifier *BitrotVerifier) (m int64, err error) {
defer func() {
if r := recover(); r != nil {
// Recover any panic from allocation, and return error.
err = bytes.ErrTooLarge
}
}() // Do not crash the server.
args := ReadFileArgs{
Vol: volume,
Path: path,
Offset: offset,
Buffer: buffer,
Verified: true, // mark read as verified by default
}
if verifier != nil {
args.Algo = verifier.algorithm
args.ExpectedHash = verifier.sum
args.Verified = verifier.IsVerified()
}
var result []byte
err = n.call("Storage.ReadFileHandler", &args, &result)
// Copy results to buffer.
copy(buffer, result)
// Return length of result, err if any.
return int64(len(result)), err
}
// ListDir - list all entries at prefix.
func (n *networkStorage) ListDir(volume, path string) (entries []string, err error) {
if err = n.call("Storage.ListDirHandler", &ListDirArgs{
Vol: volume,
Path: path,
}, &entries); err != nil {
return nil, err
}
// Return successfully unmarshalled results.
return entries, nil
}
// DeleteFile - Delete a file at path.
func (n *networkStorage) DeleteFile(volume, path string) (err error) {
reply := AuthRPCReply{}
return n.call("Storage.DeleteFileHandler", &DeleteFileArgs{
Vol: volume,
Path: path,
}, &reply)
}
// RenameFile - rename a remote file from source to destination.
func (n *networkStorage) RenameFile(srcVolume, srcPath, dstVolume, dstPath string) (err error) {
reply := AuthRPCReply{}
return n.call("Storage.RenameFileHandler", &RenameFileArgs{
SrcVol: srcVolume,
SrcPath: srcPath,
DstVol: dstVolume,
DstPath: dstPath,
}, &reply)
}