minio/cmd/fs-v1-helpers.go
Harshavardhana 12f67d47f1 Fix a possible race during PutObject() (#5376)
Under any concurrent removeObjects in progress
might have removed the parents of the same prefix
for which there is an ongoing putObject request.
An inconsistent situation may arise as explained
below even under sufficient locking.

PutObject is almost successful at the last stage when
a temporary file is renamed to its actual namespace
at `a/b/c/object1`. Concurrently a RemoveObject is
also in progress at the same prefix for an `a/b/c/object2`.

To create the object1 at location `a/b/c` PutObject has
to create all the parents recursively.

```
a/b/c - os.MkdirAll loops through has now created
        'a/' and 'b/' about to create 'c/'
a/b/c/object2 - at this point 'c/' and 'object2'
        are deleted about to delete b/
```

Now for os.MkdirAll loop the expected situation is
that top level parent 'a/b/' exists which it created
, such that it can create 'c/' - since removeObject
and putObject do not compete for lock due to holding
locks at different resources. removeObject proceeds
to delete parent 'b/' since 'c/' is not yet present,
once deleted 'os.MkdirAll' would receive an error as
syscall.ENOENT which would fail the putObject request.

This PR tries to address this issue by implementing
a safer/guarded approach where we would retry an operation
such as `os.MkdirAll` and `os.Rename` if both operations
observe syscall.ENOENT.

Fixes #5254
2018-01-13 22:43:02 +05:30

429 lines
12 KiB
Go

/*
* 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 (
"io"
"os"
pathutil "path"
"runtime"
"github.com/minio/minio/pkg/errors"
)
// Removes only the file at given path does not remove
// any parent directories, handles long paths for
// windows automatically.
func fsRemoveFile(filePath string) (err error) {
if filePath == "" {
return errors.Trace(errInvalidArgument)
}
if err = checkPathLength(filePath); err != nil {
return errors.Trace(err)
}
if err = os.Remove((filePath)); err != nil {
return osErrToFSFileErr(err)
}
return nil
}
// Removes all files and folders at a given path, handles
// long paths for windows automatically.
func fsRemoveAll(dirPath string) (err error) {
if dirPath == "" {
return errors.Trace(errInvalidArgument)
}
if err = checkPathLength(dirPath); err != nil {
return errors.Trace(err)
}
if err = os.RemoveAll(dirPath); err != nil {
if os.IsPermission(err) {
return errors.Trace(errVolumeAccessDenied)
} else if isSysErrNotEmpty(err) {
return errors.Trace(errVolumeNotEmpty)
}
return errors.Trace(err)
}
return nil
}
// Removes a directory only if its empty, handles long
// paths for windows automatically.
func fsRemoveDir(dirPath string) (err error) {
if dirPath == "" {
return errors.Trace(errInvalidArgument)
}
if err = checkPathLength(dirPath); err != nil {
return errors.Trace(err)
}
if err = os.Remove((dirPath)); err != nil {
if os.IsNotExist(err) {
return errors.Trace(errVolumeNotFound)
} else if isSysErrNotEmpty(err) {
return errors.Trace(errVolumeNotEmpty)
}
return errors.Trace(err)
}
return nil
}
// Creates a new directory, parent dir should exist
// otherwise returns an error. If directory already
// exists returns an error. Windows long paths
// are handled automatically.
func fsMkdir(dirPath string) (err error) {
if dirPath == "" {
return errors.Trace(errInvalidArgument)
}
if err = checkPathLength(dirPath); err != nil {
return errors.Trace(err)
}
if err = os.Mkdir((dirPath), 0777); err != nil {
if os.IsExist(err) {
return errors.Trace(errVolumeExists)
} else if os.IsPermission(err) {
return errors.Trace(errDiskAccessDenied)
} else if isSysErrNotDir(err) {
// File path cannot be verified since
// one of the parents is a file.
return errors.Trace(errDiskAccessDenied)
} else if isSysErrPathNotFound(err) {
// Add specific case for windows.
return errors.Trace(errDiskAccessDenied)
}
return errors.Trace(err)
}
return nil
}
// fsStat is a low level call which validates input arguments
// and checks input length upto supported maximum. Does
// not perform any higher layer interpretation of files v/s
// directories. For higher level interpretation look at
// fsStatFileDir, fsStatFile, fsStatDir.
func fsStat(statLoc string) (os.FileInfo, error) {
if statLoc == "" {
return nil, errors.Trace(errInvalidArgument)
}
if err := checkPathLength(statLoc); err != nil {
return nil, errors.Trace(err)
}
fi, err := os.Stat((statLoc))
if err != nil {
return nil, errors.Trace(err)
}
return fi, nil
}
// Lookup if volume exists, returns volume attributes upon success.
func fsStatVolume(volume string) (os.FileInfo, error) {
fi, err := fsStat(volume)
if err != nil {
err = errors.Cause(err)
if os.IsNotExist(err) {
return nil, errors.Trace(errVolumeNotFound)
} else if os.IsPermission(err) {
return nil, errors.Trace(errVolumeAccessDenied)
}
return nil, errors.Trace(err)
}
if !fi.IsDir() {
return nil, errors.Trace(errVolumeAccessDenied)
}
return fi, nil
}
// Is a one place function which converts all os.PathError
// into a more FS object layer friendly form, converts
// known errors into their typed form for top level
// interpretation.
func osErrToFSFileErr(err error) error {
if err == nil {
return nil
}
err = errors.Cause(err)
if os.IsNotExist(err) {
return errors.Trace(errFileNotFound)
}
if os.IsPermission(err) {
return errors.Trace(errFileAccessDenied)
}
if isSysErrNotDir(err) {
return errors.Trace(errFileAccessDenied)
}
if isSysErrPathNotFound(err) {
return errors.Trace(errFileNotFound)
}
return err
}
// Lookup if directory exists, returns directory attributes upon success.
func fsStatDir(statDir string) (os.FileInfo, error) {
fi, err := fsStat(statDir)
if err != nil {
return nil, osErrToFSFileErr(err)
}
if !fi.IsDir() {
return nil, errors.Trace(errFileAccessDenied)
}
return fi, nil
}
// Lookup if file exists, returns file attributes upon success.
func fsStatFile(statFile string) (os.FileInfo, error) {
fi, err := fsStat(statFile)
if err != nil {
return nil, osErrToFSFileErr(err)
}
if fi.IsDir() {
return nil, errors.Trace(errFileAccessDenied)
}
return fi, nil
}
// Opens the file at given path, optionally from an offset. Upon success returns
// a readable stream and the size of the readable stream.
func fsOpenFile(readPath string, offset int64) (io.ReadCloser, int64, error) {
if readPath == "" || offset < 0 {
return nil, 0, errors.Trace(errInvalidArgument)
}
if err := checkPathLength(readPath); err != nil {
return nil, 0, errors.Trace(err)
}
fr, err := os.Open((readPath))
if err != nil {
return nil, 0, osErrToFSFileErr(err)
}
// Stat to get the size of the file at path.
st, err := os.Stat((readPath))
if err != nil {
return nil, 0, errors.Trace(err)
}
// Verify if its not a regular file, since subsequent Seek is undefined.
if !st.Mode().IsRegular() {
return nil, 0, errors.Trace(errIsNotRegular)
}
// Seek to the requested offset.
if offset > 0 {
_, err = fr.Seek(offset, os.SEEK_SET)
if err != nil {
return nil, 0, errors.Trace(err)
}
}
// Success.
return fr, st.Size(), nil
}
// Creates a file and copies data from incoming reader. Staging buffer is used by io.CopyBuffer.
func fsCreateFile(filePath string, reader io.Reader, buf []byte, fallocSize int64) (int64, error) {
if filePath == "" || reader == nil {
return 0, errors.Trace(errInvalidArgument)
}
if err := checkPathLength(filePath); err != nil {
return 0, errors.Trace(err)
}
if err := mkdirAll(pathutil.Dir(filePath), 0777); err != nil {
return 0, errors.Trace(err)
}
if err := checkDiskFree(pathutil.Dir(filePath), fallocSize); err != nil {
return 0, errors.Trace(err)
}
writer, err := os.OpenFile((filePath), os.O_CREATE|os.O_WRONLY, 0666)
if err != nil {
return 0, osErrToFSFileErr(err)
}
defer writer.Close()
// Fallocate only if the size is final object is known.
if fallocSize > 0 {
if err = fsFAllocate(int(writer.Fd()), 0, fallocSize); err != nil {
return 0, errors.Trace(err)
}
}
var bytesWritten int64
if buf != nil {
bytesWritten, err = io.CopyBuffer(writer, reader, buf)
if err != nil {
return 0, errors.Trace(err)
}
} else {
bytesWritten, err = io.Copy(writer, reader)
if err != nil {
return 0, errors.Trace(err)
}
}
return bytesWritten, nil
}
// Removes uploadID at destination path.
func fsRemoveUploadIDPath(basePath, uploadIDPath string) error {
if basePath == "" || uploadIDPath == "" {
return errors.Trace(errInvalidArgument)
}
if err := checkPathLength(basePath); err != nil {
return errors.Trace(err)
}
if err := checkPathLength(uploadIDPath); err != nil {
return errors.Trace(err)
}
// List all the entries in uploadID.
entries, err := readDir(uploadIDPath)
if err != nil && err != errFileNotFound {
return errors.Trace(err)
}
// Delete all the entries obtained from previous readdir.
for _, entryPath := range entries {
err = fsDeleteFile(basePath, pathJoin(uploadIDPath, entryPath))
if err != nil && err != errFileNotFound {
return errors.Trace(err)
}
}
fsRemoveDir(uploadIDPath)
return nil
}
// fsFAllocate is similar to Fallocate but provides a convenient
// wrapper to handle various operating system specific errors.
func fsFAllocate(fd int, offset int64, len int64) (err error) {
e := Fallocate(fd, offset, len)
// Ignore errors when Fallocate is not supported in the current system
if e != nil && !isSysErrNoSys(e) && !isSysErrOpNotSupported(e) {
switch {
case isSysErrNoSpace(e):
err = errDiskFull
case isSysErrIO(e):
err = e
default:
// For errors: EBADF, EINTR, EINVAL, ENODEV, EPERM, ESPIPE and ETXTBSY
// Appending was failed anyway, returns unexpected error
err = errUnexpected
}
return err
}
return nil
}
// Renames source path to destination path, creates all the
// missing parents if they don't exist.
func fsRenameFile(sourcePath, destPath string) error {
if err := checkPathLength(sourcePath); err != nil {
return errors.Trace(err)
}
if err := checkPathLength(destPath); err != nil {
return errors.Trace(err)
}
// Verify if source path exists.
if _, err := os.Stat(sourcePath); err != nil {
return osErrToFSFileErr(err)
}
if err := renameAll(sourcePath, destPath); err != nil {
return errors.Trace(err)
}
return nil
}
// fsDeleteFile is a wrapper for deleteFile(), after checking the path length.
func fsDeleteFile(basePath, deletePath string) error {
if err := checkPathLength(basePath); err != nil {
return errors.Trace(err)
}
if err := checkPathLength(deletePath); err != nil {
return errors.Trace(err)
}
return deleteFile(basePath, deletePath)
}
// fsRemoveMeta safely removes a locked file and takes care of Windows special case
func fsRemoveMeta(basePath, deletePath, tmpDir string) error {
// Special case for windows please read through.
if runtime.GOOS == globalWindowsOSName {
// Ordinarily windows does not permit deletion or renaming of files still
// in use, but if all open handles to that file were opened with FILE_SHARE_DELETE
// then it can permit renames and deletions of open files.
//
// There are however some gotchas with this, and it is worth listing them here.
// Firstly, Windows never allows you to really delete an open file, rather it is
// flagged as delete pending and its entry in its directory remains visible
// (though no new file handles may be opened to it) and when the very last
// open handle to the file in the system is closed, only then is it truly
// deleted. Well, actually only sort of truly deleted, because Windows only
// appears to remove the file entry from the directory, but in fact that
// entry is merely hidden and actually still exists and attempting to create
// a file with the same name will return an access denied error. How long it
// silently exists for depends on a range of factors, but put it this way:
// if your code loops creating and deleting the same file name as you might
// when operating a lock file, you're going to see lots of random spurious
// access denied errors and truly dismal lock file performance compared to POSIX.
//
// We work-around these un-POSIX file semantics by taking a dual step to
// deleting files. Firstly, it renames the file to tmp location into multipartTmpBucket
// We always open files with FILE_SHARE_DELETE permission enabled, with that
// flag Windows permits renaming and deletion, and because the name was changed
// to a very random name somewhere not in its origin directory before deletion,
// you don't see those unexpected random errors when creating files with the
// same name as a recently deleted file as you do anywhere else on Windows.
// Because the file is probably not in its original containing directory any more,
// deletions of that directory will not fail with "directory not empty" as they
// otherwise normally would either.
tmpPath := pathJoin(tmpDir, mustGetUUID())
fsRenameFile(deletePath, tmpPath)
// Proceed to deleting the directory if empty
fsDeleteFile(basePath, pathutil.Dir(deletePath))
// Finally delete the renamed file.
return fsDeleteFile(tmpDir, tmpPath)
}
return fsDeleteFile(basePath, deletePath)
}