def process_archive_retrieval_job(job,chunk_size,output_path,friendly_name=False):
global chunk_count
filepos_limit = job.archive_size_in_bytes - 1
pad_length = len(str(job.archive_size_in_bytes // chunk_size)) + 1
current_pos = 0
job_archive_hash = job.archive_sha256_tree_hash
chunk_count = 0
failed_parts = {}
running_treehash = TreeHash(algo=hashlib.sha256)
try:
while current_pos < filepos_limit:
chunk_count += 1
end_pos = current_pos + (chunk_size - 1)
if end_pos > filepos_limit:
end_pos = filepos_limit
if not process_archive_retrieval_range(job, output_path, current_pos, end_pos, friendly_name, running_treehash):
failed_parts["part_" + str(chunk_count).zfill(pad_length)]=[current_pos, end_pos]
current_pos = end_pos + 1
if (not failed_parts) and (running_treehash.hexdigest() != job.archive_sha256_tree_hash):
failed_parts = None
failed_parts["all"] = True
except:
failed_parts["all"] = True
raise
return failed_parts
def main():
global Debug
# Parse command line options
parser = argparse.ArgumentParser()
parser.add_argument("--file", help="File Name", required=True)
parser.add_argument("--rangesize", help="Size of the range", required=True)
parser.add_argument("--chunksize", help="Size of the chunks to use for checksumming.", default=DEFAULT_CHUNK_SIZE)
parser.add_argument("--debug", help="Print Debug messages", action="store_true")
args = parser.parse_args()
Debug = args.debug
treehash = TreeHash(algo=hashlib.sha256)
statinfo = os.stat(args.file)
end_pos = statinfo.st_size
range = int(args.rangesize)
chunksize = int(args.chunksize)
current_pos = 0
while current_pos < end_pos:
current_end_pos = current_pos + range
if current_end_pos > end_pos:
current_end_pos = end_pos
running_treehash_on_file_range(treehash, args.file, current_pos, current_end_pos, chunksize)
current_pos = current_end_pos
print("TreeHash for " + args.file + " is " + treehash.hexdigest())
def test_update(self):
treehash = TreeHash()
treehash.update(TEST_DATA)
self.assertEqual(
hashlib.sha256(TEST_DATA).hexdigest(),
treehash.hexdigest()
)
def test_update(self):
tree_hash = TreeHash(TEST_CHUNK)
tree_hash.update(TEST_DATA, TEST_INDEX)
self.assertEqual(
hashlib.sha256(TEST_DATA).hexdigest(),
tree_hash.hexdigest()
)
def test_md5(self):
tree_hash = TreeHash(TEST_CHUNK, algo=hashlib.md5)
tree_hash.update(TEST_DATA, TEST_INDEX)
self.assertEqual(
hashlib.md5(TEST_DATA).hexdigest(),
tree_hash.hexdigest()
)
def treehash_on_file_range(treehash, filename, start, end, hash_chunk_size=DEFAULT_HASH_CHUNK_SIZE):
infile = open(filename, "rb")
infile.seek(start)
if Debug:
print("Treehash: Start: " + str(start) + ", End: " + str(end))
treehash_local = TreeHash(algo=hashlib.sha256)
current_pos = start
end += 1
while current_pos < end:
read_size = end - current_pos
if read_size > hash_chunk_size:
read_size = hash_chunk_size
chunk = infile.read(read_size)
if treehash:
treehash.update(chunk)
treehash_local.update(chunk)
current_pos += read_size
infile.close()
return treehash_local.hexdigest()
def hash_file(fname):
treehash = TreeHash(algo=hashlib.sha256)
with open(fname, 'rb') as f:
while True:
buf = f.read(1024 * 1024)
if not buf:
break
treehash.update(buf)
return treehash.hexdigest()
def test_tree(self):
hashlib_result = hashlib.sha256(
hashlib.sha256(TEST_DATA).digest() +
hashlib.sha256(TEST_DATA).digest()
).hexdigest()
tree_hash = TreeHash(2, block_size=1)
tree_hash.update(TEST_DATA, 0)
tree_hash.update(TEST_DATA, 1)
self.assertEqual(hashlib_result,
tree_hash.hexdigest())
def process_archive_retrieval_job(job,chunk_size,output_path,friendly_name=False):
global chunk_count
filepos_limit = job.archive_size_in_bytes - 1
current_pos = 0
job_archive_hash = job.archive_sha256_tree_hash
chunk_count = 0
archive_file_name = output_path + "/" + job.id + ".archive"
archive_file = open(archive_file_name, "wb")
treehash = TreeHash(algo=hashlib.sha256)
while current_pos < filepos_limit:
end_pos = current_pos + (chunk_size - 1)
if end_pos > filepos_limit:
end_pos = filepos_limit
range_string = "bytes=" + str(current_pos) + "-" + str(end_pos)
response = job.get_output(
range=range_string
)
if Debug:
print("process_archive_retrieval_job: job.get_output() response: " + str(response))
if HTTP_SUCCESS_LOW <= response['status'] <= HTTP_SUCCESS_HIGH:
chunk_count += 1
if Debug:
#print("Writing chunk " + str(chunk_count) + " " + range_string + " Checksum: " + response['checksum'] + " ContentRange: " + response['contentRange'] + " AcceptRanges: " + response['acceptRanges'] + " ContentType: " + response['contentType'] + " ArchiveDescription: " + response['archiveDescription'])
print("Writing chunk " + str(chunk_count) + " " + range_string + " Checksum: " + response['checksum'])
#archive_file.write(response['body'].read())
chunk_bytes=response['body'].read()
archive_file.write(chunk_bytes)
if Debug:
chunk_file = open(archive_file_name + ".chunk." + str(chunk_count), "wb")
chunk_file.write(chunk_bytes)
chunk_file.close
section_hash = sha256_on_file_range(archive_file_name, current_pos, end_pos)
running_treehash_on_file_range(treehash, archive_file_name, current_pos, end_pos)
if Debug:
print("Local checksum of chunk " + str(chunk_count) + ": " + section_hash)
print("Current running treehash is " + treehash.hexdigest())
current_pos = end_pos + 1
else:
print("Response unsuccessful. Retrying")
archive_file.close
def run(self):
client = self.get_boto_client()
logging.info("Initiating job to upload")
upload_job = client.initiate_multipart_upload(
vaultName=self.vault,
archiveDescription=self.file_name,
partSize=str(UPLOAD_PART_SIZE))
upload_id = upload_job['uploadId']
treehash = TreeHash(block_size=1024**2)
cur_file = open(self.file_location, 'rb')
i = 0
for i in range(self.numof_parts - 1):
self.cur_part += 1
self.update_task()
data = cur_file.read(UPLOAD_PART_SIZE)
treehash.update(data)
cur_range = 'bytes %d-%d/*' % (i * UPLOAD_PART_SIZE,
(i + 1) * UPLOAD_PART_SIZE - 1)
client.upload_multipart_part(vaultName=self.vault,
uploadId=upload_id,
range=cur_range,
body=data)
self.cur_part += 1
self.update_task()
data = cur_file.read(UPLOAD_PART_SIZE)
treehash.update(data)
cur_range = 'bytes %d-%d/*' % (i * UPLOAD_PART_SIZE,
self.file_size - 1)
client.upload_multipart_part(vaultName=self.vault,
uploadId=upload_id,
range=cur_range,
body=data)
cur_file.close()
hash_res = treehash.hexdigest()
client.complete_multipart_upload(vaultName=self.vault,
uploadId=upload_id,
archiveSize=str(self.file_size),
checksum=hash_res)
def main():
global Debug
# Parse command line options
parser = argparse.ArgumentParser()
parser.add_argument("--file", help="file to use for hash tests", required=True)
parser.add_argument("--outputpath", help="Path to store output", default=DEFAULT_OUTPUT_PATH)
parser.add_argument("--chunksize", help="Size of the chunks to use for download. Only valid of the job is ArchiveRetrieval.", default=DEFAULT_CHUNK_SIZE)
parser.add_argument("--debug", help="Print Debug messages", action="store_true")
args = parser.parse_args()
Debug = args.debug
ProcessJob_old.set_debug(Debug)
chunksize = int(args.chunksize)
statinfo = os.stat(args.file)
if not ProcessJob_old.is_power_of_2(chunksize):
print("Chunksize " + str(chunksize) + " is not a power of two. The next closest power of two is " + str(ProcessJob_old.next_power_of_2(chunksize)))
print("Exiting.")
sys.exit(1)
if chunksize > statinfo.st_size:
chunksize = statinfo.st_size
current_pos = 0
chunk_count = 0
treehash = TreeHash(algo=hashlib.sha256)
while current_pos < statinfo.st_size:
chunk_count += 1
end_pos = current_pos + chunksize - 1
if end_pos > statinfo.st_size:
end_pos = statinfo.st_size
if Debug:
print("Processing chunk " + str(chunk_count) + " range " + str(current_pos) + " to " + str(end_pos) )
section_hash = ProcessJob_old.sha256_on_file_range(args.file, current_pos, end_pos)
ProcessJob_old.running_treehash_on_file_range(treehash, args.file, current_pos, end_pos)
print("Range: " + str(current_pos) + " to " + str(end_pos))
print("Local checksum of chunk " + str(chunk_count) + ": " + section_hash)
print("Current running treehash is " + treehash.hexdigest())
current_pos = end_pos + 1
def sha256_on_file_range(filename, start, end, hash_chunk_size=DEFAULT_HASH_CHUNK_SIZE):
sha256 = hashlib.sha256()
infile = open(filename, "rb")
treehash = TreeHash(algo=hashlib.sha256)
infile.seek(start)
if Debug:
print("Running Hash: Start: " + str(start) + ", End: " + str(end))
current_pos = start
end += 1
while current_pos < end:
read_size = end - current_pos
if read_size > hash_chunk_size:
read_size = hash_chunk_size
chunk = infile.read(read_size)
sha256.update(chunk)
treehash.update(chunk)
current_pos += read_size
infile.close()
if Debug:
print("Running hash for this section (" + str(start) + " to " + str(end) + ") is " + sha256.hexdigest())
print("Tree hash for this section (" + str(start) + " to " + str(end) + ") is " + treehash.hexdigest())
return sha256.hexdigest()
def upload(self, client):
if (self._inventory_entry.get_state() == FileState.IN_PROGRESS):
self._upload_id = self._inventory_entry.get_upload_id()
else:
tmp_upload = client.initiate_multipart_upload(
vaultName=self._vaultName,
archiveDescription=self._inventory_entry.get_fileName(),
partSize=str(self._partSize))
self._upload_id = tmp_upload['uploadId']
if self._partSize < self._fileSizeBytes:
self._inventory_entry.set_state_from_upload(
self, FileState.IN_PROGRESS)
partBegin = self._partNumUploading * self._partSize
data = b""
with open(self._inventory_entry.get_filePath(), "rb") as f:
if partBegin:
data = f.read(partBegin)
treehash = TreeHash(data=data, block_size=self._partSize)
while partBegin < self._fileSizeBytes:
partEnd = partBegin + self._partSize - 1
if partEnd > self._fileSizeBytes:
partEnd = self._fileSizeBytes - 1
part = f.read(self._partSize)
treehash.update(part)
if not self._startTime:
self._startTime = time.time()
self._upload_part(client, part, partBegin, partEnd)
partBegin = partEnd + 1
self._partNumUploading += 1
if partEnd < self._fileSizeBytes:
self._inventory_entry.set_state_from_upload(
self, FileState.IN_PROGRESS)
completed_treehash = treehash.hexdigest()
response = client.complete_multipart_upload(
vaultName=self._vaultName,
uploadId=self._upload_id,
archiveSize=str(self._fileSizeBytes),
checksum=completed_treehash)
self._endTime = time.time()
cli.cli_progress(self._inventory_entry.get_fileName(),
self.formattedFileSize(),
self.formattedPartSize(),
self._startTime,
self._fileSizeBytes-1,
self._fileSizeBytes-1)
# Sanity check that's probably unnecessary.
if treehash.hexdigest() != response['checksum']:
raise Exception('checksum mismatch')
self._checksum = response['checksum']
self._http_status = response['ResponseMetadata']['HTTPStatusCode']
self._archive_id = response['archiveId']
self._upload_location = response['location']
# cli.pp(json.dumps(self, default=lambda o: o.__dict__))
self._inventory_entry.set_state_from_upload(self, FileState.UPLOADED)
def handle(self, *args, **options):
part_size = 8388608
print 'Contacting Amazon AWS...'
glacier = boto3.client('glacier')
multipart_upload = glacier.initiate_multipart_upload(
vaultName=settings.GLACIER_VAULT_NAME, partSize=str(part_size))
print 'Connected to Glacier Vault "' + settings.GLACIER_VAULT_NAME + '"'
upload_id = multipart_upload['uploadId']
treehash_archive = TreeHash()
db = influxdb.InfluxDBClient(settings.INFLUXDB_URI, 8086, 'root',
'root', 'seads')
archive_size = 0
for device in Device.objects.all():
start = datetime.fromtimestamp(0)
end = datetime.now() - timedelta(days=31 *
device.data_retention_policy)
start = (datetime.now() - start).total_seconds()
start = 0
end = int((datetime.now() - end).total_seconds())
end = time.time() - end
print 'Trying ' + str(device) + '...'
print 'Data Retention Policy: ' + str(
device.data_retention_policy) + ' Months'
series = 'device.' + str(device.serial)
try:
query = 'select * from ' + series + ' where time > ' + str(
start) + 's and time < ' + str(end) + 's'
points = db.query(query)
except:
print 'No data found for ' + series + '. Skipping.'
continue
print "Uploading " + series + "..."
print "Querying from " + str(datetime.fromtimestamp(
int(start))) + " to " + str(datetime.fromtimestamp(int(end)))
# store points in temporary file, break into 8MB parts
with open('/tmp/temp_archive', 'wb') as f:
f.write(json.dumps(points))
bytes_read = 0
bytes_sent = 0
with open('/tmp/temp_archive', 'rb') as f:
treehash_part = TreeHash()
part = f.read(part_size)
treehash_part.update(part)
bytes_read += len(part)
while part:
response = glacier.upload_multipart_part(
vaultName=settings.GLACIER_VAULT_NAME,
uploadId=upload_id,
range='bytes ' + str(bytes_sent) + '-' +
str(bytes_read - 1) + '/*',
body=part,
checksum=treehash_part.hexdigest())
bytes_sent += len(part)
part = f.read(part_size)
treehash_part.update(part)
bytes_read += len(part)
archive_size += 1
print "Successfully uploaded " + str(
bytes_sent) + " bytes to Glacier"
print "Deleting points from database..."
# drop from fanout series as well
series = db.query('list series')[0]['points']
rg = re.compile('device.' + str(device.serial))
for s in series:
if rg.search(s[1]):
db.query('delete from ' + s[1] + ' where time > ' +
str(start) + 's and time < ' + str(end) + 's')
print "[DONE]"
try:
with open('/tmp/temp_archive', 'rb') as f:
treehash_archive.update(f.read())
response = glacier.complete_multipart_upload(
vaultName=settings.GLACIER_VAULT_NAME,
uploadId=upload_id,
archiveSize=str(archive_size),
checksum=treehash_archive.hexdigest())
with open(settings.STATIC_PATH + 'archive_ids.log', 'a') as f:
line = {
'archiveId': response['archiveId'],
'timeEnd': str(end)
}
f.write(json.dumps(line))
f.write(';')
os.remove('/tmp/temp_archive')
print "Archival Successful"
except:
print "No data to archive. Exiting."
def test_update(self):
treehash = TreeHash()
treehash.update(TEST_DATA)
self.assertEqual(
hashlib.sha256(TEST_DATA).hexdigest(), treehash.hexdigest())
potential_zip = entry.name + '.zip'
if entry.is_dir() and re.match(r'\d+[_\-]\d+[_\-]\d+', entry.name) and potential_zip not in decoded_archived:
logger.info(entry.name + ' matches the pattern \d+[_\-]\d+[_\-]\d+ and is not currently in Glacier as a zip file.')
not_archived.append({'Zipfile' : potential_zip, 'Path' : (picture_path + '\\' + entry.name)})
logger.info(potential_zip + ' added to list of archives that need to be created uploaded.')
for n in not_archived:
logger.info('Processing ' + str(n))
with zipfile.ZipFile(n['Zipfile'], 'w', allowZip64=True) as zf:
logger.info('Writing zipfile ' + n['Path'])
writeZipfile(n['Path'], zf)
with open(n['Zipfile'], 'rb') as file:
expected_fullhash = TreeHash()
expected_fullhash.update(file.read())
expected_fullhash_value = expected_fullhash.hexdigest()
logger.debug('File: ' + n['Zipfile'] + '. Expected full file hash is ' + str(expected_fullhash_value))
file.seek(0)
#Check for multipart upload in progress
logger.debug('File: ' + n['Zipfile'] + '. Entering getExistingUploads, passing glacier object and target_vault_name: ' + target_vault_name)
existing_uploads = getExistingUploads(glacier, target_vault_name)
start_byte = 0
file_size = 0
chunksize = 8388608
fullhash = TreeHash()
description = '{"Path":"' + n['Zipfile'] + ', "ExpectedTreeHash":"' + expected_fullhash_value + '"}'
logger.info('File: ' + n['Zipfile'] + '. Starting file upload process. File description: ' + description)
logger.info('File: ' + n['Zipfile'] + '. Chunksize: ' + str(chunksize))
if opt['Verbose']: print("## upload id " + upid)
if upid[0] != '-':
break
(rp, scnt, ft, awsids) = (0, 0, open(fn, 'rb'),
awstmp + ' --upload-id {}'.format(upid))
for x in range(0, fs, csize):
rs = ft.read(csize)
(fl, fp) = (len(rs), opt['TempFile'].format(fcnt, scnt))
with open(fp, 'wb') as fw:
fw.write(rs)
mp = TreeHash()
mp.update(rs)
(s,r)=cmd(awsids.format('upload-multipart-part')+' --body \'{}\''.format(fp)+\
' --range \"bytes {}-{}/*\" --checksum \"{}\"'.format(rp,rp+fl-1,mp.hexdigest()))
if s != 0: errorexit(r)
if opt['Verbose']:
print('## done part {:2}, {:6.2f} GB ({:12} b)'.format(
scnt, fl / gbyte, fl),
flush=True)
elif scnt == 0:
print('## done part {:2}'.format(scnt), end='', flush=True)
else:
print(' {:2}'.format(scnt), end='', flush=True)
(rp, scnt) = (rp + fl, scnt + 1)
os.remove(fp)
ft.close()
print('')
(s, r) = cmd(
def main():
# Parse command line options
parser = argparse.ArgumentParser()
parser.add_argument("--file", help="File to upload", required=True)
parser.add_argument("--size", help="Chunk size in bytes", required=True)
parser.add_argument("--account", help="Account ID", default="-")
parser.add_argument("--vault", help="Vault Name", required=True)
args = parser.parse_args()
in_file = open(args.file, "rb")
in_file_size = os.path.getsize(args.file)
#in_file_sha256 = hashlib.sha256(open(args.file, "rb").read()).hexdigest()
treehash = TreeHash(algo=hashlib.sha256)
with open(args.file, "rb") as treehash_input:
while True:
data = treehash_input.read(hash_chunks)
treehash.update(data)
if len(data) < hash_chunks:
break
in_file_tree_sha256 = treehash.hexdigest()
chunk_size = int(args.size)
if chunk_size < min_chunk_size:
print("Supplied chunk size (" + args.size + ") is less than minimum. Setting chunk size to " + str(min_chunk_size))
chunk_size = min_chunk_size
glacier = boto3.resource('glacier')
client = boto3.client('glacier')
#
multiupload_request = client.initiate_multipart_upload(
vaultName=args.vault,
archiveDescription=args.file,
partSize=str(chunk_size)
)
multipart_upload = glacier.MultipartUpload(args.account, args.vault, multiupload_request['uploadId'])
print("MultiUpload ID: " + multiupload_request['uploadId'])
print("Size: " + str(in_file_size))
#print("Hash: " + in_file_sha256)
print("Tree Hash: " + in_file_tree_sha256)
position = 0
chunk = in_file.read(chunk_size)
while chunk:
print("Length: " + str(len(chunk)))
print("Current range: bytes " + str(position) + "-" + str(position + len(chunk) - 1) + "/*")
response = multipart_upload.upload_part(
range="bytes " + str(position) + "-" + str(position + len(chunk) - 1) + "/*",
body=chunk
)
print("Uploaded Checksum: " + response['checksum'])
position += len(chunk)
chunk = in_file.read(chunk_size)
response = multipart_upload.complete(
archiveSize=str(in_file_size),
checksum=in_file_tree_sha256
)
print("Upload Complete.")
print("Location: " + response['location'])
print("Checksum: " + response['checksum'])
print("Archive ID: " + response['archiveId'])
print("Account ID: " + multipart_upload.account_id)
print("Vault name: " + multipart_upload.vault_name)
print("ID: " + multipart_upload.id)