def test_do_stat_err(self):
def mock_os_stat_eacces(path):
raise OSError(errno.EACCES, os.strerror(errno.EACCES))
try:
with patch('os.stat', mock_os_stat_eacces):
fs.do_stat('/tmp')
except GlusterFileSystemOSError:
pass
else:
self.fail("GlusterFileSystemOSError expected")
def test_do_stat_eio_ten(self):
def mock_os_stat_eio(path):
raise OSError(errno.EIO, os.strerror(errno.EIO))
try:
with patch('os.stat', mock_os_stat_eio):
fs.do_stat('/tmp')
except GlusterFileSystemOSError:
pass
else:
self.fail("GlusterFileSystemOSError expected")
def test_do_stat_eio_twice(self):
count = [0]
_os_stat = os.stat
def mock_os_stat_eio(path):
count[0] += 1
if count[0] <= 2:
raise OSError(errno.EIO, os.strerror(errno.EIO))
return _os_stat(path)
with patch('os.stat', mock_os_stat_eio):
fs.do_stat('/tmp') is not None
def test_do_stat_eio_ten(self):
def mock_os_stat_eio(path):
raise OSError(errno.EIO, os.strerror(errno.EIO))
try:
with patch('os.stat', mock_os_stat_eio):
fs.do_stat('/tmp')
except GlusterFileSystemOSError:
pass
else:
self.fail("GlusterFileSystemOSError expected")
def test_do_stat_eio_twice(self):
count = [0]
_os_stat = os.stat
def mock_os_stat_eio(path):
count[0] += 1
if count[0] <= 2:
raise OSError(errno.EIO, os.strerror(errno.EIO))
return _os_stat(path)
with patch('os.stat', mock_os_stat_eio):
fs.do_stat('/tmp') is not None
def test_do_stat_err(self):
def mock_os_stat_eacces(path):
raise OSError(errno.EACCES, os.strerror(errno.EACCES))
try:
with patch('os.stat', mock_os_stat_eacces):
fs.do_stat('/tmp')
except GlusterFileSystemOSError:
pass
else:
self.fail("GlusterFileSystemOSError expected")
def get_object_metadata(obj_path_or_fd, stats=None):
"""
Return metadata of object.
"""
if not stats:
if isinstance(obj_path_or_fd, int):
# We are given a file descriptor, so this is an invocation from the
# DiskFile.open() method.
stats = do_fstat(obj_path_or_fd)
else:
# We are given a path to the object when the
# DiskDir.list_objects_iter method invokes us.
stats = do_stat(obj_path_or_fd)
if not stats:
metadata = {}
else:
is_dir = stat.S_ISDIR(stats.st_mode)
metadata = {
X_TYPE: OBJECT,
X_TIMESTAMP: normalize_timestamp(stats.st_ctime),
X_CONTENT_TYPE: DIR_TYPE if is_dir else FILE_TYPE,
X_OBJECT_TYPE: DIR_NON_OBJECT if is_dir else FILE,
X_CONTENT_LENGTH: 0 if is_dir else stats.st_size,
X_MTIME: 0 if is_dir else normalize_timestamp(stats.st_mtime),
X_ETAG: md5().hexdigest() if is_dir else _get_etag(obj_path_or_fd)
}
return metadata
def get_object_metadata(obj_path_or_fd):
"""
Return metadata of object.
"""
if isinstance(obj_path_or_fd, int):
# We are given a file descriptor, so this is an invocation from the
# DiskFile.open() method.
stats = do_fstat(obj_path_or_fd)
else:
# We are given a path to the object when the DiskDir.list_objects_iter
# method invokes us.
stats = do_stat(obj_path_or_fd)
if not stats:
metadata = {}
else:
is_dir = stat.S_ISDIR(stats.st_mode)
metadata = {
X_TYPE: OBJECT,
X_TIMESTAMP: normalize_timestamp(stats.st_ctime),
X_CONTENT_TYPE: DIR_TYPE if is_dir else FILE_TYPE,
X_OBJECT_TYPE: DIR_NON_OBJECT if is_dir else FILE,
X_CONTENT_LENGTH: 0 if is_dir else stats.st_size,
X_ETAG: md5().hexdigest() if is_dir else _get_etag(obj_path_or_fd)}
return metadata
def test_do_stat(self):
tmpdir = mkdtemp()
try:
fd, tmpfile = mkstemp(dir=tmpdir)
buf1 = os.stat(tmpfile)
buf2 = fs.do_stat(tmpfile)
assert buf1 == buf2
finally:
os.close(fd)
os.remove(tmpfile)
os.rmdir(tmpdir)
def test_do_stat(self):
tmpdir = mkdtemp()
try:
fd, tmpfile = mkstemp(dir=tmpdir)
buf1 = os.stat(tmpfile)
buf2 = fs.do_stat(tmpfile)
assert buf1 == buf2
finally:
os.close(fd)
os.remove(tmpfile)
os.rmdir(tmpdir)
def read_metadata(self):
"""
Return the metadata for an object without requiring the caller to open
the object first.
This method is invoked by Swift code in POST, PUT, HEAD and DELETE path
metadata = disk_file.read_metadata()
The operations performed here is very similar to those made in open().
This is to avoid opening and closing of file (two syscalls over
network). IOW, this optimization addresses the case where the fd
returned by open() isn't going to be used i.e the file is not read (GET
or metadata recalculation)
:returns: metadata dictionary for an object
:raises DiskFileError: this implementation will raise the same
errors as the `open()` method.
"""
try:
self._metadata = read_metadata(self._data_file)
except (OSError, IOError) as err:
if err.errno in (errno.ENOENT, errno.ESTALE):
raise DiskFileNotExist
raise err
if self._metadata and self._is_object_expired(self._metadata):
raise DiskFileExpired(metadata=self._metadata)
try:
self._stat = do_stat(self._data_file)
self._is_dir = stat.S_ISDIR(self._stat.st_mode)
except (OSError, IOError) as err:
if err.errno in (errno.ENOENT, errno.ESTALE):
raise DiskFileNotExist
raise err
if not validate_object(self._metadata, self._stat):
# Metadata is stale/invalid. So open the object for reading
# to update Etag and other metadata.
with self.open():
return self.get_metadata()
else:
# Metadata is valid. Don't have to open the file.
self._filter_metadata()
return self._metadata
def read_metadata(self):
"""
Return the metadata for an object without requiring the caller to open
the object first.
This method is invoked by Swift code in POST, PUT, HEAD and DELETE path
metadata = disk_file.read_metadata()
The operations performed here is very similar to those made in open().
This is to avoid opening and closing of file (two syscalls over
network). IOW, this optimization addresses the case where the fd
returned by open() isn't going to be used i.e the file is not read (GET
or metadata recalculation)
:returns: metadata dictionary for an object
:raises DiskFileError: this implementation will raise the same
errors as the `open()` method.
"""
try:
self._metadata = read_metadata(self._data_file)
except (OSError, IOError) as err:
if err.errno in (errno.ENOENT, errno.ESTALE):
raise DiskFileNotExist
raise err
if self._metadata and self._is_object_expired(self._metadata):
raise DiskFileExpired(metadata=self._metadata)
try:
self._stat = do_stat(self._data_file)
self._is_dir = stat.S_ISDIR(self._stat.st_mode)
except (OSError, IOError) as err:
if err.errno in (errno.ENOENT, errno.ESTALE):
raise DiskFileNotExist
raise err
if not validate_object(self._metadata, self._stat):
# Metadata is stale/invalid. So open the object for reading
# to update Etag and other metadata.
with self.open():
return self.get_metadata()
else:
# Metadata is valid. Don't have to open the file.
self._filter_metadata()
return self._metadata
def make_directory(full_path, uid, gid, metadata=None):
"""
Make a directory and change the owner ship as specified, and potentially
creating the object metadata if requested.
"""
try:
do_mkdir(full_path)
except OSError as err:
if err.errno == errno.ENOENT:
# Tell the caller some directory of the parent path does not
# exist.
return False, metadata
elif err.errno == errno.EEXIST:
# Possible race, in that the caller invoked this method when it
# had previously determined the file did not exist.
#
# FIXME: When we are confident, remove this stat() call as it is
# not necessary.
try:
stats = do_stat(full_path)
except SwiftOnFileSystemOSError as serr:
# FIXME: Ideally we'd want to return an appropriate error
# message and code in the PUT Object REST API response.
raise DiskFileError("make_directory: mkdir failed"
" because path %s already exists, and"
" a subsequent stat on that same"
" path failed (%s)" %
(full_path, str(serr)))
else:
is_dir = stat.S_ISDIR(stats.st_mode)
if not is_dir:
# FIXME: Ideally we'd want to return an appropriate error
# message and code in the PUT Object REST API response.
raise AlreadyExistsAsFile("make_directory:"
" mkdir failed on path %s"
" because it already exists"
" but not as a directory" %
(full_path))
return True, metadata
elif err.errno == errno.ENOTDIR:
# FIXME: Ideally we'd want to return an appropriate error
# message and code in the PUT Object REST API response.
raise AlreadyExistsAsFile("make_directory:"
" mkdir failed because some "
"part of path %s is not in fact"
" a directory" % (full_path))
elif err.errno == errno.EIO:
# Sometimes Fuse will return an EIO error when it does not know
# how to handle an unexpected, but transient situation. It is
# possible the directory now exists, stat() it to find out after a
# short period of time.
_random_sleep()
try:
stats = do_stat(full_path)
except SwiftOnFileSystemOSError as serr:
if serr.errno == errno.ENOENT:
errmsg = "make_directory: mkdir failed on" \
" path %s (EIO), and a subsequent stat on" \
" that same path did not find the file." % (
full_path,)
else:
errmsg = "make_directory: mkdir failed on" \
" path %s (%s), and a subsequent stat on" \
" that same path failed as well (%s)" % (
full_path, str(err), str(serr))
raise DiskFileError(errmsg)
else:
if not stats:
errmsg = "make_directory: mkdir failed on" \
" path %s (EIO), and a subsequent stat on" \
" that same path did not find the file." % (
full_path,)
raise DiskFileError(errmsg)
else:
# The directory at least exists now
is_dir = stat.S_ISDIR(stats.st_mode)
if is_dir:
# Dump the stats to the log with the original exception
logging.warn("make_directory: mkdir initially"
" failed on path %s (%s) but a stat()"
" following that succeeded: %r" %
(full_path, str(err), stats))
# Assume another entity took care of the proper setup.
return True, metadata
else:
raise DiskFileError("make_directory: mkdir"
" initially failed on path %s (%s)"
" but now we see that it exists"
" but is not a directory (%r)" %
(full_path, str(err), stats))
else:
# Some other potentially rare exception occurred that does not
# currently warrant a special log entry to help diagnose.
raise DiskFileError("make_directory: mkdir failed on"
" path %s (%s)" % (full_path, str(err)))
else:
if metadata:
# We were asked to set the initial metadata for this object.
metadata_orig = get_object_metadata(full_path)
metadata_orig.update(metadata)
write_metadata(full_path, metadata_orig)
metadata = metadata_orig
# We created it, so we are reponsible for always setting the proper
# ownership.
do_chown(full_path, uid, gid)
return True, metadata
def test_do_stat_enoent(self):
res = fs.do_stat(os.path.join('/tmp', str(random.random())))
assert res is None
def make_directory(full_path, uid, gid, metadata=None):
"""
Make a directory and change the owner ship as specified, and potentially
creating the object metadata if requested.
"""
try:
do_mkdir(full_path)
except OSError as err:
if err.errno == errno.ENOENT:
# Tell the caller some directory of the parent path does not
# exist.
return False, metadata
elif err.errno == errno.EEXIST:
# Possible race, in that the caller invoked this method when it
# had previously determined the file did not exist.
#
# FIXME: When we are confident, remove this stat() call as it is
# not necessary.
try:
stats = do_stat(full_path)
except SwiftOnFileSystemOSError as serr:
# FIXME: Ideally we'd want to return an appropriate error
# message and code in the PUT Object REST API response.
raise DiskFileError("make_directory: mkdir failed"
" because path %s already exists, and"
" a subsequent stat on that same"
" path failed (%s)" % (full_path,
str(serr)))
else:
is_dir = stat.S_ISDIR(stats.st_mode)
if not is_dir:
# FIXME: Ideally we'd want to return an appropriate error
# message and code in the PUT Object REST API response.
raise AlreadyExistsAsFile("make_directory:"
" mkdir failed on path %s"
" because it already exists"
" but not as a directory"
% (full_path))
return True, metadata
elif err.errno == errno.ENOTDIR:
# FIXME: Ideally we'd want to return an appropriate error
# message and code in the PUT Object REST API response.
raise AlreadyExistsAsFile("make_directory:"
" mkdir failed because some "
"part of path %s is not in fact"
" a directory" % (full_path))
elif err.errno == errno.EIO:
# Sometimes Fuse will return an EIO error when it does not know
# how to handle an unexpected, but transient situation. It is
# possible the directory now exists, stat() it to find out after a
# short period of time.
_random_sleep()
try:
stats = do_stat(full_path)
except SwiftOnFileSystemOSError as serr:
if serr.errno == errno.ENOENT:
errmsg = "make_directory: mkdir failed on" \
" path %s (EIO), and a subsequent stat on" \
" that same path did not find the file." % (
full_path,)
else:
errmsg = "make_directory: mkdir failed on" \
" path %s (%s), and a subsequent stat on" \
" that same path failed as well (%s)" % (
full_path, str(err), str(serr))
raise DiskFileError(errmsg)
else:
if not stats:
errmsg = "make_directory: mkdir failed on" \
" path %s (EIO), and a subsequent stat on" \
" that same path did not find the file." % (
full_path,)
raise DiskFileError(errmsg)
else:
# The directory at least exists now
is_dir = stat.S_ISDIR(stats.st_mode)
if is_dir:
# Dump the stats to the log with the original exception
logging.warn("make_directory: mkdir initially"
" failed on path %s (%s) but a stat()"
" following that succeeded: %r" %
(full_path, str(err), stats))
# Assume another entity took care of the proper setup.
return True, metadata
else:
raise DiskFileError("make_directory: mkdir"
" initially failed on path %s (%s)"
" but now we see that it exists"
" but is not a directory (%r)" %
(full_path, str(err), stats))
else:
# Some other potentially rare exception occurred that does not
# currently warrant a special log entry to help diagnose.
raise DiskFileError("make_directory: mkdir failed on"
" path %s (%s)" % (full_path, str(err)))
else:
if metadata:
# We were asked to set the initial metadata for this object.
metadata_orig = get_object_metadata(full_path)
metadata_orig.update(metadata)
write_metadata(full_path, metadata_orig)
metadata = metadata_orig
# We created it, so we are reponsible for always setting the proper
# ownership.
do_chown(full_path, uid, gid)
return True, metadata
def _finalize_put(self, metadata):
# Write out metadata before fsync() to ensure it is also forced to
# disk.
write_metadata(self._fd, metadata)
# We call fsync() before calling drop_cache() to lower the
# amount of redundant work the drop cache code will perform on
# the pages (now that after fsync the pages will be all
# clean).
do_fsync(self._fd)
# From the Department of the Redundancy Department, make sure
# we call drop_cache() after fsync() to avoid redundant work
# (pages all clean).
do_fadvise64(self._fd, self._last_sync, self._upload_size)
# At this point we know that the object's full directory path
# exists, so we can just rename it directly without using Swift's
# swift.common.utils.renamer(), which makes the directory path and
# adds extra stat() calls.
df = self._disk_file
attempts = 1
while True:
try:
do_rename(self._tmppath, df._data_file)
except OSError as err:
if err.errno in (errno.ENOENT, errno.EIO) \
and attempts < MAX_RENAME_ATTEMPTS:
# FIXME: Why either of these two error conditions is
# happening is unknown at this point. This might be a
# FUSE issue of some sort or a possible race
# condition. So let's sleep on it, and double check
# the environment after a good nap.
_random_sleep()
# Tease out why this error occurred. The man page for
# rename reads:
# "The link named by tmppath does not exist; or, a
# directory component in data_file does not exist;
# or, tmppath or data_file is an empty string."
assert len(self._tmppath) > 0 and len(df._data_file) > 0
tpstats = do_stat(self._tmppath)
tfstats = do_fstat(self._fd)
assert tfstats
if not tpstats or tfstats.st_ino != tpstats.st_ino:
# Temporary file name conflict
raise DiskFileError(
'DiskFile.put(): temporary file, %s, was'
' already renamed (targeted for %s)' % (
self._tmppath, df._data_file))
else:
# Data file target name now has a bad path!
dfstats = do_stat(df._put_datadir)
if not dfstats:
raise DiskFileError(
'DiskFile.put(): path to object, %s, no'
' longer exists (targeted for %s)' % (
df._put_datadir, df._data_file))
else:
is_dir = stat.S_ISDIR(dfstats.st_mode)
if not is_dir:
raise DiskFileError(
'DiskFile.put(): path to object, %s,'
' no longer a directory (targeted for'
' %s)' % (self._put_datadir,
df._data_file))
else:
# Let's retry since everything looks okay
logging.warn(
"DiskFile.put(): rename('%s','%s')"
" initially failed (%s) but a"
" stat('%s') following that succeeded:"
" %r" % (
self._tmppath, df._data_file, str(err),
df._put_datadir, dfstats))
attempts += 1
continue
else:
raise SwiftOnFileSystemOSError(
err.errno, "%s, rename('%s', '%s')" % (
err.strerror, self._tmppath, df._data_file))
else:
# Success!
break
# Close here so the calling context does not have to perform this
# in a thread.
self.close()
def _finalize_put(self, metadata):
# Write out metadata before fsync() to ensure it is also forced to
# disk.
write_metadata(self._fd, metadata)
# We call fsync() before calling drop_cache() to lower the
# amount of redundant work the drop cache code will perform on
# the pages (now that after fsync the pages will be all
# clean).
do_fsync(self._fd)
# From the Department of the Redundancy Department, make sure
# we call drop_cache() after fsync() to avoid redundant work
# (pages all clean).
do_fadvise64(self._fd, self._last_sync, self._upload_size)
# At this point we know that the object's full directory path
# exists, so we can just rename it directly without using Swift's
# swift.common.utils.renamer(), which makes the directory path and
# adds extra stat() calls.
df = self._disk_file
attempts = 1
while True:
try:
do_rename(self._tmppath, df._data_file)
except OSError as err:
if err.errno in (errno.ENOENT, errno.EIO) \
and attempts < MAX_RENAME_ATTEMPTS:
# FIXME: Why either of these two error conditions is
# happening is unknown at this point. This might be a
# FUSE issue of some sort or a possible race
# condition. So let's sleep on it, and double check
# the environment after a good nap.
_random_sleep()
# Tease out why this error occurred. The man page for
# rename reads:
# "The link named by tmppath does not exist; or, a
# directory component in data_file does not exist;
# or, tmppath or data_file is an empty string."
assert len(self._tmppath) > 0 and len(df._data_file) > 0
tpstats = do_stat(self._tmppath)
tfstats = do_fstat(self._fd)
assert tfstats
if not tpstats or tfstats.st_ino != tpstats.st_ino:
# Temporary file name conflict
raise DiskFileError(
'DiskFile.put(): temporary file, %s, was'
' already renamed (targeted for %s)' %
(self._tmppath, df._data_file))
else:
# Data file target name now has a bad path!
dfstats = do_stat(df._put_datadir)
if not dfstats:
raise DiskFileError(
'DiskFile.put(): path to object, %s, no'
' longer exists (targeted for %s)' %
(df._put_datadir, df._data_file))
else:
is_dir = stat.S_ISDIR(dfstats.st_mode)
if not is_dir:
raise DiskFileError(
'DiskFile.put(): path to object, %s,'
' no longer a directory (targeted for'
' %s)' %
(self._put_datadir, df._data_file))
else:
# Let's retry since everything looks okay
logging.warn(
"DiskFile.put(): rename('%s','%s')"
" initially failed (%s) but a"
" stat('%s') following that succeeded:"
" %r" %
(self._tmppath, df._data_file, str(err),
df._put_datadir, dfstats))
attempts += 1
continue
else:
raise SwiftOnFileSystemOSError(
err.errno, "%s, rename('%s', '%s')" %
(err.strerror, self._tmppath, df._data_file))
else:
# Success!
break
# Close here so the calling context does not have to perform this
# in a thread.
self.close()
def test_do_stat_enoent(self):
res = fs.do_stat(os.path.join('/tmp', str(random.random())))
assert res is None
