Bazaar

Merge lp:~gz/bzr/transport_post_connect_hook into lp:bzr

  1. transport_post_connect_hook
  2. Merge into bzr.dev
Proposed by Martin Packman
Status: Superseded
Proposed branch: lp:~gz/bzr/transport_post_connect_hook
Merge into: lp:bzr
Prerequisite: lp:~spiv/bzr/hooks-refactoring
Diff against target: 280 lines (+62/-108)
6 files modified
bzrlib/hooks.py (+1/-0)
bzrlib/tests/__init__.py (+14/-9)
bzrlib/tests/test_transport.py (+23/-0)
bzrlib/tests/transport_util.py (+4/-98)
bzrlib/transport/__init__.py (+18/-1)
bzrlib/transport/remote.py (+2/-0)
To merge this branch: bzr merge lp:~gz/bzr/transport_post_connect_hook
Reviewer Review Type Date Requested Status
Vincent Ladeuil Approve
Review via email: mp+38431@code.launchpad.net

This proposal has been superseded by a proposal from 2011-12-14.

Description of the change

Creates a new post_connect hook for transports, and uses it in the testing framework to ensure transports are disconnected when the test finishes so we don't leak resources. This replaces the current hack that overrides get_transport which was incomplete and causing other problems. See the mailing list thread for more:

<https://lists.ubuntu.com/archives/bazaar/2010q4/070386.html>

I'm not really sure how ready this is, but putting up an mp seems as good a way of getting feedback as bugging people on IRC.

The hook is hurled together based on lp:~parthm/bzr/173274-export-hooks if there are opinions on how it should be written differently I'd like to know.

Todo:
  News. I'm leaving this till landing actually happens to lessen pain of flux.
  Other documentation changes?
  Some tests that don't suck. Suggestions welcome.

To post a comment you must log in.
Revision history for this message
Vincent Ladeuil (vila) wrote :

129 + for hook in self.hooks["post_connect"]:
130 + # GZ 2010-10-14: Should the hook be passed the new connection and
131 + # credentials too or does opaque really mean that?
132 + hook(self)

The hook already receives 'self' so it can access the connection/credentials if needed.
But they are specific to each transport class...

The tests are ok. We know the hook is heavily exercised or we get leaks anyway.

If you really really want to add tests you can check what happens when several transports share a connection, but even that sounds overkill.

Write the NEWS entry, we'll see what is available when your patch lands. Don't forget the hooks-help.txt file.

I'll ping people on the pre-requisites.

review: Approve
Revision history for this message
Martin Packman (gz) wrote :

So, very nearly this exact hook exists in bzrlib.tests.transport_util but in a hackier tests only fashion. Basically that whole file can now go, but Vincent pointed out it has some sftp-or-ftp logic depending on if paramiko is installed that may as well stay put for the moment.

Revision history for this message
Martin Packman (gz) wrote :

Just having a hook turned out to be insufficient, and results in worse hangs than the old get_transport hack as can be seen from the nearly four hour babune runtime last night:

<http://babune.ladeuil.net:24842/job/selftest-subset-windows/11/>

The problem is that RemoteTransport classes don't use _set_connection and generally behave rather differently. Throwing in a hook point in __init__ when a new medium is built gets us to a leak-free 40 minute runtime:

<http://babune.ladeuil.net:24842/job/selftest-subset-windows/13/>

However the semantics are not quite correct there. Even if we don't have to worry about remote reconnections (do we?) the post_connect hook is happening before the real connection, and potentially twice in the RemoteHTTPTransport case. Combined with the existing confusion over how many times disconnect gets called, it suggests this hook probably isn't sane enough to be generally useful. We do want the leak problem fixed asap though, so if anyone has any clever ideas...

Revision history for this message
Vincent Ladeuil (vila) wrote :
Download full text (4.0 KiB)

>>>>> Martin [gz] <email address hidden> writes:

    > Just having a hook turned out to be insufficient, and results in
    > worse hangs than the old get_transport hack as can be seen from
    > the nearly four hour babune runtime last night:
    > <http://babune.ladeuil.net:24842/job/selftest-subset-windows/11/>

    > The problem is that RemoteTransport classes don't use
    > _set_connection and generally behave rather differently. Throwing
    > in a hook point in __init__ when a new medium is built gets us to
    > a leak-free 40 minute runtime:

    > <http://babune.ladeuil.net:24842/job/selftest-subset-windows/13/>

Ok, so this validates the approach of calling disconnect for all
transports that have been able to connect to their server and ensures
all code paths are covered. As such, I'm tempted to accept the patch as
is waiting for a better solution in a followon.

From a design point of view, I think this outlines the divergence
between the smart transport and the others and came from the time where
we implemented connection sharing.

The smart transport has a medium object that implements
_ensure_connection() and disconnect() whereas the other transports use a
_SharedConnection object which is used a data container by the
transport, but the transport object implements connect_xxx() and
disconnect() while calling _set_connection() when the connection is
established (including reconnections).

So while the post_connect hook can be implemented for the transports
objects, it can't be properly implemented for smart transports where the
transport object is not available at the medium level when the
connection occur (in theory we could pass it done but I'm worried about
ref cycles there).

This hints as connection hook instead of a transport hook but this
doesn't play well with the transport objects who consider the connection
as an opaque attribute so far.

    > However the semantics are not quite correct there. Even if we
    > don't have to worry about remote reconnections (do we?) the
    > post_connect hook is happening before the real connection, and
    > potentially twice in the RemoteHTTPTransport case. Combined with
    > the existing confusion over how many times disconnect gets called,

I think the confusion comes the fact that disconnect() should be
implemented at the transport level since it is defined as closing the
connection even if there are other transports sharing this connection
(as opposed to closing the connection when the *last* transport using
the connection requires it).

In the test suite, all transports sharing a connection calls
disconnect() to ensure we don't get leaks but that's the expected
behaviour.

    > it suggests this hook probably isn't sane enough to be generally
    > useful.

Indeed, it lies for the smart transports as it's called *before* the
connection occurs.

    > We do want the leak problem fixed asap though, so if anyone has
    > any clever ideas...

One alternative would be to add a 'created' hook for transports which,
for the tests, will also call disconnect(). From your hack_transport
branch we know that calling disconnect() for all created transports is
enough to fix the leaks ev...

Read more...

Preview Diff

[H/L] Next/Prev Comment, [J/K] Next/Prev File, [N/P] Next/Prev Hunk
1=== modified file 'bzrlib/hooks.py'
2--- bzrlib/hooks.py 2010-09-28 07:34:29 +0000
3+++ bzrlib/hooks.py 2010-10-15 14:01:50 +0000
4@@ -80,6 +80,7 @@
5 ('bzrlib.smart.client', '_SmartClient.hooks', 'SmartClientHooks'),
6 ('bzrlib.smart.server', 'SmartTCPServer.hooks', 'SmartServerHooks'),
7 ('bzrlib.status', 'hooks', 'StatusHooks'),
8+ ('bzrlib.transport', 'Transport.hooks', 'TransportHooks'),
9 ('bzrlib.version_info_formats.format_rio', 'RioVersionInfoBuilder.hooks',
10 'RioVersionInfoBuilderHooks'),
11 ('bzrlib.merge_directive', 'BaseMergeDirective.hooks',
12
13=== modified file 'bzrlib/tests/__init__.py'
14--- bzrlib/tests/__init__.py 2010-10-15 11:20:45 +0000
15+++ bzrlib/tests/__init__.py 2010-10-15 14:01:50 +0000
16@@ -2444,15 +2444,20 @@
17
18 def setUp(self):
19 super(TestCaseWithMemoryTransport, self).setUp()
20- # Ensure that ConnectedTransport doesn't leak sockets
21- def get_transport_with_cleanup(*args, **kwargs):
22- t = orig_get_transport(*args, **kwargs)
23- if isinstance(t, _mod_transport.ConnectedTransport):
24- self.addCleanup(t.disconnect)
25- return t
26-
27- orig_get_transport = self.overrideAttr(_mod_transport, 'get_transport',
28- get_transport_with_cleanup)
29+
30+ def _add_disconnect_cleanup(transport):
31+ """Schedule disconnection of given transport at test cleanup
32+
33+ This needs to happen for all connected transports or leaks occur.
34+
35+ Note reconnections may mean we call disconnect multiple times per
36+ transport which is suboptimal but seems harmless.
37+ """
38+ self.addCleanup(transport.disconnect)
39+
40+ _mod_transport.Transport.hooks.install_named_hook('post_connect',
41+ _add_disconnect_cleanup, None)
42+
43 self._make_test_root()
44 self.addCleanup(os.chdir, os.getcwdu())
45 self.makeAndChdirToTestDir()
46
47=== modified file 'bzrlib/tests/test_transport.py'
48--- bzrlib/tests/test_transport.py 2010-10-08 07:17:16 +0000
49+++ bzrlib/tests/test_transport.py 2010-10-15 14:01:50 +0000
50@@ -465,6 +465,29 @@
51 server.stop_server()
52
53
54+class TestHooks(tests.TestCase):
55+ """Basic tests for transport hooks"""
56+
57+ def _get_connected_transport(self):
58+ return transport.ConnectedTransport("bogus:nowhere")
59+
60+ def test_transporthooks_initialisation(self):
61+ """Check all expected transport hook points are set up"""
62+ hookpoint = transport.TransportHooks()
63+ self.assertTrue("post_connect" in hookpoint,
64+ "post_connect not in %s" % (hookpoint,))
65+
66+ def test_post_connect(self):
67+ """Ensure the post_connect hook is called when _set_transport is"""
68+ calls = []
69+ transport.Transport.hooks.install_named_hook("post_connect",
70+ calls.append, None)
71+ t = self._get_connected_transport()
72+ self.assertLength(0, calls)
73+ t._set_connection("connection", "auth")
74+ self.assertEqual(calls, [t])
75+
76+
77 class PathFilteringDecoratorTransportTest(tests.TestCase):
78 """Pathfilter decoration specific tests."""
79
80
81=== modified file 'bzrlib/tests/transport_util.py'
82--- bzrlib/tests/transport_util.py 2010-02-23 07:43:11 +0000
83+++ bzrlib/tests/transport_util.py 2010-10-15 14:01:50 +0000
84@@ -14,128 +14,34 @@
85 # along with this program; if not, write to the Free Software
86 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
87
88-import bzrlib.hooks
89 from bzrlib.tests import features
90
91 # SFTPTransport offers better performances but relies on paramiko, if paramiko
92 # is not available, we fallback to FtpTransport
93 if features.paramiko.available():
94 from bzrlib.tests import test_sftp_transport
95- from bzrlib.transport import sftp
96+ from bzrlib.transport import sftp, Transport
97 _backing_scheme = 'sftp'
98 _backing_transport_class = sftp.SFTPTransport
99 _backing_test_class = test_sftp_transport.TestCaseWithSFTPServer
100 else:
101- from bzrlib.transport import ftp
102+ from bzrlib.transport import ftp, Transport
103 from bzrlib.tests import test_ftp_transport
104 _backing_scheme = 'ftp'
105 _backing_transport_class = ftp.FtpTransport
106 _backing_test_class = test_ftp_transport.TestCaseWithFTPServer
107
108-from bzrlib.transport import (
109- ConnectedTransport,
110- get_transport,
111- register_transport,
112- register_urlparse_netloc_protocol,
113- unregister_transport,
114- _unregister_urlparse_netloc_protocol,
115- )
116-
117-
118-
119-class TransportHooks(bzrlib.hooks.Hooks):
120- """Dict-mapping hook name to a list of callables for transport hooks"""
121-
122- def __init__(self):
123- super(TransportHooks, self).__init__()
124- # Invoked when the transport has just created a new connection.
125- # The api signature is (transport, connection, credentials)
126- self['_set_connection'] = []
127-
128-_hooked_scheme = 'hooked'
129-
130-def _change_scheme_in(url, actual, desired):
131- if not url.startswith(actual + '://'):
132- raise AssertionError('url "%r" does not start with "%r]"'
133- % (url, actual))
134- return desired + url[len(actual):]
135-
136-
137-class InstrumentedTransport(_backing_transport_class):
138- """Instrumented transport class to test commands behavior"""
139-
140- hooks = TransportHooks()
141-
142- def __init__(self, base, _from_transport=None):
143- if not base.startswith(_hooked_scheme + '://'):
144- raise ValueError(base)
145- # We need to trick the backing transport class about the scheme used
146- # We'll do the reverse when we need to talk to the backing server
147- fake_base = _change_scheme_in(base, _hooked_scheme, _backing_scheme)
148- super(InstrumentedTransport, self).__init__(
149- fake_base, _from_transport=_from_transport)
150- # The following is needed to minimize the effects of our trick above
151- # while retaining the best compatibility.
152- self._scheme = _hooked_scheme
153- base = self._unsplit_url(self._scheme,
154- self._user, self._password,
155- self._host, self._port,
156- self._path)
157- super(ConnectedTransport, self).__init__(base)
158-
159-
160-class ConnectionHookedTransport(InstrumentedTransport):
161- """Transport instrumented to inspect connections"""
162-
163- def _set_connection(self, connection, credentials):
164- """Called when a new connection is created """
165- super(ConnectionHookedTransport, self)._set_connection(connection,
166- credentials)
167- for hook in self.hooks['_set_connection']:
168- hook(self, connection, credentials)
169-
170
171 class TestCaseWithConnectionHookedTransport(_backing_test_class):
172
173 def setUp(self):
174- register_urlparse_netloc_protocol(_hooked_scheme)
175- register_transport(_hooked_scheme, ConnectionHookedTransport)
176- self.addCleanup(unregister_transport, _hooked_scheme,
177- ConnectionHookedTransport)
178- self.addCleanup(_unregister_urlparse_netloc_protocol, _hooked_scheme)
179 super(TestCaseWithConnectionHookedTransport, self).setUp()
180 self.reset_connections()
181- # Add the 'hooked' url to the permitted url list.
182- # XXX: See TestCase.start_server. This whole module shouldn't need to
183- # exist - a bug has been filed on that. once its cleanedup/removed, the
184- # standard test support code will work and permit the server url
185- # correctly.
186- url = self.get_url()
187- t = get_transport(url)
188- if t.base.endswith('work/'):
189- t = t.clone('../..')
190- self.permit_url(t.base)
191-
192- def get_url(self, relpath=None):
193- super_self = super(TestCaseWithConnectionHookedTransport, self)
194- url = super_self.get_url(relpath)
195- # Replace the backing scheme by our own (see
196- # InstrumentedTransport.__init__)
197- url = _change_scheme_in(url, _backing_scheme, _hooked_scheme)
198- return url
199
200 def start_logging_connections(self):
201- self.overrideAttr(InstrumentedTransport, 'hooks', TransportHooks())
202- # We preserved the hooks class attribute. Now we install our hook.
203- ConnectionHookedTransport.hooks.install_named_hook(
204- '_set_connection', self._collect_connection, None)
205+ Transport.hooks.install_named_hook('post_connect',
206+ self.connections.append, None)
207
208 def reset_connections(self):
209 self.connections = []
210
211- def _collect_connection(self, transport, connection, credentials):
212- # Note: uncomment the following line and use 'bt' under pdb, that will
213- # identify all the connections made including the extraneous ones.
214- # import pdb; pdb.set_trace()
215- self.connections.append(connection)
216-
217
218=== modified file 'bzrlib/transport/__init__.py'
219--- bzrlib/transport/__init__.py 2010-06-18 15:38:20 +0000
220+++ bzrlib/transport/__init__.py 2010-10-15 14:01:50 +0000
221@@ -52,7 +52,10 @@
222 from bzrlib.trace import (
223 mutter,
224 )
225-from bzrlib import registry
226+from bzrlib import (
227+ hooks,
228+ registry,
229+ )
230
231
232 # a dictionary of open file streams. Keys are absolute paths, values are
233@@ -267,6 +270,16 @@
234 self.transport.append_bytes(self.relpath, bytes)
235
236
237+class TransportHooks(hooks.Hooks):
238+ """Mapping of hook names to registered callbacks for transport hooks"""
239+ def __init__(self):
240+ super(TransportHooks, self).__init__()
241+ self.create_hook(hooks.HookPoint("post_connect",
242+ "Called after a new connection is established or a reconnect "
243+ "occurs. The connected transport instance is the sole argument "
244+ "passed.", (2, 3), None))
245+
246+
247 class Transport(object):
248 """This class encapsulates methods for retrieving or putting a file
249 from/to a storage location.
250@@ -291,6 +304,8 @@
251 # where the biggest benefit between combining reads and
252 # and seeking is. Consider a runtime auto-tune.
253 _bytes_to_read_before_seek = 0
254+
255+ hooks = TransportHooks()
256
257 def __init__(self, base):
258 super(Transport, self).__init__()
259@@ -1492,6 +1507,8 @@
260 """
261 self._shared_connection.connection = connection
262 self._shared_connection.credentials = credentials
263+ for hook in self.hooks["post_connect"]:
264+ hook(self)
265
266 def _get_connection(self):
267 """Returns the transport specific connection object."""
268
269=== modified file 'bzrlib/transport/remote.py'
270--- bzrlib/transport/remote.py 2010-06-18 15:38:20 +0000
271+++ bzrlib/transport/remote.py 2010-10-15 14:01:50 +0000
272@@ -111,6 +111,8 @@
273 if 'hpss' in debug.debug_flags:
274 trace.mutter('hpss: Built a new medium: %s',
275 medium.__class__.__name__)
276+ for hook in self.hooks["post_connect"]:
277+ hook(self)
278 self._shared_connection = transport._SharedConnection(medium,
279 credentials,
280 self.base)