Add streaming download support

So we don't run out of memory when fetching blobs, use cherrypy's
streaming support.

Change-Id: I6c05240a81f0b6b2dd0992e42d097883ca4fdc2e
This commit is contained in:
James E. Blair 2019-10-03 10:27:08 -07:00
parent 6d2894de8b
commit 61af2796b2
5 changed files with 56 additions and 6 deletions

View File

@ -57,6 +57,17 @@ class FilesystemDriver(storageutils.StorageDriver):
with open(path, 'rb') as f:
return f.read()
def stream_object(self, path):
path = os.path.join(self.root, path)
if not os.path.exists(path):
return None
with open(path, 'rb') as f:
while True:
chunk = f.read(4096)
if not chunk:
return
yield chunk
def delete_object(self, path):
path = os.path.join(self.root, path)
if os.path.exists(path):

View File

@ -106,20 +106,29 @@ class RegistryAPI:
@cherrypy.expose
@cherrypy.config(**{'tools.auth_basic.checkpassword': require_read})
def get_blob(self, repository, digest):
def head_blob(self, repository, digest):
namespace = self.get_namespace()
method = cherrypy.request.method
self.log.info('%s blob %s %s', method, repository, digest)
self.log.info('Head blob %s %s', repository, digest)
size = self.storage.blob_size(namespace, digest)
if size is None:
return self.not_found()
res = cherrypy.response
res.headers['Docker-Content-Digest'] = digest
if method != 'HEAD':
data = self.storage.get_blob(namespace, digest)
return data
return {}
@cherrypy.expose
@cherrypy.config(**{'tools.auth_basic.checkpassword': require_read,
'response.stream': True})
def get_blob(self, repository, digest):
namespace = self.get_namespace()
self.log.info('Get blob %s %s', repository, digest)
size = self.storage.blob_size(namespace, digest)
if size is None:
return self.not_found()
res = cherrypy.response
res.headers['Docker-Content-Digest'] = digest
return self.storage.stream_blob(namespace, digest)
@cherrypy.expose
@cherrypy.config(**{'tools.auth_basic.checkpassword': require_write})
def start_upload(self, repository, digest=None):
@ -260,6 +269,10 @@ class RegistryServer:
conditions=dict(method=['GET']),
controller=api, action='get_manifest')
route_map.connect('api', '/v2/{repository:.*}/blobs/{digest}',
conditions=dict(method=['HEAD']),
controller=api, action='head_blob')
route_map.connect('api', '/v2/{repository:.*}/blobs/{digest}',
conditions=dict(method=['GET']),
controller=api, action='get_blob')
conf = {

View File

@ -139,6 +139,11 @@ class Storage:
path = os.path.join(namespace, 'blobs', digest, 'data')
return self.backend.get_object(path)
def stream_blob(self, namespace, digest):
path = os.path.join(namespace, 'blobs',
self._path_from_digest(digest), 'data')
return self.backend.stream_object(path)
def start_upload(self, namespace):
"""Start an upload.

View File

@ -91,6 +91,19 @@ class StorageDriver(metaclass=ABCMeta):
"""
pass
@abstractmethod
def stream_object(self, path):
"""Retrieve an object, streaming.
Return a generator with the content of the object at `path`.
:arg str path: The object path.
:returns: The contents of the object.
:rtype: generator of bytearray
"""
pass
@abstractmethod
def delete_object(self, path):
"""Delete an object.

View File

@ -120,6 +120,14 @@ class SwiftDriver(storageutils.StorageDriver):
return None
return ret.content
def stream_object(self, path):
try:
ret = retry_function(
lambda: self.conn.session.get(self.get_url(path), stream=True))
except keystoneauth1.exceptions.http.NotFound:
return None
return ret.iter_content(chunk_size=4096)
def delete_object(self, path):
retry_function(
lambda: self.conn.session.delete(