Sat Oct 17 18:30:13 PDT 2009 Kevan Carstensen * Alter NoNetworkGrid to allow the creation of readonly servers for testing purposes. Fri Oct 30 02:19:08 PDT 2009 "Kevan Carstensen" * Refactor some behavior into a mixin, and add tests for the behavior described in #778 Tue Nov 3 19:36:02 PST 2009 Kevan Carstensen * Alter tests to use the new form of set_shareholders Tue Nov 3 19:42:32 PST 2009 Kevan Carstensen * Minor tweak to an existing test -- make the first server read-write, instead of read-only Wed Nov 4 03:13:24 PST 2009 Kevan Carstensen * Add a test for upload.shares_by_server Wed Nov 4 03:28:49 PST 2009 Kevan Carstensen * Add more tests for comment:53 in ticket #778 Sun Nov 8 16:37:35 PST 2009 Kevan Carstensen * Test Tahoe2PeerSelector to make sure that it recognizeses existing shares on readonly servers Mon Nov 16 11:23:34 PST 2009 Kevan Carstensen * Re-work 'test_upload.py' to be more readable; add more tests for #778 Sun Nov 22 17:20:08 PST 2009 Kevan Carstensen * Add tests for the behavior described in #834. Fri Dec 4 20:34:53 PST 2009 Kevan Carstensen * Replace "UploadHappinessError" with "UploadUnhappinessError" in tests. Thu Jan 7 10:13:25 PST 2010 Kevan Carstensen * Alter various unit tests to work with the new happy behavior Mon Jan 18 13:02:38 PST 2010 Kevan Carstensen * Revisions of the #778 tests, per reviewers' comments - Fix comments and confusing naming. - Add tests for the new error messages suggested by David-Sarah and Zooko. - Alter existing tests for new error messages. - Make sure that the tests continue to work with the trunk. - Add a test for a mutual disjointedness assertion that I added to upload.servers_of_happiness. New patches: [Alter NoNetworkGrid to allow the creation of readonly servers for testing purposes. Kevan Carstensen **20091018013013 Ignore-this: e12cd7c4ddeb65305c5a7e08df57c754 ] { hunk ./src/allmydata/test/no_network.py 204 c.setServiceParent(self) self.clients.append(c) - def make_server(self, i): + def make_server(self, i, readonly=False): serverid = hashutil.tagged_hash("serverid", str(i))[:20] serverdir = os.path.join(self.basedir, "servers", idlib.shortnodeid_b2a(serverid)) hunk ./src/allmydata/test/no_network.py 209 fileutil.make_dirs(serverdir) - ss = StorageServer(serverdir, serverid, stats_provider=SimpleStats()) + ss = StorageServer(serverdir, serverid, stats_provider=SimpleStats(), + readonly_storage=readonly) return ss def add_server(self, i, ss): } [Refactor some behavior into a mixin, and add tests for the behavior described in #778 "Kevan Carstensen" **20091030091908 Ignore-this: a6f9797057ca135579b249af3b2b66ac ] { hunk ./src/allmydata/test/test_upload.py 2 -import os +import os, shutil from cStringIO import StringIO from twisted.trial import unittest from twisted.python.failure import Failure hunk ./src/allmydata/test/test_upload.py 12 import allmydata # for __full_version__ from allmydata import uri, monitor, client -from allmydata.immutable import upload +from allmydata.immutable import upload, encode from allmydata.interfaces import FileTooLargeError, NoSharesError, \ NotEnoughSharesError from allmydata.util.assertutil import precondition hunk ./src/allmydata/test/test_upload.py 20 from no_network import GridTestMixin from common_util import ShouldFailMixin from allmydata.storage_client import StorageFarmBroker +from allmydata.storage.server import storage_index_to_dir MiB = 1024*1024 hunk ./src/allmydata/test/test_upload.py 91 class ServerError(Exception): pass +class SetDEPMixin: + def set_encoding_parameters(self, k, happy, n, max_segsize=1*MiB): + p = {"k": k, + "happy": happy, + "n": n, + "max_segment_size": max_segsize, + } + self.node.DEFAULT_ENCODING_PARAMETERS = p + class FakeStorageServer: def __init__(self, mode): self.mode = mode hunk ./src/allmydata/test/test_upload.py 247 u = upload.FileHandle(fh, convergence=None) return uploader.upload(u) -class GoodServer(unittest.TestCase, ShouldFailMixin): +class GoodServer(unittest.TestCase, ShouldFailMixin, SetDEPMixin): def setUp(self): self.node = FakeClient(mode="good") self.u = upload.Uploader() hunk ./src/allmydata/test/test_upload.py 254 self.u.running = True self.u.parent = self.node - def set_encoding_parameters(self, k, happy, n, max_segsize=1*MiB): - p = {"k": k, - "happy": happy, - "n": n, - "max_segment_size": max_segsize, - } - self.node.DEFAULT_ENCODING_PARAMETERS = p - def _check_small(self, newuri, size): u = uri.from_string(newuri) self.failUnless(isinstance(u, uri.LiteralFileURI)) hunk ./src/allmydata/test/test_upload.py 377 d.addCallback(self._check_large, SIZE_LARGE) return d -class ServerErrors(unittest.TestCase, ShouldFailMixin): +class ServerErrors(unittest.TestCase, ShouldFailMixin, SetDEPMixin): def make_node(self, mode, num_servers=10): self.node = FakeClient(mode, num_servers) self.u = upload.Uploader() hunk ./src/allmydata/test/test_upload.py 677 d.addCallback(_done) return d -class EncodingParameters(GridTestMixin, unittest.TestCase): +class EncodingParameters(GridTestMixin, unittest.TestCase, SetDEPMixin, + ShouldFailMixin): + def _do_upload_with_broken_servers(self, servers_to_break): + """ + I act like a normal upload, but before I send the results of + Tahoe2PeerSelector to the Encoder, I break the first servers_to_break + PeerTrackers in the used_peers part of the return result. + """ + assert self.g, "I tried to find a grid at self.g, but failed" + broker = self.g.clients[0].storage_broker + sh = self.g.clients[0]._secret_holder + data = upload.Data("data" * 10000, convergence="") + data.encoding_param_k = 3 + data.encoding_param_happy = 4 + data.encoding_param_n = 10 + uploadable = upload.EncryptAnUploadable(data) + encoder = encode.Encoder() + encoder.set_encrypted_uploadable(uploadable) + status = upload.UploadStatus() + selector = upload.Tahoe2PeerSelector("dglev", "test", status) + storage_index = encoder.get_param("storage_index") + share_size = encoder.get_param("share_size") + block_size = encoder.get_param("block_size") + num_segments = encoder.get_param("num_segments") + d = selector.get_shareholders(broker, sh, storage_index, + share_size, block_size, num_segments, + 10, 4) + def _have_shareholders((used_peers, already_peers)): + assert servers_to_break <= len(used_peers) + for index in xrange(servers_to_break): + server = list(used_peers)[index] + for share in server.buckets.keys(): + server.buckets[share].abort() + buckets = {} + for peer in used_peers: + buckets.update(peer.buckets) + encoder.set_shareholders(buckets) + d = encoder.start() + return d + d.addCallback(_have_shareholders) + return d + + def _add_server_with_share(self, server_number, share_number=None, + readonly=False): + assert self.g, "I tried to find a grid at self.g, but failed" + assert self.shares, "I tried to find shares at self.shares, but failed" + ss = self.g.make_server(server_number, readonly) + self.g.add_server(server_number, ss) + if share_number: + # Copy share i from the directory associated with the first + # storage server to the directory associated with this one. + old_share_location = self.shares[share_number][2] + new_share_location = os.path.join(ss.storedir, "shares") + si = uri.from_string(self.uri).get_storage_index() + new_share_location = os.path.join(new_share_location, + storage_index_to_dir(si)) + if not os.path.exists(new_share_location): + os.makedirs(new_share_location) + new_share_location = os.path.join(new_share_location, + str(share_number)) + shutil.copy(old_share_location, new_share_location) + shares = self.find_shares(self.uri) + # Make sure that the storage server has the share. + self.failUnless((share_number, ss.my_nodeid, new_share_location) + in shares) + + def _setup_and_upload(self): + """ + I set up a NoNetworkGrid with a single server and client, + upload a file to it, store its uri in self.uri, and store its + sharedata in self.shares. + """ + self.set_up_grid(num_clients=1, num_servers=1) + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = 1 + data = upload.Data("data" * 10000, convergence="") + self.data = data + d = client.upload(data) + def _store_uri(ur): + self.uri = ur.uri + d.addCallback(_store_uri) + d.addCallback(lambda ign: + self.find_shares(self.uri)) + def _store_shares(shares): + self.shares = shares + d.addCallback(_store_shares) + return d + def test_configure_parameters(self): self.basedir = self.mktemp() hooks = {0: self._set_up_nodes_extra_config} hunk ./src/allmydata/test/test_upload.py 784 d.addCallback(_check) return d + def _setUp(self, ns): + # Used by test_happy_semantics and test_prexisting_share_behavior + # to set up the grid. + self.node = FakeClient(mode="good", num_servers=ns) + self.u = upload.Uploader() + self.u.running = True + self.u.parent = self.node + + def test_happy_semantics(self): + self._setUp(2) + DATA = upload.Data("kittens" * 10000, convergence="") + # These parameters are unsatisfiable with the client that we've made + # -- we'll use them to test that the semnatics work correctly. + self.set_encoding_parameters(k=3, happy=5, n=10) + d = self.shouldFail(NotEnoughSharesError, "test_happy_semantics", + "shares could only be placed on 2 servers " + "(5 were requested)", + self.u.upload, DATA) + # Let's reset the client to have 10 servers + d.addCallback(lambda ign: + self._setUp(10)) + # These parameters are satisfiable with the client we've made. + d.addCallback(lambda ign: + self.set_encoding_parameters(k=3, happy=5, n=10)) + # this should work + d.addCallback(lambda ign: + self.u.upload(DATA)) + # Let's reset the client to have 7 servers + # (this is less than n, but more than h) + d.addCallback(lambda ign: + self._setUp(7)) + # These encoding parameters should still be satisfiable with our + # client setup + d.addCallback(lambda ign: + self.set_encoding_parameters(k=3, happy=5, n=10)) + # This, then, should work. + d.addCallback(lambda ign: + self.u.upload(DATA)) + return d + + def test_problem_layouts(self): + self.basedir = self.mktemp() + # This scenario is at + # http://allmydata.org/trac/tahoe/ticket/778#comment:52 + # + # The scenario in comment:52 proposes that we have a layout + # like: + # server 1: share 1 + # server 2: share 1 + # server 3: share 1 + # server 4: shares 2 - 10 + # To get access to the shares, we will first upload to one + # server, which will then have shares 1 - 10. We'll then + # add three new servers, configure them to not accept any new + # shares, then write share 1 directly into the serverdir of each. + # Then each of servers 1 - 3 will report that they have share 1, + # and will not accept any new share, while server 4 will report that + # it has shares 2 - 10 and will accept new shares. + # We'll then set 'happy' = 4, and see that an upload fails + # (as it should) + d = self._setup_and_upload() + d.addCallback(lambda ign: + self._add_server_with_share(1, 0, True)) + d.addCallback(lambda ign: + self._add_server_with_share(2, 0, True)) + d.addCallback(lambda ign: + self._add_server_with_share(3, 0, True)) + # Remove the first share from server 0. + def _remove_share_0(): + share_location = self.shares[0][2] + os.remove(share_location) + d.addCallback(lambda ign: + _remove_share_0()) + # Set happy = 4 in the client. + def _prepare(): + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = 4 + return client + d.addCallback(lambda ign: + _prepare()) + # Uploading data should fail + d.addCallback(lambda client: + self.shouldFail(NotEnoughSharesError, "test_happy_semantics", + "shares could only be placed on 1 servers " + "(4 were requested)", + client.upload, upload.Data("data" * 10000, + convergence=""))) + + + # This scenario is at + # http://allmydata.org/trac/tahoe/ticket/778#comment:53 + # + # Set up the grid to have one server + def _change_basedir(ign): + self.basedir = self.mktemp() + d.addCallback(_change_basedir) + d.addCallback(lambda ign: + self._setup_and_upload()) + # We want to have a layout like this: + # server 1: share 1 + # server 2: share 2 + # server 3: share 3 + # server 4: shares 1 - 10 + # (this is an expansion of Zooko's example because it is easier + # to code, but it will fail in the same way) + # To start, we'll create a server with shares 1-10 of the data + # we're about to upload. + # Next, we'll add three new servers to our NoNetworkGrid. We'll add + # one share from our initial upload to each of these. + # The counterintuitive ordering of the share numbers is to deal with + # the permuting of these servers -- distributing the shares this + # way ensures that the Tahoe2PeerSelector sees them in the order + # described above. + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1, share_number=2)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2, share_number=0)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3, share_number=1)) + # So, we now have the following layout: + # server 0: shares 1 - 10 + # server 1: share 0 + # server 2: share 1 + # server 3: share 2 + # We want to change the 'happy' parameter in the client to 4. + # We then want to feed the upload process a list of peers that + # server 0 is at the front of, so we trigger Zooko's scenario. + # Ideally, a reupload of our original data should work. + def _reset_encoding_parameters(ign): + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = 4 + return client + d.addCallback(_reset_encoding_parameters) + # We need this to get around the fact that the old Data + # instance already has a happy parameter set. + d.addCallback(lambda client: + client.upload(upload.Data("data" * 10000, convergence=""))) + return d + + + def test_dropped_servers_in_encoder(self): + def _set_basedir(ign=None): + self.basedir = self.mktemp() + _set_basedir() + d = self._setup_and_upload(); + # Add 5 servers, with one share each from the original + # Add a readonly server + def _do_server_setup(ign): + self._add_server_with_share(1, 1, True) + self._add_server_with_share(2) + self._add_server_with_share(3) + self._add_server_with_share(4) + self._add_server_with_share(5) + d.addCallback(_do_server_setup) + # remove the original server + # (necessary to ensure that the Tahoe2PeerSelector will distribute + # all the shares) + def _remove_server(ign): + server = self.g.servers_by_number[0] + self.g.remove_server(server.my_nodeid) + d.addCallback(_remove_server) + # This should succeed. + d.addCallback(lambda ign: + self._do_upload_with_broken_servers(1)) + # Now, do the same thing over again, but drop 2 servers instead + # of 1. This should fail. + d.addCallback(_set_basedir) + d.addCallback(lambda ign: + self._setup_and_upload()) + d.addCallback(_do_server_setup) + d.addCallback(_remove_server) + d.addCallback(lambda ign: + self.shouldFail(NotEnoughSharesError, + "test_dropped_server_in_encoder", "", + self._do_upload_with_broken_servers, 2)) + return d + + + def test_servers_with_unique_shares(self): + # servers_with_unique_shares expects a dict of + # shnum => peerid as a preexisting shares argument. + test1 = { + 1 : "server1", + 2 : "server2", + 3 : "server3", + 4 : "server4" + } + unique_servers = upload.servers_with_unique_shares(test1) + self.failUnlessEqual(4, len(unique_servers)) + for server in ["server1", "server2", "server3", "server4"]: + self.failUnlessIn(server, unique_servers) + test1[4] = "server1" + # Now there should only be 3 unique servers. + unique_servers = upload.servers_with_unique_shares(test1) + self.failUnlessEqual(3, len(unique_servers)) + for server in ["server1", "server2", "server3"]: + self.failUnlessIn(server, unique_servers) + # servers_with_unique_shares expects a set of PeerTracker + # instances as a used_peers argument, but only uses the peerid + # instance variable to assess uniqueness. So we feed it some fake + # PeerTrackers whose only important characteristic is that they + # have peerid set to something. + class FakePeerTracker: + pass + trackers = [] + for server in ["server5", "server6", "server7", "server8"]: + t = FakePeerTracker() + t.peerid = server + trackers.append(t) + # Recall that there are 3 unique servers in test1. Since none of + # those overlap with the ones in trackers, we should get 7 back + unique_servers = upload.servers_with_unique_shares(test1, set(trackers)) + self.failUnlessEqual(7, len(unique_servers)) + expected_servers = ["server" + str(i) for i in xrange(1, 9)] + expected_servers.remove("server4") + for server in expected_servers: + self.failUnlessIn(server, unique_servers) + # Now add an overlapping server to trackers. + t = FakePeerTracker() + t.peerid = "server1" + trackers.append(t) + unique_servers = upload.servers_with_unique_shares(test1, set(trackers)) + self.failUnlessEqual(7, len(unique_servers)) + for server in expected_servers: + self.failUnlessIn(server, unique_servers) + + def _set_up_nodes_extra_config(self, clientdir): cfgfn = os.path.join(clientdir, "tahoe.cfg") oldcfg = open(cfgfn, "r").read() } [Alter tests to use the new form of set_shareholders Kevan Carstensen **20091104033602 Ignore-this: 3deac11fc831618d11441317463ef830 ] { hunk ./src/allmydata/test/test_encode.py 301 (NUM_SEGMENTS-1)*segsize, len(data), NUM_SEGMENTS*segsize) shareholders = {} + servermap = {} for shnum in range(NUM_SHARES): peer = FakeBucketReaderWriterProxy() shareholders[shnum] = peer hunk ./src/allmydata/test/test_encode.py 305 + servermap[shnum] = str(shnum) all_shareholders.append(peer) hunk ./src/allmydata/test/test_encode.py 307 - e.set_shareholders(shareholders) + e.set_shareholders(shareholders, servermap) return e.start() d.addCallback(_ready) merger 0.0 ( hunk ./src/allmydata/test/test_encode.py 462 - all_peers = [] hunk ./src/allmydata/test/test_encode.py 463 + servermap = {} ) hunk ./src/allmydata/test/test_encode.py 467 mode = bucket_modes.get(shnum, "good") peer = FakeBucketReaderWriterProxy(mode) shareholders[shnum] = peer - e.set_shareholders(shareholders) + servermap[shnum] = str(shnum) + e.set_shareholders(shareholders, servermap) return e.start() d.addCallback(_ready) def _sent(res): hunk ./src/allmydata/test/test_upload.py 711 for share in server.buckets.keys(): server.buckets[share].abort() buckets = {} + servermap = already_peers.copy() for peer in used_peers: buckets.update(peer.buckets) hunk ./src/allmydata/test/test_upload.py 714 - encoder.set_shareholders(buckets) + for bucket in peer.buckets: + servermap[bucket] = peer.peerid + encoder.set_shareholders(buckets, servermap) d = encoder.start() return d d.addCallback(_have_shareholders) hunk ./src/allmydata/test/test_upload.py 933 _set_basedir() d = self._setup_and_upload(); # Add 5 servers, with one share each from the original - # Add a readonly server def _do_server_setup(ign): self._add_server_with_share(1, 1, True) self._add_server_with_share(2) } [Minor tweak to an existing test -- make the first server read-write, instead of read-only Kevan Carstensen **20091104034232 Ignore-this: a951a46c93f7f58dd44d93d8623b2aee ] hunk ./src/allmydata/test/test_upload.py 934 d = self._setup_and_upload(); # Add 5 servers, with one share each from the original def _do_server_setup(ign): - self._add_server_with_share(1, 1, True) + self._add_server_with_share(1, 1) self._add_server_with_share(2) self._add_server_with_share(3) self._add_server_with_share(4) [Add a test for upload.shares_by_server Kevan Carstensen **20091104111324 Ignore-this: f9802e82d6982a93e00f92e0b276f018 ] hunk ./src/allmydata/test/test_upload.py 1013 self.failUnlessIn(server, unique_servers) + def test_shares_by_server(self): + test = { + 1 : "server1", + 2 : "server2", + 3 : "server3", + 4 : "server4" + } + shares_by_server = upload.shares_by_server(test) + self.failUnlessEqual(set([1]), shares_by_server["server1"]) + self.failUnlessEqual(set([2]), shares_by_server["server2"]) + self.failUnlessEqual(set([3]), shares_by_server["server3"]) + self.failUnlessEqual(set([4]), shares_by_server["server4"]) + test1 = { + 1 : "server1", + 2 : "server1", + 3 : "server1", + 4 : "server2", + 5 : "server2" + } + shares_by_server = upload.shares_by_server(test1) + self.failUnlessEqual(set([1, 2, 3]), shares_by_server["server1"]) + self.failUnlessEqual(set([4, 5]), shares_by_server["server2"]) + + def _set_up_nodes_extra_config(self, clientdir): cfgfn = os.path.join(clientdir, "tahoe.cfg") oldcfg = open(cfgfn, "r").read() [Add more tests for comment:53 in ticket #778 Kevan Carstensen **20091104112849 Ignore-this: 3bb2edd299a944cc9586e14d5d83ec8c ] { hunk ./src/allmydata/test/test_upload.py 722 d.addCallback(_have_shareholders) return d - def _add_server_with_share(self, server_number, share_number=None, - readonly=False): + def _add_server(self, server_number, readonly=False): assert self.g, "I tried to find a grid at self.g, but failed" assert self.shares, "I tried to find shares at self.shares, but failed" ss = self.g.make_server(server_number, readonly) hunk ./src/allmydata/test/test_upload.py 727 self.g.add_server(server_number, ss) + + def _add_server_with_share(self, server_number, share_number=None, + readonly=False): + self._add_server(server_number, readonly) if share_number: hunk ./src/allmydata/test/test_upload.py 732 - # Copy share i from the directory associated with the first - # storage server to the directory associated with this one. - old_share_location = self.shares[share_number][2] - new_share_location = os.path.join(ss.storedir, "shares") - si = uri.from_string(self.uri).get_storage_index() - new_share_location = os.path.join(new_share_location, - storage_index_to_dir(si)) - if not os.path.exists(new_share_location): - os.makedirs(new_share_location) - new_share_location = os.path.join(new_share_location, - str(share_number)) - shutil.copy(old_share_location, new_share_location) - shares = self.find_shares(self.uri) - # Make sure that the storage server has the share. - self.failUnless((share_number, ss.my_nodeid, new_share_location) - in shares) + self._copy_share_to_server(share_number, server_number) + + def _copy_share_to_server(self, share_number, server_number): + ss = self.g.servers_by_number[server_number] + # Copy share i from the directory associated with the first + # storage server to the directory associated with this one. + assert self.g, "I tried to find a grid at self.g, but failed" + assert self.shares, "I tried to find shares at self.shares, but failed" + old_share_location = self.shares[share_number][2] + new_share_location = os.path.join(ss.storedir, "shares") + si = uri.from_string(self.uri).get_storage_index() + new_share_location = os.path.join(new_share_location, + storage_index_to_dir(si)) + if not os.path.exists(new_share_location): + os.makedirs(new_share_location) + new_share_location = os.path.join(new_share_location, + str(share_number)) + shutil.copy(old_share_location, new_share_location) + shares = self.find_shares(self.uri) + # Make sure that the storage server has the share. + self.failUnless((share_number, ss.my_nodeid, new_share_location) + in shares) + def _setup_and_upload(self): """ hunk ./src/allmydata/test/test_upload.py 917 d.addCallback(lambda ign: self._add_server_with_share(server_number=3, share_number=1)) # So, we now have the following layout: - # server 0: shares 1 - 10 + # server 0: shares 0 - 9 # server 1: share 0 # server 2: share 1 # server 3: share 2 hunk ./src/allmydata/test/test_upload.py 934 # instance already has a happy parameter set. d.addCallback(lambda client: client.upload(upload.Data("data" * 10000, convergence=""))) + + + # This scenario is basically comment:53, but with the order reversed; + # this means that the Tahoe2PeerSelector sees + # server 0: shares 1-10 + # server 1: share 1 + # server 2: share 2 + # server 3: share 3 + d.addCallback(_change_basedir) + d.addCallback(lambda ign: + self._setup_and_upload()) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2, share_number=0)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3, share_number=1)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1, share_number=2)) + # Copy all of the other shares to server number 2 + def _copy_shares(ign): + for i in xrange(1, 10): + self._copy_share_to_server(i, 2) + d.addCallback(_copy_shares) + # Remove the first server, and add a placeholder with share 0 + d.addCallback(lambda ign: + self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=0, share_number=0)) + # Now try uploading. + d.addCallback(_reset_encoding_parameters) + d.addCallback(lambda client: + client.upload(upload.Data("data" * 10000, convergence=""))) + # Try the same thing, but with empty servers after the first one + # We want to make sure that Tahoe2PeerSelector will redistribute + # shares as necessary, not simply discover an existing layout. + d.addCallback(_change_basedir) + d.addCallback(lambda ign: + self._setup_and_upload()) + d.addCallback(lambda ign: + self._add_server(server_number=2)) + d.addCallback(lambda ign: + self._add_server(server_number=3)) + d.addCallback(lambda ign: + self._add_server(server_number=1)) + d.addCallback(_copy_shares) + d.addCallback(lambda ign: + self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) + d.addCallback(lambda ign: + self._add_server(server_number=0)) + d.addCallback(_reset_encoding_parameters) + d.addCallback(lambda client: + client.upload(upload.Data("data" * 10000, convergence=""))) + # Try the following layout + # server 0: shares 1-10 + # server 1: share 1, read-only + # server 2: share 2, read-only + # server 3: share 3, read-only + d.addCallback(_change_basedir) + d.addCallback(lambda ign: + self._setup_and_upload()) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2, share_number=0)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3, share_number=1, + readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1, share_number=2, + readonly=True)) + # Copy all of the other shares to server number 2 + d.addCallback(_copy_shares) + # Remove server 0, and add another in its place + d.addCallback(lambda ign: + self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=0, share_number=0, + readonly=True)) + d.addCallback(_reset_encoding_parameters) + d.addCallback(lambda client: + client.upload(upload.Data("data" * 10000, convergence=""))) return d } [Test Tahoe2PeerSelector to make sure that it recognizeses existing shares on readonly servers Kevan Carstensen **20091109003735 Ignore-this: 12f9b4cff5752fca7ed32a6ebcff6446 ] hunk ./src/allmydata/test/test_upload.py 1125 self.failUnlessEqual(set([4, 5]), shares_by_server["server2"]) + def test_existing_share_detection(self): + self.basedir = self.mktemp() + d = self._setup_and_upload() + # Our final setup should look like this: + # server 1: shares 1 - 10, read-only + # server 2: empty + # server 3: empty + # server 4: empty + # The purpose of this test is to make sure that the peer selector + # knows about the shares on server 1, even though it is read-only. + # It used to simply filter these out, which would cause the test + # to fail when servers_of_happiness = 4. + d.addCallback(lambda ign: + self._add_server_with_share(1, 0, True)) + d.addCallback(lambda ign: + self._add_server_with_share(2)) + d.addCallback(lambda ign: + self._add_server_with_share(3)) + d.addCallback(lambda ign: + self._add_server_with_share(4)) + def _copy_shares(ign): + for i in xrange(1, 10): + self._copy_share_to_server(i, 1) + d.addCallback(_copy_shares) + d.addCallback(lambda ign: + self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) + def _prepare_client(ign): + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = 4 + return client + d.addCallback(_prepare_client) + d.addCallback(lambda client: + client.upload(upload.Data("data" * 10000, convergence=""))) + return d + + def _set_up_nodes_extra_config(self, clientdir): cfgfn = os.path.join(clientdir, "tahoe.cfg") oldcfg = open(cfgfn, "r").read() [Re-work 'test_upload.py' to be more readable; add more tests for #778 Kevan Carstensen **20091116192334 Ignore-this: 7e8565f92fe51dece5ae28daf442d659 ] { hunk ./src/allmydata/test/test_upload.py 722 d.addCallback(_have_shareholders) return d + def _add_server(self, server_number, readonly=False): assert self.g, "I tried to find a grid at self.g, but failed" assert self.shares, "I tried to find shares at self.shares, but failed" hunk ./src/allmydata/test/test_upload.py 729 ss = self.g.make_server(server_number, readonly) self.g.add_server(server_number, ss) + def _add_server_with_share(self, server_number, share_number=None, readonly=False): self._add_server(server_number, readonly) hunk ./src/allmydata/test/test_upload.py 733 - if share_number: + if share_number is not None: self._copy_share_to_server(share_number, server_number) hunk ./src/allmydata/test/test_upload.py 736 + def _copy_share_to_server(self, share_number, server_number): ss = self.g.servers_by_number[server_number] # Copy share i from the directory associated with the first hunk ./src/allmydata/test/test_upload.py 752 os.makedirs(new_share_location) new_share_location = os.path.join(new_share_location, str(share_number)) - shutil.copy(old_share_location, new_share_location) + if old_share_location != new_share_location: + shutil.copy(old_share_location, new_share_location) shares = self.find_shares(self.uri) # Make sure that the storage server has the share. self.failUnless((share_number, ss.my_nodeid, new_share_location) hunk ./src/allmydata/test/test_upload.py 782 d.addCallback(_store_shares) return d + def test_configure_parameters(self): self.basedir = self.mktemp() hooks = {0: self._set_up_nodes_extra_config} hunk ./src/allmydata/test/test_upload.py 802 d.addCallback(_check) return d + def _setUp(self, ns): # Used by test_happy_semantics and test_prexisting_share_behavior # to set up the grid. hunk ./src/allmydata/test/test_upload.py 811 self.u.running = True self.u.parent = self.node + def test_happy_semantics(self): self._setUp(2) DATA = upload.Data("kittens" * 10000, convergence="") hunk ./src/allmydata/test/test_upload.py 844 self.u.upload(DATA)) return d - def test_problem_layouts(self): - self.basedir = self.mktemp() + + def test_problem_layout_comment_52(self): + def _basedir(): + self.basedir = self.mktemp() + _basedir() # This scenario is at # http://allmydata.org/trac/tahoe/ticket/778#comment:52 # hunk ./src/allmydata/test/test_upload.py 890 # Uploading data should fail d.addCallback(lambda client: self.shouldFail(NotEnoughSharesError, "test_happy_semantics", - "shares could only be placed on 1 servers " + "shares could only be placed on 2 servers " "(4 were requested)", client.upload, upload.Data("data" * 10000, convergence=""))) hunk ./src/allmydata/test/test_upload.py 895 + # Do comment:52, but like this: + # server 2: empty + # server 3: share 0, read-only + # server 1: share 0, read-only + # server 0: shares 0-9 + d.addCallback(lambda ign: + _basedir()) + d.addCallback(lambda ign: + self._setup_and_upload()) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3, share_number=0, + readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1, share_number=0, + readonly=True)) + def _prepare2(): + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = 3 + return client + d.addCallback(lambda ign: + _prepare2()) + d.addCallback(lambda client: + self.shouldFail(NotEnoughSharesError, "test_happy_sematics", + "shares could only be placed on 2 servers " + "(3 were requested)", + client.upload, upload.Data("data" * 10000, + convergence=""))) + return d + hunk ./src/allmydata/test/test_upload.py 927 + def test_problem_layout_comment_53(self): # This scenario is at # http://allmydata.org/trac/tahoe/ticket/778#comment:53 # hunk ./src/allmydata/test/test_upload.py 934 # Set up the grid to have one server def _change_basedir(ign): self.basedir = self.mktemp() - d.addCallback(_change_basedir) - d.addCallback(lambda ign: - self._setup_and_upload()) - # We want to have a layout like this: - # server 1: share 1 - # server 2: share 2 - # server 3: share 3 - # server 4: shares 1 - 10 - # (this is an expansion of Zooko's example because it is easier - # to code, but it will fail in the same way) - # To start, we'll create a server with shares 1-10 of the data - # we're about to upload. + _change_basedir(None) + d = self._setup_and_upload() + # We start by uploading all of the shares to one server (which has + # already been done above). # Next, we'll add three new servers to our NoNetworkGrid. We'll add # one share from our initial upload to each of these. # The counterintuitive ordering of the share numbers is to deal with hunk ./src/allmydata/test/test_upload.py 952 self._add_server_with_share(server_number=3, share_number=1)) # So, we now have the following layout: # server 0: shares 0 - 9 - # server 1: share 0 - # server 2: share 1 - # server 3: share 2 + # server 1: share 2 + # server 2: share 0 + # server 3: share 1 # We want to change the 'happy' parameter in the client to 4. hunk ./src/allmydata/test/test_upload.py 956 - # We then want to feed the upload process a list of peers that - # server 0 is at the front of, so we trigger Zooko's scenario. + # The Tahoe2PeerSelector will see the peers permuted as: + # 2, 3, 1, 0 # Ideally, a reupload of our original data should work. hunk ./src/allmydata/test/test_upload.py 959 - def _reset_encoding_parameters(ign): + def _reset_encoding_parameters(ign, happy=4): client = self.g.clients[0] hunk ./src/allmydata/test/test_upload.py 961 - client.DEFAULT_ENCODING_PARAMETERS['happy'] = 4 + client.DEFAULT_ENCODING_PARAMETERS['happy'] = happy return client d.addCallback(_reset_encoding_parameters) hunk ./src/allmydata/test/test_upload.py 964 - # We need this to get around the fact that the old Data - # instance already has a happy parameter set. d.addCallback(lambda client: client.upload(upload.Data("data" * 10000, convergence=""))) hunk ./src/allmydata/test/test_upload.py 970 # This scenario is basically comment:53, but with the order reversed; # this means that the Tahoe2PeerSelector sees - # server 0: shares 1-10 - # server 1: share 1 - # server 2: share 2 - # server 3: share 3 + # server 2: shares 1-10 + # server 3: share 1 + # server 1: share 2 + # server 4: share 3 d.addCallback(_change_basedir) d.addCallback(lambda ign: self._setup_and_upload()) hunk ./src/allmydata/test/test_upload.py 992 d.addCallback(lambda ign: self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) d.addCallback(lambda ign: - self._add_server_with_share(server_number=0, share_number=0)) + self._add_server_with_share(server_number=4, share_number=0)) # Now try uploading. d.addCallback(_reset_encoding_parameters) d.addCallback(lambda client: hunk ./src/allmydata/test/test_upload.py 1013 d.addCallback(lambda ign: self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) d.addCallback(lambda ign: - self._add_server(server_number=0)) + self._add_server(server_number=4)) d.addCallback(_reset_encoding_parameters) d.addCallback(lambda client: client.upload(upload.Data("data" * 10000, convergence=""))) hunk ./src/allmydata/test/test_upload.py 1017 + return d + + + def test_happiness_with_some_readonly_peers(self): # Try the following layout hunk ./src/allmydata/test/test_upload.py 1022 - # server 0: shares 1-10 - # server 1: share 1, read-only - # server 2: share 2, read-only - # server 3: share 3, read-only - d.addCallback(_change_basedir) - d.addCallback(lambda ign: - self._setup_and_upload()) + # server 2: shares 0-9 + # server 4: share 0, read-only + # server 3: share 1, read-only + # server 1: share 2, read-only + self.basedir = self.mktemp() + d = self._setup_and_upload() d.addCallback(lambda ign: self._add_server_with_share(server_number=2, share_number=0)) d.addCallback(lambda ign: hunk ./src/allmydata/test/test_upload.py 1037 self._add_server_with_share(server_number=1, share_number=2, readonly=True)) # Copy all of the other shares to server number 2 + def _copy_shares(ign): + for i in xrange(1, 10): + self._copy_share_to_server(i, 2) d.addCallback(_copy_shares) # Remove server 0, and add another in its place d.addCallback(lambda ign: hunk ./src/allmydata/test/test_upload.py 1045 self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) d.addCallback(lambda ign: - self._add_server_with_share(server_number=0, share_number=0, + self._add_server_with_share(server_number=4, share_number=0, readonly=True)) hunk ./src/allmydata/test/test_upload.py 1047 + def _reset_encoding_parameters(ign, happy=4): + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = happy + return client + d.addCallback(_reset_encoding_parameters) + d.addCallback(lambda client: + client.upload(upload.Data("data" * 10000, convergence=""))) + return d + + + def test_happiness_with_all_readonly_peers(self): + # server 3: share 1, read-only + # server 1: share 2, read-only + # server 2: shares 0-9, read-only + # server 4: share 0, read-only + # The idea with this test is to make sure that the survey of + # read-only peers doesn't undercount servers of happiness + self.basedir = self.mktemp() + d = self._setup_and_upload() + d.addCallback(lambda ign: + self._add_server_with_share(server_number=4, share_number=0, + readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3, share_number=1, + readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1, share_number=2, + readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2, share_number=0, + readonly=True)) + def _copy_shares(ign): + for i in xrange(1, 10): + self._copy_share_to_server(i, 2) + d.addCallback(_copy_shares) + d.addCallback(lambda ign: + self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) + def _reset_encoding_parameters(ign, happy=4): + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = happy + return client d.addCallback(_reset_encoding_parameters) d.addCallback(lambda client: client.upload(upload.Data("data" * 10000, convergence=""))) hunk ./src/allmydata/test/test_upload.py 1099 self.basedir = self.mktemp() _set_basedir() d = self._setup_and_upload(); - # Add 5 servers, with one share each from the original + # Add 5 servers def _do_server_setup(ign): hunk ./src/allmydata/test/test_upload.py 1101 - self._add_server_with_share(1, 1) + self._add_server_with_share(1) self._add_server_with_share(2) self._add_server_with_share(3) self._add_server_with_share(4) hunk ./src/allmydata/test/test_upload.py 1126 d.addCallback(_remove_server) d.addCallback(lambda ign: self.shouldFail(NotEnoughSharesError, - "test_dropped_server_in_encoder", "", + "test_dropped_servers_in_encoder", + "lost too many servers during upload " + "(still have 3, want 4)", + self._do_upload_with_broken_servers, 2)) + # Now do the same thing over again, but make some of the servers + # readonly, break some of the ones that aren't, and make sure that + # happiness accounting is preserved. + d.addCallback(_set_basedir) + d.addCallback(lambda ign: + self._setup_and_upload()) + def _do_server_setup_2(ign): + self._add_server_with_share(1) + self._add_server_with_share(2) + self._add_server_with_share(3) + self._add_server_with_share(4, 7, readonly=True) + self._add_server_with_share(5, 8, readonly=True) + d.addCallback(_do_server_setup_2) + d.addCallback(_remove_server) + d.addCallback(lambda ign: + self._do_upload_with_broken_servers(1)) + d.addCallback(_set_basedir) + d.addCallback(lambda ign: + self._setup_and_upload()) + d.addCallback(_do_server_setup_2) + d.addCallback(_remove_server) + d.addCallback(lambda ign: + self.shouldFail(NotEnoughSharesError, + "test_dropped_servers_in_encoder", + "lost too many servers during upload " + "(still have 3, want 4)", self._do_upload_with_broken_servers, 2)) return d hunk ./src/allmydata/test/test_upload.py 1179 self.failUnlessEqual(3, len(unique_servers)) for server in ["server1", "server2", "server3"]: self.failUnlessIn(server, unique_servers) - # servers_with_unique_shares expects a set of PeerTracker - # instances as a used_peers argument, but only uses the peerid - # instance variable to assess uniqueness. So we feed it some fake - # PeerTrackers whose only important characteristic is that they - # have peerid set to something. + # servers_with_unique_shares expects to receive some object with + # a peerid attribute. So we make a FakePeerTracker whose only + # job is to have a peerid attribute. class FakePeerTracker: pass trackers = [] hunk ./src/allmydata/test/test_upload.py 1185 - for server in ["server5", "server6", "server7", "server8"]: + for (i, server) in [(i, "server%d" % i) for i in xrange(5, 9)]: t = FakePeerTracker() t.peerid = server hunk ./src/allmydata/test/test_upload.py 1188 + t.buckets = [i] trackers.append(t) # Recall that there are 3 unique servers in test1. Since none of # those overlap with the ones in trackers, we should get 7 back hunk ./src/allmydata/test/test_upload.py 1201 # Now add an overlapping server to trackers. t = FakePeerTracker() t.peerid = "server1" + t.buckets = [1] trackers.append(t) unique_servers = upload.servers_with_unique_shares(test1, set(trackers)) self.failUnlessEqual(7, len(unique_servers)) hunk ./src/allmydata/test/test_upload.py 1207 for server in expected_servers: self.failUnlessIn(server, unique_servers) + test = {} + unique_servers = upload.servers_with_unique_shares(test) + self.failUnlessEqual(0, len(test)) def test_shares_by_server(self): hunk ./src/allmydata/test/test_upload.py 1213 - test = { - 1 : "server1", - 2 : "server2", - 3 : "server3", - 4 : "server4" - } + test = dict([(i, "server%d" % i) for i in xrange(1, 5)]) shares_by_server = upload.shares_by_server(test) self.failUnlessEqual(set([1]), shares_by_server["server1"]) self.failUnlessEqual(set([2]), shares_by_server["server2"]) hunk ./src/allmydata/test/test_upload.py 1267 return d + def test_should_add_server(self): + shares = dict([(i, "server%d" % i) for i in xrange(10)]) + self.failIf(upload.should_add_server(shares, "server1", 4)) + shares[4] = "server1" + self.failUnless(upload.should_add_server(shares, "server4", 4)) + shares = {} + self.failUnless(upload.should_add_server(shares, "server1", 1)) + + def _set_up_nodes_extra_config(self, clientdir): cfgfn = os.path.join(clientdir, "tahoe.cfg") oldcfg = open(cfgfn, "r").read() } [Add tests for the behavior described in #834. Kevan Carstensen **20091123012008 Ignore-this: d8e0aa0f3f7965ce9b5cea843c6d6f9f ] { hunk ./src/allmydata/test/test_encode.py 12 from allmydata.util.assertutil import _assert from allmydata.util.consumer import MemoryConsumer from allmydata.interfaces import IStorageBucketWriter, IStorageBucketReader, \ - NotEnoughSharesError, IStorageBroker + NotEnoughSharesError, IStorageBroker, UploadHappinessError from allmydata.monitor import Monitor import common_util as testutil hunk ./src/allmydata/test/test_encode.py 794 d = self.send_and_recover((4,8,10), bucket_modes=modemap) def _done(res): self.failUnless(isinstance(res, Failure)) - self.failUnless(res.check(NotEnoughSharesError), res) + self.failUnless(res.check(UploadHappinessError), res) d.addBoth(_done) return d hunk ./src/allmydata/test/test_encode.py 805 d = self.send_and_recover((4,8,10), bucket_modes=modemap) def _done(res): self.failUnless(isinstance(res, Failure)) - self.failUnless(res.check(NotEnoughSharesError)) + self.failUnless(res.check(UploadHappinessError)) d.addBoth(_done) return d hunk ./src/allmydata/test/test_upload.py 13 import allmydata # for __full_version__ from allmydata import uri, monitor, client from allmydata.immutable import upload, encode -from allmydata.interfaces import FileTooLargeError, NoSharesError, \ - NotEnoughSharesError +from allmydata.interfaces import FileTooLargeError, UploadHappinessError from allmydata.util.assertutil import precondition from allmydata.util.deferredutil import DeferredListShouldSucceed from no_network import GridTestMixin hunk ./src/allmydata/test/test_upload.py 402 def test_first_error_all(self): self.make_node("first-fail") - d = self.shouldFail(NoSharesError, "first_error_all", + d = self.shouldFail(UploadHappinessError, "first_error_all", "peer selection failed", upload_data, self.u, DATA) def _check((f,)): hunk ./src/allmydata/test/test_upload.py 434 def test_second_error_all(self): self.make_node("second-fail") - d = self.shouldFail(NotEnoughSharesError, "second_error_all", + d = self.shouldFail(UploadHappinessError, "second_error_all", "peer selection failed", upload_data, self.u, DATA) def _check((f,)): hunk ./src/allmydata/test/test_upload.py 452 self.u.parent = self.node def _should_fail(self, f): - self.failUnless(isinstance(f, Failure) and f.check(NoSharesError), f) + self.failUnless(isinstance(f, Failure) and f.check(UploadHappinessError), f) def test_data_large(self): data = DATA hunk ./src/allmydata/test/test_upload.py 817 # These parameters are unsatisfiable with the client that we've made # -- we'll use them to test that the semnatics work correctly. self.set_encoding_parameters(k=3, happy=5, n=10) - d = self.shouldFail(NotEnoughSharesError, "test_happy_semantics", + d = self.shouldFail(UploadHappinessError, "test_happy_semantics", "shares could only be placed on 2 servers " "(5 were requested)", self.u.upload, DATA) hunk ./src/allmydata/test/test_upload.py 888 _prepare()) # Uploading data should fail d.addCallback(lambda client: - self.shouldFail(NotEnoughSharesError, "test_happy_semantics", + self.shouldFail(UploadHappinessError, "test_happy_semantics", "shares could only be placed on 2 servers " "(4 were requested)", client.upload, upload.Data("data" * 10000, hunk ./src/allmydata/test/test_upload.py 918 d.addCallback(lambda ign: _prepare2()) d.addCallback(lambda client: - self.shouldFail(NotEnoughSharesError, "test_happy_sematics", + self.shouldFail(UploadHappinessError, "test_happy_sematics", "shares could only be placed on 2 servers " "(3 were requested)", client.upload, upload.Data("data" * 10000, hunk ./src/allmydata/test/test_upload.py 1124 d.addCallback(_do_server_setup) d.addCallback(_remove_server) d.addCallback(lambda ign: - self.shouldFail(NotEnoughSharesError, + self.shouldFail(UploadHappinessError, "test_dropped_servers_in_encoder", "lost too many servers during upload " "(still have 3, want 4)", hunk ./src/allmydata/test/test_upload.py 1151 d.addCallback(_do_server_setup_2) d.addCallback(_remove_server) d.addCallback(lambda ign: - self.shouldFail(NotEnoughSharesError, + self.shouldFail(UploadHappinessError, "test_dropped_servers_in_encoder", "lost too many servers during upload " "(still have 3, want 4)", hunk ./src/allmydata/test/test_upload.py 1275 self.failUnless(upload.should_add_server(shares, "server1", 1)) + def test_exception_messages_during_peer_selection(self): + # server 1: readonly, no shares + # server 2: readonly, no shares + # server 3: readonly, no shares + # server 4: readonly, no shares + # server 5: readonly, no shares + # This will fail, but we want to make sure that the log messages + # are informative about why it has failed. + self.basedir = self.mktemp() + d = self._setup_and_upload() + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1, readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2, readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3, readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=4, readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=5, readonly=True)) + d.addCallback(lambda ign: + self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) + def _reset_encoding_parameters(ign): + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = 4 + return client + d.addCallback(_reset_encoding_parameters) + d.addCallback(lambda client: + self.shouldFail(UploadHappinessError, "test_selection_exceptions", + "peer selection failed for : placed 0 shares out of 10 " + "total (10 homeless), want to place on 4 servers," + " sent 5 queries to 5 peers, 0 queries placed " + "some shares, 5 placed none " + "(of which 5 placed none due to the server being " + "full and 0 placed none due to an error)", + client.upload, + upload.Data("data" * 10000, convergence=""))) + + + # server 1: readonly, no shares + # server 2: broken, no shares + # server 3: readonly, no shares + # server 4: readonly, no shares + # server 5: readonly, no shares + def _reset(ign): + self.basedir = self.mktemp() + d.addCallback(_reset) + d.addCallback(lambda ign: + self._setup_and_upload()) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1, readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2)) + def _break_server_2(ign): + server = self.g.servers_by_number[2].my_nodeid + # We have to break the server in servers_by_id, + # because the ones in servers_by_number isn't wrapped, + # and doesn't look at its broken attribute + self.g.servers_by_id[server].broken = True + d.addCallback(_break_server_2) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3, readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=4, readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=5, readonly=True)) + d.addCallback(lambda ign: + self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) + def _reset_encoding_parameters(ign): + client = self.g.clients[0] + client.DEFAULT_ENCODING_PARAMETERS['happy'] = 4 + return client + d.addCallback(_reset_encoding_parameters) + d.addCallback(lambda client: + self.shouldFail(UploadHappinessError, "test_selection_exceptions", + "peer selection failed for : placed 0 shares out of 10 " + "total (10 homeless), want to place on 4 servers," + " sent 5 queries to 5 peers, 0 queries placed " + "some shares, 5 placed none " + "(of which 4 placed none due to the server being " + "full and 1 placed none due to an error)", + client.upload, + upload.Data("data" * 10000, convergence=""))) + return d + + def _set_up_nodes_extra_config(self, clientdir): cfgfn = os.path.join(clientdir, "tahoe.cfg") oldcfg = open(cfgfn, "r").read() } [Replace "UploadHappinessError" with "UploadUnhappinessError" in tests. Kevan Carstensen **20091205043453 Ignore-this: 83f4bc50c697d21b5f4e2a4cd91862ca ] { replace ./src/allmydata/test/test_encode.py [A-Za-z_0-9] UploadHappinessError UploadUnhappinessError replace ./src/allmydata/test/test_upload.py [A-Za-z_0-9] UploadHappinessError UploadUnhappinessError } [Alter various unit tests to work with the new happy behavior Kevan Carstensen **20100107181325 Ignore-this: 132032bbf865e63a079f869b663be34a ] { hunk ./src/allmydata/test/common.py 915 # We need multiple segments to test crypttext hash trees that are # non-trivial (i.e. they have more than just one hash in them). cl0.DEFAULT_ENCODING_PARAMETERS['max_segment_size'] = 12 + # Tests that need to test servers of happiness using this should + # set their own value for happy -- the default (7) breaks stuff. + cl0.DEFAULT_ENCODING_PARAMETERS['happy'] = 1 d2 = cl0.upload(immutable.upload.Data(TEST_DATA, convergence="")) def _after_upload(u): filecap = u.uri hunk ./src/allmydata/test/test_checker.py 283 self.basedir = "checker/AddLease/875" self.set_up_grid(num_servers=1) c0 = self.g.clients[0] + c0.DEFAULT_ENCODING_PARAMETERS['happy'] = 1 self.uris = {} DATA = "data" * 100 d = c0.upload(Data(DATA, convergence="")) hunk ./src/allmydata/test/test_system.py 93 d = self.set_up_nodes() def _check_connections(res): for c in self.clients: + c.DEFAULT_ENCODING_PARAMETERS['happy'] = 5 all_peerids = c.get_storage_broker().get_all_serverids() self.failUnlessEqual(len(all_peerids), self.numclients) sb = c.storage_broker hunk ./src/allmydata/test/test_system.py 205 add_to_sparent=True)) def _added(extra_node): self.extra_node = extra_node + self.extra_node.DEFAULT_ENCODING_PARAMETERS['happy'] = 5 d.addCallback(_added) HELPER_DATA = "Data that needs help to upload" * 1000 hunk ./src/allmydata/test/test_system.py 705 self.basedir = "system/SystemTest/test_filesystem" self.data = LARGE_DATA d = self.set_up_nodes(use_stats_gatherer=True) + def _new_happy_semantics(ign): + for c in self.clients: + c.DEFAULT_ENCODING_PARAMETERS['happy'] = 1 + d.addCallback(_new_happy_semantics) d.addCallback(self._test_introweb) d.addCallback(self.log, "starting publish") d.addCallback(self._do_publish1) hunk ./src/allmydata/test/test_system.py 1129 d.addCallback(self.failUnlessEqual, "new.txt contents") # and again with something large enough to use multiple segments, # and hopefully trigger pauseProducing too + def _new_happy_semantics(ign): + for c in self.clients: + # these get reset somewhere? Whatever. + c.DEFAULT_ENCODING_PARAMETERS['happy'] = 1 + d.addCallback(_new_happy_semantics) d.addCallback(lambda res: self.PUT(public + "/subdir3/big.txt", "big" * 500000)) # 1.5MB d.addCallback(lambda res: self.GET(public + "/subdir3/big.txt")) hunk ./src/allmydata/test/test_upload.py 178 class FakeClient: DEFAULT_ENCODING_PARAMETERS = {"k":25, - "happy": 75, + "happy": 25, "n": 100, "max_segment_size": 1*MiB, } hunk ./src/allmydata/test/test_upload.py 316 data = self.get_data(SIZE_LARGE) segsize = int(SIZE_LARGE / 2.5) # we want 3 segments, since that's not a power of two - self.set_encoding_parameters(25, 75, 100, segsize) + self.set_encoding_parameters(25, 25, 100, segsize) d = upload_data(self.u, data) d.addCallback(extract_uri) d.addCallback(self._check_large, SIZE_LARGE) hunk ./src/allmydata/test/test_upload.py 395 def test_first_error(self): mode = dict([(0,"good")] + [(i,"first-fail") for i in range(1,10)]) self.make_node(mode) + self.set_encoding_parameters(k=25, happy=1, n=50) d = upload_data(self.u, DATA) d.addCallback(extract_uri) d.addCallback(self._check_large, SIZE_LARGE) hunk ./src/allmydata/test/test_upload.py 513 self.make_client() data = self.get_data(SIZE_LARGE) - self.set_encoding_parameters(50, 75, 100) + # if there are 50 peers, then happy needs to be <= 50 + self.set_encoding_parameters(50, 50, 100) d = upload_data(self.u, data) d.addCallback(extract_uri) d.addCallback(self._check_large, SIZE_LARGE) hunk ./src/allmydata/test/test_upload.py 560 self.make_client() data = self.get_data(SIZE_LARGE) - self.set_encoding_parameters(100, 150, 200) + # if there are 50 peers, then happy should be no more than 50 if + # we want this to work. + self.set_encoding_parameters(100, 50, 200) d = upload_data(self.u, data) d.addCallback(extract_uri) d.addCallback(self._check_large, SIZE_LARGE) hunk ./src/allmydata/test/test_upload.py 580 self.make_client(3) data = self.get_data(SIZE_LARGE) - self.set_encoding_parameters(3, 5, 10) + self.set_encoding_parameters(3, 3, 10) d = upload_data(self.u, data) d.addCallback(extract_uri) d.addCallback(self._check_large, SIZE_LARGE) hunk ./src/allmydata/test/test_web.py 3581 self.basedir = "web/Grid/exceptions" self.set_up_grid(num_clients=1, num_servers=2) c0 = self.g.clients[0] + c0.DEFAULT_ENCODING_PARAMETERS['happy'] = 2 self.fileurls = {} DATA = "data" * 100 d = c0.create_dirnode() } [Revisions of the #778 tests, per reviewers' comments Kevan Carstensen **20100118210238 Ignore-this: 6c0d9de3b5d6d7965540df8d7c79a5df - Fix comments and confusing naming. - Add tests for the new error messages suggested by David-Sarah and Zooko. - Alter existing tests for new error messages. - Make sure that the tests continue to work with the trunk. - Add a test for a mutual disjointedness assertion that I added to upload.servers_of_happiness. ] { hunk ./src/allmydata/test/test_encode.py 462 def _ready(res): k,happy,n = e.get_param("share_counts") assert n == NUM_SHARES # else we'll be completely confused - all_peers = [] + servermap = {} for shnum in range(NUM_SHARES): mode = bucket_modes.get(shnum, "good") peer = FakeBucketReaderWriterProxy(mode) hunk ./src/allmydata/test/test_upload.py 706 num_segments = encoder.get_param("num_segments") d = selector.get_shareholders(broker, sh, storage_index, share_size, block_size, num_segments, - 10, 4) + 10, 3, 4) def _have_shareholders((used_peers, already_peers)): assert servers_to_break <= len(used_peers) for index in xrange(servers_to_break): hunk ./src/allmydata/test/test_upload.py 762 self.failUnless((share_number, ss.my_nodeid, new_share_location) in shares) + def _setup_grid(self): + """ + I set up a NoNetworkGrid with a single server and client. + """ + self.set_up_grid(num_clients=1, num_servers=1) def _setup_and_upload(self): """ hunk ./src/allmydata/test/test_upload.py 774 upload a file to it, store its uri in self.uri, and store its sharedata in self.shares. """ - self.set_up_grid(num_clients=1, num_servers=1) + self._setup_grid() client = self.g.clients[0] client.DEFAULT_ENCODING_PARAMETERS['happy'] = 1 data = upload.Data("data" * 10000, convergence="") hunk ./src/allmydata/test/test_upload.py 812 def _setUp(self, ns): - # Used by test_happy_semantics and test_prexisting_share_behavior + # Used by test_happy_semantics and test_preexisting_share_behavior # to set up the grid. self.node = FakeClient(mode="good", num_servers=ns) self.u = upload.Uploader() hunk ./src/allmydata/test/test_upload.py 823 def test_happy_semantics(self): self._setUp(2) DATA = upload.Data("kittens" * 10000, convergence="") - # These parameters are unsatisfiable with the client that we've made - # -- we'll use them to test that the semnatics work correctly. + # These parameters are unsatisfiable with only 2 servers. self.set_encoding_parameters(k=3, happy=5, n=10) d = self.shouldFail(UploadUnhappinessError, "test_happy_semantics", hunk ./src/allmydata/test/test_upload.py 826 - "shares could only be placed on 2 servers " - "(5 were requested)", + "shares could only be placed or found on 2 " + "server(s). We were asked to place shares on " + "at least 5 server(s) such that any 3 of them " + "have enough shares to recover the file", self.u.upload, DATA) # Let's reset the client to have 10 servers d.addCallback(lambda ign: hunk ./src/allmydata/test/test_upload.py 834 self._setUp(10)) - # These parameters are satisfiable with the client we've made. + # These parameters are satisfiable with 10 servers. d.addCallback(lambda ign: self.set_encoding_parameters(k=3, happy=5, n=10)) # this should work hunk ./src/allmydata/test/test_upload.py 844 # (this is less than n, but more than h) d.addCallback(lambda ign: self._setUp(7)) - # These encoding parameters should still be satisfiable with our - # client setup + # These parameters are satisfiable with 7 servers. d.addCallback(lambda ign: self.set_encoding_parameters(k=3, happy=5, n=10)) # This, then, should work. hunk ./src/allmydata/test/test_upload.py 862 # # The scenario in comment:52 proposes that we have a layout # like: - # server 1: share 1 - # server 2: share 1 - # server 3: share 1 - # server 4: shares 2 - 10 + # server 0: shares 1 - 9 + # server 1: share 0 + # server 2: share 0 + # server 3: share 0 # To get access to the shares, we will first upload to one hunk ./src/allmydata/test/test_upload.py 867 - # server, which will then have shares 1 - 10. We'll then + # server, which will then have shares 0 - 9. We'll then # add three new servers, configure them to not accept any new hunk ./src/allmydata/test/test_upload.py 869 - # shares, then write share 1 directly into the serverdir of each. - # Then each of servers 1 - 3 will report that they have share 1, - # and will not accept any new share, while server 4 will report that - # it has shares 2 - 10 and will accept new shares. + # shares, then write share 0 directly into the serverdir of each, + # and then remove share 0 from server 0 in the same way. + # Then each of servers 1 - 3 will report that they have share 0, + # and will not accept any new share, while server 0 will report that + # it has shares 1 - 9 and will accept new shares. # We'll then set 'happy' = 4, and see that an upload fails # (as it should) d = self._setup_and_upload() hunk ./src/allmydata/test/test_upload.py 878 d.addCallback(lambda ign: - self._add_server_with_share(1, 0, True)) + self._add_server_with_share(server_number=1, share_number=0, + readonly=True)) d.addCallback(lambda ign: hunk ./src/allmydata/test/test_upload.py 881 - self._add_server_with_share(2, 0, True)) + self._add_server_with_share(server_number=2, share_number=0, + readonly=True)) d.addCallback(lambda ign: hunk ./src/allmydata/test/test_upload.py 884 - self._add_server_with_share(3, 0, True)) + self._add_server_with_share(server_number=3, share_number=0, + readonly=True)) # Remove the first share from server 0. hunk ./src/allmydata/test/test_upload.py 887 - def _remove_share_0(): + def _remove_share_0_from_server_0(): share_location = self.shares[0][2] os.remove(share_location) d.addCallback(lambda ign: hunk ./src/allmydata/test/test_upload.py 891 - _remove_share_0()) + _remove_share_0_from_server_0()) # Set happy = 4 in the client. def _prepare(): client = self.g.clients[0] hunk ./src/allmydata/test/test_upload.py 902 # Uploading data should fail d.addCallback(lambda client: self.shouldFail(UploadUnhappinessError, "test_happy_semantics", - "shares could only be placed on 2 servers " - "(4 were requested)", + "shares could be placed or found on 4 server(s), " + "but they are not spread out evenly enough to " + "ensure that any 3 of these servers would have " + "enough shares to recover the file. " + "We were asked to place shares on at " + "least 4 servers such that any 3 of them have " + "enough shares to recover the file", client.upload, upload.Data("data" * 10000, convergence=""))) hunk ./src/allmydata/test/test_upload.py 931 readonly=True)) def _prepare2(): client = self.g.clients[0] - client.DEFAULT_ENCODING_PARAMETERS['happy'] = 3 + client.DEFAULT_ENCODING_PARAMETERS['happy'] = 4 return client d.addCallback(lambda ign: _prepare2()) hunk ./src/allmydata/test/test_upload.py 937 d.addCallback(lambda client: self.shouldFail(UploadUnhappinessError, "test_happy_sematics", - "shares could only be placed on 2 servers " - "(3 were requested)", + "shares could only be placed on 3 server(s) such " + "that any 3 of them have enough shares to recover " + "the file, but we were asked to use at least 4 " + "such servers.", client.upload, upload.Data("data" * 10000, convergence=""))) return d hunk ./src/allmydata/test/test_upload.py 954 def _change_basedir(ign): self.basedir = self.mktemp() _change_basedir(None) - d = self._setup_and_upload() - # We start by uploading all of the shares to one server (which has - # already been done above). + # We start by uploading all of the shares to one server. # Next, we'll add three new servers to our NoNetworkGrid. We'll add # one share from our initial upload to each of these. # The counterintuitive ordering of the share numbers is to deal with hunk ./src/allmydata/test/test_upload.py 960 # the permuting of these servers -- distributing the shares this # way ensures that the Tahoe2PeerSelector sees them in the order - # described above. + # described below. + d = self._setup_and_upload() d.addCallback(lambda ign: self._add_server_with_share(server_number=1, share_number=2)) d.addCallback(lambda ign: hunk ./src/allmydata/test/test_upload.py 973 # server 1: share 2 # server 2: share 0 # server 3: share 1 - # We want to change the 'happy' parameter in the client to 4. + # We change the 'happy' parameter in the client to 4. # The Tahoe2PeerSelector will see the peers permuted as: # 2, 3, 1, 0 # Ideally, a reupload of our original data should work. hunk ./src/allmydata/test/test_upload.py 986 client.upload(upload.Data("data" * 10000, convergence=""))) - # This scenario is basically comment:53, but with the order reversed; - # this means that the Tahoe2PeerSelector sees - # server 2: shares 1-10 - # server 3: share 1 - # server 1: share 2 - # server 4: share 3 + # This scenario is basically comment:53, but changed so that the + # Tahoe2PeerSelector sees the server with all of the shares before + # any of the other servers. + # The layout is: + # server 2: shares 0 - 9 + # server 3: share 0 + # server 1: share 1 + # server 4: share 2 + # The Tahoe2PeerSelector sees the peers permuted as: + # 2, 3, 1, 4 + # Note that server 0 has been replaced by server 4; this makes it + # easier to ensure that the last server seen by Tahoe2PeerSelector + # has only one share. d.addCallback(_change_basedir) d.addCallback(lambda ign: self._setup_and_upload()) hunk ./src/allmydata/test/test_upload.py 1022 d.addCallback(_reset_encoding_parameters) d.addCallback(lambda client: client.upload(upload.Data("data" * 10000, convergence=""))) + + # Try the same thing, but with empty servers after the first one # We want to make sure that Tahoe2PeerSelector will redistribute # shares as necessary, not simply discover an existing layout. hunk ./src/allmydata/test/test_upload.py 1027 + # The layout is: + # server 2: shares 0 - 9 + # server 3: empty + # server 1: empty + # server 4: empty d.addCallback(_change_basedir) d.addCallback(lambda ign: self._setup_and_upload()) hunk ./src/allmydata/test/test_upload.py 1127 def test_dropped_servers_in_encoder(self): + # The Encoder does its own "servers_of_happiness" check if it + # happens to lose a bucket during an upload (it assumes that + # the layout presented to it satisfies "servers_of_happiness" + # until a failure occurs) + # + # This test simulates an upload where servers break after peer + # selection, but before they are written to. def _set_basedir(ign=None): self.basedir = self.mktemp() _set_basedir() hunk ./src/allmydata/test/test_upload.py 1140 d = self._setup_and_upload(); # Add 5 servers def _do_server_setup(ign): - self._add_server_with_share(1) - self._add_server_with_share(2) - self._add_server_with_share(3) - self._add_server_with_share(4) - self._add_server_with_share(5) + self._add_server_with_share(server_number=1) + self._add_server_with_share(server_number=2) + self._add_server_with_share(server_number=3) + self._add_server_with_share(server_number=4) + self._add_server_with_share(server_number=5) d.addCallback(_do_server_setup) # remove the original server # (necessary to ensure that the Tahoe2PeerSelector will distribute hunk ./src/allmydata/test/test_upload.py 1153 server = self.g.servers_by_number[0] self.g.remove_server(server.my_nodeid) d.addCallback(_remove_server) - # This should succeed. + # This should succeed; we still have 4 servers, and the + # happiness of the upload is 4. d.addCallback(lambda ign: self._do_upload_with_broken_servers(1)) # Now, do the same thing over again, but drop 2 servers instead hunk ./src/allmydata/test/test_upload.py 1158 - # of 1. This should fail. + # of 1. This should fail, because servers_of_happiness is 4 and + # we can't satisfy that. d.addCallback(_set_basedir) d.addCallback(lambda ign: self._setup_and_upload()) hunk ./src/allmydata/test/test_upload.py 1201 return d - def test_servers_with_unique_shares(self): - # servers_with_unique_shares expects a dict of + def test_servers_of_happiness_utility_function(self): + # This function is concerned with the one-to-one or injective + # function upload.servers_of_happiness(); many + # issues that relate to other aspects of the servers of happiness + # behavior that happen to use this function are tested elsewhere. + # These tests exist to ensure that upload.servers_of_happiness doesn't + # under or overcount the happiness value for given inputs. + + # servers_of_happiness expects a dict of # shnum => peerid as a preexisting shares argument. test1 = { 1 : "server1", hunk ./src/allmydata/test/test_upload.py 1217 3 : "server3", 4 : "server4" } - unique_servers = upload.servers_with_unique_shares(test1) - self.failUnlessEqual(4, len(unique_servers)) - for server in ["server1", "server2", "server3", "server4"]: - self.failUnlessIn(server, unique_servers) + happy = upload.servers_of_happiness(test1) + self.failUnlessEqual(4, happy) test1[4] = "server1" hunk ./src/allmydata/test/test_upload.py 1220 - # Now there should only be 3 unique servers. - unique_servers = upload.servers_with_unique_shares(test1) - self.failUnlessEqual(3, len(unique_servers)) - for server in ["server1", "server2", "server3"]: - self.failUnlessIn(server, unique_servers) - # servers_with_unique_shares expects to receive some object with - # a peerid attribute. So we make a FakePeerTracker whose only - # job is to have a peerid attribute. + # We've added a duplicate server, so now servers_of_happiness + # should be 3 instead of 4. + happy = upload.servers_of_happiness(test1) + self.failUnlessEqual(3, happy) + # The second argument of servers_of_happiness should be a list of + # objects with peerid and buckets as attributes. In actual use, + # these will be PeerTracker instances, but for testing it is fine + # to make a FakePeerTracker whose job is to hold those instance + # variables to test that part. class FakePeerTracker: pass trackers = [] hunk ./src/allmydata/test/test_upload.py 1237 t.peerid = server t.buckets = [i] trackers.append(t) - # Recall that there are 3 unique servers in test1. Since none of - # those overlap with the ones in trackers, we should get 7 back - unique_servers = upload.servers_with_unique_shares(test1, set(trackers)) - self.failUnlessEqual(7, len(unique_servers)) - expected_servers = ["server" + str(i) for i in xrange(1, 9)] - expected_servers.remove("server4") - for server in expected_servers: - self.failUnlessIn(server, unique_servers) - # Now add an overlapping server to trackers. + # Recall that test1 is a server layout with servers_of_happiness = 3. + # Since there isn't any overlap between the server => share + # correspondences in test1 and those in trackers, the result here + # should be 7. + happy = upload.servers_of_happiness(test1, set(trackers)) + self.failUnlessEqual(7, happy) + # Now add an overlapping server to trackers. This is redundant, so it + # should not cause the previously reported happiness value to change. t = FakePeerTracker() t.peerid = "server1" t.buckets = [1] hunk ./src/allmydata/test/test_upload.py 1249 trackers.append(t) - unique_servers = upload.servers_with_unique_shares(test1, set(trackers)) - self.failUnlessEqual(7, len(unique_servers)) - for server in expected_servers: - self.failUnlessIn(server, unique_servers) + happy = upload.servers_of_happiness(test1, set(trackers)) + self.failUnlessEqual(7, happy) test = {} hunk ./src/allmydata/test/test_upload.py 1252 - unique_servers = upload.servers_with_unique_shares(test) - self.failUnlessEqual(0, len(test)) + happy = upload.servers_of_happiness(test) + self.failUnlessEqual(0, happy) + # Test a more substantial overlap between the trackers and the + # existing assignments. + test = { + 1 : 'server1', + 2 : 'server2', + 3 : 'server3', + 4 : 'server4', + } + trackers = [] + t = FakePeerTracker() + t.peerid = 'server5' + t.buckets = [4] + trackers.append(t) + t = FakePeerTracker() + t.peerid = 'server6' + t.buckets = [3, 5] + trackers.append(t) + # The value returned by upload.servers_of_happiness is the size + # of the domain of the one-to-one function that + # upload.servers_of_happiness makes between peerids and share numbers. + # It should make something like this: + # server 1: share 1 + # server 2: share 2 + # server 3: share 3 + # server 5: share 4 + # server 6: share 5 + # + # and, since there are 5 servers in the domain of this function, it + # should return 5. + happy = upload.servers_of_happiness(test, set(trackers)) + self.failUnlessEqual(5, happy) + # upload.servers_of_happiness assumes that the buckets attributes + # of the PeerTrackers in its used_peers argument are mutually + # disjoint; that is, they contain no shares in common. It should + # refuse to work at all if this is not the case. + t = FakePeerTracker() + t.peerid = "server7" + t.buckets = [3, 4, 5] + trackers.append(t) + self.shouldFail(AssertionError, + "test_servers_of_happiness_utility_function", + "", + upload.servers_of_happiness, test, trackers) def test_shares_by_server(self): hunk ./src/allmydata/test/test_upload.py 1355 def test_should_add_server(self): + # upload.should_add_server tests whether or not the addition of + # shnum => peerid mapping to the existing_shares dictionary + # would make the dictionary happier. shares = dict([(i, "server%d" % i) for i in xrange(10)]) hunk ./src/allmydata/test/test_upload.py 1359 - self.failIf(upload.should_add_server(shares, "server1", 4)) + + + # Attempting to add a shnum => peerid mapping for a shnum that + # isn't already in the dictionary should return true. + self.failUnless(upload.should_add_server(shares, "server11", 11)) + + + # Attempting to add an identical entry should return false + # (since the dictionary would be no happier than it already is) + self.failIf(upload.should_add_server(shares, "server1", 1)) + + + # shnum 1 maps to "server1", which occurs nowhere else + # in shares. Attempting to map shnum 1 to "server4", which occurs + # elsewhere in shares, should return false because the resulting + # dictionary will be less happy than it is now. + self.failIf(upload.should_add_server(shares, "server4", 1)) + + + # Now map shnum 4 to "server1", then check to see that + # upload.should_add_server thinks that mapping this to "server4" + # makes shares happier. shares[4] = "server1" self.failUnless(upload.should_add_server(shares, "server4", 4)) hunk ./src/allmydata/test/test_upload.py 1383 + + + # attempting to add a mapping to an empty dictionary should always + # make the dictionary happier, so this should return true. shares = {} self.failUnless(upload.should_add_server(shares, "server1", 1)) hunk ./src/allmydata/test/test_upload.py 1422 self.shouldFail(UploadUnhappinessError, "test_selection_exceptions", "peer selection failed for : placed 0 shares out of 10 " - "total (10 homeless), want to place on 4 servers," - " sent 5 queries to 5 peers, 0 queries placed " + "total (10 homeless), want to place shares on at " + "least 4 servers such that any 3 of them have " + "enough shares to recover the file, " + "sent 5 queries to 5 peers, 0 queries placed " "some shares, 5 placed none " "(of which 5 placed none due to the server being " "full and 0 placed none due to an error)", hunk ./src/allmydata/test/test_upload.py 1462 self._add_server_with_share(server_number=5, readonly=True)) d.addCallback(lambda ign: self.g.remove_server(self.g.servers_by_number[0].my_nodeid)) - def _reset_encoding_parameters(ign): + def _reset_encoding_parameters(ign, happy=4): client = self.g.clients[0] hunk ./src/allmydata/test/test_upload.py 1464 - client.DEFAULT_ENCODING_PARAMETERS['happy'] = 4 + client.DEFAULT_ENCODING_PARAMETERS['happy'] = happy return client d.addCallback(_reset_encoding_parameters) d.addCallback(lambda client: hunk ./src/allmydata/test/test_upload.py 1471 self.shouldFail(UploadUnhappinessError, "test_selection_exceptions", "peer selection failed for : placed 0 shares out of 10 " - "total (10 homeless), want to place on 4 servers," - " sent 5 queries to 5 peers, 0 queries placed " + "total (10 homeless), want to place shares on at " + "least 4 servers such that any 3 of them have " + "enough shares to recover the file, " + "sent 5 queries to 5 peers, 0 queries placed " "some shares, 5 placed none " "(of which 4 placed none due to the server being " "full and 1 placed none due to an error)", hunk ./src/allmydata/test/test_upload.py 1480 client.upload, upload.Data("data" * 10000, convergence=""))) + # server 0, server 1 = empty, accepting shares + # This should place all of the shares, but still fail with happy=4. + # We want to make sure that the exception message is worded correctly. + d.addCallback(_reset) + d.addCallback(lambda ign: + self._setup_grid()) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1)) + d.addCallback(_reset_encoding_parameters) + d.addCallback(lambda client: + self.shouldFail(UploadUnhappinessError, "test_selection_exceptions", + "shares could only be placed or found on 2 " + "server(s). We were asked to place shares on at " + "least 4 server(s) such that any 3 of them have " + "enough shares to recover the file.", + client.upload, upload.Data("data" * 10000, + convergence=""))) + # servers 0 - 4 = empty, accepting shares + # This too should place all the shares, and this too should fail, + # but since the effective happiness is more than the k encoding + # parameter, it should trigger a different error message than the one + # above. + d.addCallback(_reset) + d.addCallback(lambda ign: + self._setup_grid()) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=4)) + d.addCallback(_reset_encoding_parameters, happy=7) + d.addCallback(lambda client: + self.shouldFail(UploadUnhappinessError, "test_selection_exceptions", + "shares could only be placed on 5 server(s) such " + "that any 3 of them have enough shares to recover " + "the file, but we were asked to use at least 7 " + "such servers.", + client.upload, upload.Data("data" * 10000, + convergence=""))) + # server 0 : shares 0 - 9 + # servers 1 - 3: share 0, readonly + # This should place all of the shares, but fail with happy=4. + # Since the number of servers with shares is more than the number + # necessary to reconstitute the file, this will trigger a different + # error message than either of those above. + d.addCallback(_reset) + d.addCallback(lambda ign: + self._setup_and_upload()) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=1, share_number=0, + readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=2, share_number=0, + readonly=True)) + d.addCallback(lambda ign: + self._add_server_with_share(server_number=3, share_number=0, + readonly=True)) + d.addCallback(_reset_encoding_parameters, happy=7) + d.addCallback(lambda client: + self.shouldFail(UploadUnhappinessError, "test_selection_exceptions", + "shares could be placed or found on 4 server(s), " + "but they are not spread out evenly enough to " + "ensure that any 3 of these servers would have " + "enough shares to recover the file. We were asked " + "to place shares on at least 7 servers such that " + "any 3 of them have enough shares to recover the " + "file", + client.upload, upload.Data("data" * 10000, + convergence=""))) return d } Context: [tahoe_add_alias.py: minor refactoring Brian Warner **20100115064220 Ignore-this: 29910e81ad11209c9e493d65fd2dab9b ] [test_dirnode.py: reduce scope of a Client instance, suggested by Kevan. Brian Warner **20100115062713 Ignore-this: b35efd9e6027e43de6c6f509bfb4ccaa ] [test_provisioning: STAN is not always a list. Fix by David-Sarah Hopwood. Brian Warner **20100115014632 Ignore-this: 9989de7f1e00907706d2b63153138219 ] [web/directory.py mkdir-immutable: hush pyflakes, add TODO for #903 behavior Brian Warner **20100114222804 Ignore-this: 717cd3b9a1c8aeee76938c9641db7356 ] [hush pyflakes-0.4.0 warnings: slightly less-trivial fixes. Closes #900. Brian Warner **20100114221719 Ignore-this: f774f4637e256ad55502659413a811a8 This includes one fix (in test_web) which was testing the wrong thing. ] [hush pyflakes-0.4.0 warnings: remove trivial unused variables. For #900. Brian Warner **20100114221529 Ignore-this: e96106c8f1a99fbf93306fbfe9a294cf ] [tahoe add-alias/create-alias: don't corrupt non-newline-terminated alias Brian Warner **20100114210246 Ignore-this: 9c994792e53a85159d708760a9b1b000 file. Closes #741. ] [change docs and --help to use "grid" instead of "virtual drive": closes #892. Brian Warner **20100114201119 Ignore-this: a20d4a4dcc4de4e3b404ff72d40fc29b Thanks to David-Sarah Hopwood for the patch. ] [backupdb.txt: fix ST_CTIME reference Brian Warner **20100114194052 Ignore-this: 5a189c7a1181b07dd87f0a08ea31b6d3 ] [client.py: fix/update comments on KeyGenerator Brian Warner **20100113004226 Ignore-this: 2208adbb3fd6a911c9f44e814583cabd ] [Clean up log.err calls, for one of the issues in #889. Brian Warner **20100112013343 Ignore-this: f58455ce15f1fda647c5fb25d234d2db allmydata.util.log.err() either takes a Failure as the first positional argument, or takes no positional arguments and must be invoked in an exception handler. Fixed its signature to match both foolscap.logging.log.err and twisted.python.log.err . Included a brief unit test. ] [tidy up DeadReferenceError handling, ignore them in add_lease calls Brian Warner **20100112000723 Ignore-this: 72f1444e826fd0b9db6d318f89603c38 Stop checking separately for ConnectionDone/ConnectionLost, since those have been folded into DeadReferenceError since foolscap-0.3.1 . Write rrefutil.trap_deadref() in terms of rrefutil.trap_and_discard() to improve code coverage. ] [NEWS: improve "tahoe backup" notes, mention first-backup-after-upgrade duration Brian Warner **20100111190132 Ignore-this: 10347c590b3375964579ba6c2b0edb4f Thanks to Francois Deppierraz for the suggestion. ] [test_repairer: add (commented-out) test_each_byte, to see exactly what the Brian Warner **20100110203552 Ignore-this: 8e84277d5304752edeff052b97821815 Verifier misses The results (described in #819) match our expectations: it misses corruption in unused share fields and in most container fields (which are only visible to the storage server, not the client). 1265 bytes of a 2753 byte share (hosting a 56-byte file with an artifically small segment size) are unused, mostly in the unused tail of the overallocated UEB space (765 bytes), and the allocated-but-unwritten plaintext_hash_tree (480 bytes). ] [repairer: fix some wrong offsets in the randomized verifier tests, debugged by Brian zooko@zooko.com**20100110203721 Ignore-this: 20604a609db8706555578612c1c12feb fixes #819 ] [test_repairer: fix colliding basedir names, which caused test inconsistencies Brian Warner **20100110084619 Ignore-this: b1d56dd27e6ab99a7730f74ba10abd23 ] [repairer: add deterministic test for #819, mark as TODO zooko@zooko.com**20100110013619 Ignore-this: 4cb8bb30b25246de58ed2b96fa447d68 ] [contrib/fuse/runtests.py: Tolerate the tahoe CLI returning deprecation warnings francois@ctrlaltdel.ch**20100109175946 Ignore-this: 419c354d9f2f6eaec03deb9b83752aee Depending on the versions of external libraries such as Twisted of Foolscap, the tahoe CLI can display deprecation warnings on stdout. The tests should not interpret those warnings as a failure if the node is in fact correctly started. See http://allmydata.org/trac/tahoe/ticket/859 for an example of deprecation warnings. fixes #876 ] [contrib: fix fuse_impl_c to use new Python API zooko@zooko.com**20100109174956 Ignore-this: 51ca1ec7c2a92a0862e9b99e52542179 original patch by Thomas Delaet, fixed by François, reviewed by Brian, committed by me ] [docs: CREDITS: add David-Sarah to the CREDITS file zooko@zooko.com**20100109060435 Ignore-this: 896062396ad85f9d2d4806762632f25a ] [mutable/publish: don't loop() right away upon DeadReferenceError. Closes #877 Brian Warner **20100102220841 Ignore-this: b200e707b3f13aa8251981362b8a3e61 The bug was that a disconnected server could cause us to re-enter the initial loop() call, sending multiple queries to a single server, provoking an incorrect UCWE. To fix it, stall the loop() with an eventual.fireEventually() ] [immutable/checker.py: oops, forgot some imports. Also hush pyflakes. Brian Warner **20091229233909 Ignore-this: 4d61bd3f8113015a4773fd4768176e51 ] [mutable repair: return successful=False when numshares**20091229233746 Ignore-this: d881c3275ff8c8bee42f6a80ca48441e instead of weird errors. Closes #874 and #786. Previously, if the file had 0 shares, this would raise TypeError as it tried to call download_version(None). If the file had some shares but fewer than 'k', it would incorrectly raise MustForceRepairError. Added get_successful() to the IRepairResults API, to give repair() a place to report non-code-bug problems like this. ] [node.py/interfaces.py: minor docs fixes Brian Warner **20091229230409 Ignore-this: c86ad6342ef0f95d50639b4f99cd4ddf ] [NEWS: fix 1.4.1 announcement w.r.t. add-lease behavior in older releases Brian Warner **20091229230310 Ignore-this: bbbbb9c961f3bbcc6e5dbe0b1594822 ] [checker: don't let failures in add-lease affect checker results. Closes #875. Brian Warner **20091229230108 Ignore-this: ef1a367b93e4d01298c2b1e6ca59c492 Mutable servermap updates and the immutable checker, when run with add_lease=True, send both the do-you-have-block and add-lease commands in parallel, to avoid an extra round trip time. Many older servers have problems with add-lease and raise various exceptions, which don't generally matter. The client-side code was catching+ignoring some of them, but unrecognized exceptions were passed through to the DYHB code, concealing the DYHB results from the checker, making it think the server had no shares. The fix is to separate the code paths. Both commands are sent at the same time, but the errback path from add-lease is handled separately. Known exceptions are ignored, the others (both unknown-remote and all-local) are logged (log.WEIRD, which will trigger an Incident), but neither will affect the DYHB results. The add-lease message is sent first, and we know that the server handles them synchronously. So when the checker is done, we can be sure that all the add-lease messages have been retired. This makes life easier for unit tests. ] [test_cli: verify fix for "tahoe get" not creating empty file on error (#121) Brian Warner **20091227235444 Ignore-this: 6444d52413b68eb7c11bc3dfdc69c55f ] [addendum to "Fix 'tahoe ls' on files (#771)" Brian Warner **20091227232149 Ignore-this: 6dd5e25f8072a3153ba200b7fdd49491 tahoe_ls.py: tolerate missing metadata web/filenode.py: minor cleanups test_cli.py: test 'tahoe ls FILECAP' ] [Fix 'tahoe ls' on files (#771). Patch adapted from Kevan Carstensen. Brian Warner **20091227225443 Ignore-this: 8bf8c7b1cd14ea4b0ebd453434f4fe07 web/filenode.py: also serve edge metadata when using t=json on a DIRCAP/childname object. tahoe_ls.py: list file objects as if we were listing one-entry directories. Show edge metadata if we have it, which will be true when doing 'tahoe ls DIRCAP/filename' and false when doing 'tahoe ls FILECAP' ] [tahoe_get: don't create the output file on error. Closes #121. Brian Warner **20091227220404 Ignore-this: 58d5e793a77ec6e87d9394ade074b926 ] [webapi: don't accept zero-length childnames during traversal. Closes #358, #676. Brian Warner **20091227201043 Ignore-this: a9119dec89e1c7741f2289b0cad6497b This forbids operations that would implicitly create a directory with a zero-length (empty string) name, like what you'd get if you did "tahoe put local /oops/blah" (#358) or "POST /uri/CAP//?t=mkdir" (#676). The error message is fairly friendly too. Also added code to "tahoe put" to catch this error beforehand and suggest the correct syntax (i.e. without the leading slash). ] [CLI: send 'Accept:' header to ask for text/plain tracebacks. Closes #646. Brian Warner **20091227195828 Ignore-this: 44c258d4d4c7dac0ed58adb22f73331 The webapi has been looking for an Accept header since 1.4.0, but it treats a missing header as equal to */* (to honor RFC2616). This change finally modifies our CLI tools to ask for "text/plain, application/octet-stream", which seems roughly correct (we either want a plain-text traceback or error message, or an uninterpreted chunk of binary data to save to disk). Some day we'll figure out how JSON fits into this scheme. ] [Makefile: upload-tarballs: switch from xfer-client to flappclient, closes #350 Brian Warner **20091227163703 Ignore-this: 3beeecdf2ad9c2438ab57f0e33dcb357 I've also set up a new flappserver on source@allmydata.org to receive the tarballs. We still need to replace the gutsy buildslave (which is where the tarballs used to be generated+uploaded) and give it the new FURL. ] [misc/ringsim.py: make it deterministic, more detail about grid-is-full behavior Brian Warner **20091227024832 Ignore-this: a691cc763fb2e98a4ce1767c36e8e73f ] [misc/ringsim.py: tool to discuss #302 Brian Warner **20091226060339 Ignore-this: fc171369b8f0d97afeeb8213e29d10ed ] [docs: fix helper.txt to describe new config style zooko@zooko.com**20091224223522 Ignore-this: 102e7692dc414a4b466307f7d78601fe ] [docs/stats.txt: add TOC, notes about controlling gatherer's listening port Brian Warner **20091224202133 Ignore-this: 8eef63b0e18db5aa8249c2eafde02c05 Thanks to Jody Harris for the suggestions. ] [Add docs/stats.py, explaining Tahoe stats, the gatherer, and the munin plugins. Brian Warner **20091223052400 Ignore-this: 7c9eeb6e5644eceda98b59a67730ccd5 ] [more #859: avoid deprecation warning for unit tests too, hush pyflakes Brian Warner **20091215000147 Ignore-this: 193622e24d31077da825a11ed2325fd3 * factor maybe-import-sha logic into util.hashutil ] [use hashlib module if available, thus avoiding a DeprecationWarning for importing the old sha module; fixes #859 zooko@zooko.com**20091214212703 Ignore-this: 8d0f230a4bf8581dbc1b07389d76029c ] [docs: reflow architecture.txt to 78-char lines zooko@zooko.com**20091208232943 Ignore-this: 88f55166415f15192e39407815141f77 ] [docs: update the about.html a little zooko@zooko.com**20091208212737 Ignore-this: 3fe2d9653c6de0727d3e82bd70f2a8ed ] [docs: remove obsolete doc file "codemap.txt" zooko@zooko.com**20091113163033 Ignore-this: 16bc21a1835546e71d1b344c06c61ebb I started to update this to reflect the current codebase, but then I thought (a) nobody seemed to notice that it hasn't been updated since December 2007, and (b) it will just bit-rot again, so I'm removing it. ] [mutable/retrieve.py: stop reaching into private MutableFileNode attributes Brian Warner **20091208172921 Ignore-this: 61e548798c1105aed66a792bf26ceef7 ] [mutable/servermap.py: stop reaching into private MutableFileNode attributes Brian Warner **20091208172608 Ignore-this: b40a6b62f623f9285ad96fda139c2ef2 ] [mutable/servermap.py: oops, query N+e servers in MODE_WRITE, not k+e Brian Warner **20091208171156 Ignore-this: 3497f4ab70dae906759007c3cfa43bc under normal conditions, this wouldn't cause any problems, but if the shares are really sparse (perhaps because new servers were added), then file-modifies might stop looking too early and leave old shares in place ] [control.py: fix speedtest: use download_best_version (not read) on mutable nodes Brian Warner **20091207060512 Ignore-this: 7125eabfe74837e05f9291dd6414f917 ] [FTP-and-SFTP.txt: fix ssh-keygen pointer Brian Warner **20091207052803 Ignore-this: bc2a70ee8c58ec314e79c1262ccb22f7 ] [setup: ignore _darcs in the "test-clean" test and make the "clean" step remove all .egg's in the root dir zooko@zooko.com**20091206184835 Ignore-this: 6066bd160f0db36d7bf60aba405558d2 ] [remove MutableFileNode.download(), prefer download_best_version() instead Brian Warner **20091201225438 Ignore-this: 5733eb373a902063e09fd52cc858dec0 ] [Simplify immutable download API: use just filenode.read(consumer, offset, size) Brian Warner **20091201225330 Ignore-this: bdedfb488ac23738bf52ae6d4ab3a3fb * remove Downloader.download_to_data/download_to_filename/download_to_filehandle * remove download.Data/FileName/FileHandle targets * remove filenode.download/download_to_data/download_to_filename methods * leave Downloader.download (the whole Downloader will go away eventually) * add util.consumer.MemoryConsumer/download_to_data, for convenience (this is mostly used by unit tests, but it gets used by enough non-test code to warrant putting it in allmydata.util) * update tests * removes about 180 lines of code. Yay negative code days! Overall plan is to rewrite immutable/download.py and leave filenode.read() as the sole read-side API. ] [server.py: undo my bogus 'correction' of David-Sarah's comment fix Brian Warner **20091201024607 Ignore-this: ff4bb58f6a9e045b900ac3a89d6f506a and move it to a better line ] [Implement more coherent behavior when copying with dircaps/filecaps (closes #761). Patch by Kevan Carstensen. "Brian Warner "**20091130211009] [storage.py: update comment "Brian Warner "**20091130195913] [storage server: detect disk space usage on Windows too (fixes #637) david-sarah@jacaranda.org**20091121055644 Ignore-this: 20fb30498174ce997befac7701fab056 ] [make status of finished operations consistently "Finished" david-sarah@jacaranda.org**20091121061543 Ignore-this: 97d483e8536ccfc2934549ceff7055a3 ] [NEWS: update with all user-visible changes since the last release Brian Warner **20091127224217 Ignore-this: 741da6cd928e939fb6d21a61ea3daf0b ] [update "tahoe backup" docs, and webapi.txt's mkdir-with-children Brian Warner **20091127055900 Ignore-this: defac1fb9a2335b0af3ef9dbbcc67b7e ] [Add dirnodes to backupdb and "tahoe backup", closes #606. Brian Warner **20091126234257 Ignore-this: fa88796fcad1763c6a2bf81f56103223 * backups now share dirnodes with any previous backup, in any location, so renames and moves are handled very efficiently * "tahoe backup" no longer bothers reading the previous snapshot * if you switch grids, you should delete ~/.tahoe/private/backupdb.sqlite, to force new uploads of all files and directories ] [webapi: fix t=check for DIR2-LIT (i.e. empty immutable directories) Brian Warner **20091126232731 Ignore-this: 8513c890525c69c1eca0e80d53a231f8 ] [PipelineError: fix str() on python2.4 . Closes #842. Brian Warner **20091124212512 Ignore-this: e62c92ea9ede2ab7d11fe63f43b9c942 ] [test_uri.py: s/NewDirnode/Dirnode/ , now that they aren't "new" anymore Brian Warner **20091120075553 Ignore-this: 61c8ef5e45a9d966873a610d8349b830 ] [interface name cleanups: IFileNode, IImmutableFileNode, IMutableFileNode Brian Warner **20091120075255 Ignore-this: e3d193c229e2463e1d0b0c92306de27f The proper hierarchy is: IFilesystemNode +IFileNode ++IMutableFileNode ++IImmutableFileNode +IDirectoryNode Also expand test_client.py (NodeMaker) to hit all IFilesystemNode types. ] [class name cleanups: s/FileNode/ImmutableFileNode/ Brian Warner **20091120072239 Ignore-this: 4b3218f2d0e585c62827e14ad8ed8ac1 also fix test/bench_dirnode.py for recent dirnode changes ] [Use DIR-IMM and t=mkdir-immutable for "tahoe backup", for #828 Brian Warner **20091118192813 Ignore-this: a4720529c9bc6bc8b22a3d3265925491 ] [web/directory.py: use "DIR-IMM" to describe immutable directories, not DIR-RO Brian Warner **20091118191832 Ignore-this: aceafd6ab4bf1cc0c2a719ef7319ac03 ] [web/info.py: hush pyflakes Brian Warner **20091118191736 Ignore-this: edc5f128a2b8095fb20686a75747c8 ] [make get_size/get_current_size consistent for all IFilesystemNode classes Brian Warner **20091118191624 Ignore-this: bd3449cf96e4827abaaf962672c1665a * stop caching most_recent_size in dirnode, rely upon backing filenode for it * start caching most_recent_size in MutableFileNode * return None when you don't know, not "?" * only render None as "?" in the web "more info" page * add get_size/get_current_size to UnknownNode ] [ImmutableDirectoryURIVerifier: fix verifycap handling Brian Warner **20091118164238 Ignore-this: 6bba5c717b54352262eabca6e805d590 ] [Add t=mkdir-immutable to the webapi. Closes #607. Brian Warner **20091118070900 Ignore-this: 311e5fab9a5f28b9e8a28d3d08f3c0d * change t=mkdir-with-children to not use multipart/form encoding. Instead, the request body is all JSON. t=mkdir-immutable uses this format too. * make nodemaker.create_immutable_dirnode() get convergence from SecretHolder, but let callers override it * raise NotDeepImmutableError instead of using assert() * add mutable= argument to DirectoryNode.create_subdirectory(), default True ] [move convergence secret into SecretHolder, next to lease secret Brian Warner **20091118015444 Ignore-this: 312f85978a339f2d04deb5bcb8f511bc ] [nodemaker: implement immutable directories (internal interface), for #607 Brian Warner **20091112002233 Ignore-this: d09fccf41813fdf7e0db177ed9e5e130 * nodemaker.create_from_cap() now handles DIR2-CHK and DIR2-LIT * client.create_immutable_dirnode() is used to create them * no webapi yet ] [stop using IURI()/etc as an adapter Brian Warner **20091111224542 Ignore-this: 9611da7ea6a4696de2a3b8c08776e6e0 ] [clean up uri-vs-cap terminology, emphasize cap instances instead of URI strings Brian Warner **20091111222619 Ignore-this: 93626385f6e7f039ada71f54feefe267 * "cap" means a python instance which encapsulates a filecap/dircap (uri.py) * "uri" means a string with a "URI:" prefix * FileNode instances are created with (and retain) a cap instance, and generate uri strings on demand * .get_cap/get_readcap/get_verifycap/get_repaircap return cap instances * .get_uri/get_readonly_uri return uri strings * add filenode.download_to_filename() for control.py, should find a better way * use MutableFileNode.init_from_cap, not .init_from_uri * directory URI instances: use get_filenode_cap, not get_filenode_uri * update/cleanup bench_dirnode.py to match, add Makefile target to run it ] [add parser for immutable directory caps: DIR2-CHK, DIR2-LIT, DIR2-CHK-Verifier Brian Warner **20091104181351 Ignore-this: 854398cc7a75bada57fa97c367b67518 ] [wui: s/TahoeLAFS/Tahoe-LAFS/ zooko@zooko.com**20091029035050 Ignore-this: 901e64cd862e492ed3132bd298583c26 ] [tests: bump up the timeout on test_repairer to see if 120 seconds was too short for François's ARM box to do the test even when it was doing it right. zooko@zooko.com**20091027224800 Ignore-this: 95e93dc2e018b9948253c2045d506f56 ] [dirnode.pack_children(): add deep_immutable= argument Brian Warner **20091026162809 Ignore-this: d5a2371e47662c4bc6eff273e8181b00 This will be used by DIR2:CHK to enforce the deep-immutability requirement. ] [webapi: use t=mkdir-with-children instead of a children= arg to t=mkdir . Brian Warner **20091026011321 Ignore-this: 769cab30b6ab50db95000b6c5a524916 This is safer: in the earlier API, an old webapi server would silently ignore the initial children, and clients trying to set them would have to fetch the newly-created directory to discover the incompatibility. In the new API, clients using t=mkdir-with-children against an old webapi server will get a clear error. ] [nodemaker.create_new_mutable_directory: pack_children() in initial_contents= Brian Warner **20091020005118 Ignore-this: bd43c4eefe06fd32b7492bcb0a55d07e instead of creating an empty file and then adding the children later. This should speed up mkdir(initial_children) considerably, removing two roundtrips and an entire read-modify-write cycle, probably bringing it down to a single roundtrip. A quick test (against the volunteergrid) suggests a 30% speedup. test_dirnode: add new tests to enforce the restrictions that interfaces.py claims for create_new_mutable_directory(): no UnknownNodes, metadata dicts ] [test_dirnode.py: add tests of initial_children= args to client.create_dirnode Brian Warner **20091017194159 Ignore-this: 2e2da28323a4d5d815466387914abc1b and nodemaker.create_new_mutable_directory ] [update many dirnode interfaces to accept dict-of-nodes instead of dict-of-caps Brian Warner **20091017192829 Ignore-this: b35472285143862a856bf4b361d692f0 interfaces.py: define INodeMaker, document argument values, change create_new_mutable_directory() to take dict-of-nodes. Change dirnode.set_nodes() and dirnode.create_subdirectory() too. nodemaker.py: use INodeMaker, update create_new_mutable_directory() client.py: have create_dirnode() delegate initial_children= to nodemaker dirnode.py (Adder): take dict-of-nodes instead of list-of-nodes, which updates set_nodes() and create_subdirectory() web/common.py (convert_initial_children_json): create dict-of-nodes web/directory.py: same web/unlinked.py: same test_dirnode.py: update tests to match ] [dirnode.py: move pack_children() out to a function, for eventual use by others Brian Warner **20091017180707 Ignore-this: 6a823fb61f2c180fd38d6742d3196a7a ] [move dirnode.CachingDict to dictutil.AuxValueDict, generalize method names, Brian Warner **20091017180005 Ignore-this: b086933cf429df0fcea16a308d2640dd improve tests. Let dirnode _pack_children accept either dict or AuxValueDict. ] [test/common.py: update FakeMutableFileNode to new contents= callable scheme Brian Warner **20091013052154 Ignore-this: 62f00a76454a2190d1c8641c5993632f ] [The initial_children= argument to nodemaker.create_new_mutable_directory is Brian Warner **20091013031922 Ignore-this: 72e45317c21f9eb9ec3bd79bd4311f48 now enabled. ] [client.create_mutable_file(contents=) now accepts a callable, which is Brian Warner **20091013031232 Ignore-this: 3c89d2f50c1e652b83f20bd3f4f27c4b invoked with the new MutableFileNode and is supposed to return the initial contents. This can be used by e.g. a new dirnode which needs the filenode's writekey to encrypt its initial children. create_mutable_file() still accepts a bytestring too, or None for an empty file. ] [webapi: t=mkdir now accepts initial children, using the same JSON that t=json Brian Warner **20091013023444 Ignore-this: 574a46ed46af4251abf8c9580fd31ef7 emits. client.create_dirnode(initial_children=) now works. ] [replace dirnode.create_empty_directory() with create_subdirectory(), which Brian Warner **20091013021520 Ignore-this: 6b57cb51bcfcc6058d0df569fdc8a9cf takes an initial_children= argument ] [dirnode.set_children: change return value: fire with self instead of None Brian Warner **20091013015026 Ignore-this: f1d14e67e084e4b2a4e25fa849b0e753 ] [dirnode.set_nodes: change return value: fire with self instead of None Brian Warner **20091013014546 Ignore-this: b75b3829fb53f7399693f1c1a39aacae ] [dirnode.set_children: take a dict, not a list Brian Warner **20091013002440 Ignore-this: 540ce72ce2727ee053afaae1ff124e21 ] [dirnode.set_uri/set_children: change signature to take writecap+readcap Brian Warner **20091012235126 Ignore-this: 5df617b2d379a51c79148a857e6026b1 instead of a single cap. The webapi t=set_children call benefits too. ] [replace Client.create_empty_dirnode() with create_dirnode(), in anticipation Brian Warner **20091012224506 Ignore-this: cbdaa4266ecb3c6496ffceab4f95709d of adding initial_children= argument. Includes stubbed-out initial_children= support. ] [test_web.py: use a less-fake client, making test harness smaller Brian Warner **20091012222808 Ignore-this: 29e95147f8c94282885c65b411d100bb ] [webapi.txt: document t=set_children, other small edits Brian Warner **20091009200446 Ignore-this: 4d7e76b04a7b8eaa0a981879f778ea5d ] [Verifier: check the full cryptext-hash tree on each share. Removed .todos Brian Warner **20091005221849 Ignore-this: 6fb039c5584812017d91725e687323a5 from the last few test_repairer tests that were waiting on this. ] [Verifier: check the full block-hash-tree on each share Brian Warner **20091005214844 Ignore-this: 3f7ccf6d253f32340f1bf1da27803eee Removed the .todo from two test_repairer tests that check this. The only remaining .todos are on the three crypttext-hash-tree tests. ] [Verifier: check the full share-hash chain on each share Brian Warner **20091005213443 Ignore-this: 3d30111904158bec06a4eac22fd39d17 Removed the .todo from two test_repairer tests that check this. ] [test_repairer: rename Verifier test cases to be more precise and less verbose Brian Warner **20091005201115 Ignore-this: 64be7094e33338c7c2aea9387e138771 ] [immutable/checker.py: rearrange code a little bit, make it easier to follow Brian Warner **20091005200252 Ignore-this: 91cc303fab66faf717433a709f785fb5 ] [test/common.py: wrap docstrings to 80cols so I can read them more easily Brian Warner **20091005200143 Ignore-this: b180a3a0235cbe309c87bd5e873cbbb3 ] [immutable/download.py: wrap to 80cols, no functional changes Brian Warner **20091005192542 Ignore-this: 6b05fe3dc6d78832323e708b9e6a1fe ] [CHK-hashes.svg: cross out plaintext hashes, since we don't include Brian Warner **20091005010803 Ignore-this: bea2e953b65ec7359363aa20de8cb603 them (until we finish #453) ] [docs: a few licensing clarifications requested by Ubuntu zooko@zooko.com**20090927033226 Ignore-this: 749fc8c9aeb6dc643669854a3e81baa7 ] [setup: remove binary WinFUSE modules zooko@zooko.com**20090924211436 Ignore-this: 8aefc571d2ae22b9405fc650f2c2062 I would prefer to have just source code, or indications of what 3rd-party packages are required, under revision control, and have the build process generate o r acquire the binaries as needed. Also, having these in our release tarballs is interfering with getting Tahoe-LAFS uploaded into Ubuntu Karmic. (Technicall y, they would accept binary modules as long as they came with the accompanying source so that they could satisfy their obligations under GPL2+ and TGPPL1+, bu t it is easier for now to remove the binaries from the source tree.) In this case, the binaries are from the tahoe-w32-client project: http://allmydata.org/trac/tahoe-w32-client , from which you can also get the source. ] [setup: remove binary _fusemodule.so 's zooko@zooko.com**20090924211130 Ignore-this: 74487bbe27d280762ac5dd5f51e24186 I would prefer to have just source code, or indications of what 3rd-party packages are required, under revision control, and have the build process generate or acquire the binaries as needed. Also, having these in our release tarballs is interfering with getting Tahoe-LAFS uploaded into Ubuntu Karmic. (Technically, they would accept binary modules as long as they came with the accompanying source so that they could satisfy their obligations under GPL2+ and TGPPL1+, but it is easier for now to remove the binaries from the source tree.) In this case, these modules come from the MacFUSE project: http://code.google.com/p/macfuse/ ] [doc: add a copy of LGPL2 for documentation purposes for ubuntu zooko@zooko.com**20090924054218 Ignore-this: 6a073b48678a7c84dc4fbcef9292ab5b ] [setup: remove a convenience copy of figleaf, to ease inclusion into Ubuntu Karmic Koala zooko@zooko.com**20090924053215 Ignore-this: a0b0c990d6e2ee65c53a24391365ac8d We need to carefully document the licence of figleaf in order to get Tahoe-LAFS into Ubuntu Karmic Koala. However, figleaf isn't really a part of Tahoe-LAFS per se -- this is just a "convenience copy" of a development tool. The quickest way to make Tahoe-LAFS acceptable for Karmic then, is to remove figleaf from the Tahoe-LAFS tarball itself. People who want to run figleaf on Tahoe-LAFS (as everyone should want) can install figleaf themselves. I haven't tested this -- there may be incompatibilities between upstream figleaf and the copy that we had here... ] [setup: shebang for misc/build-deb.py to fail quickly zooko@zooko.com**20090819135626 Ignore-this: 5a1b893234d2d0bb7b7346e84b0a6b4d Without this patch, when I ran "chmod +x ./misc/build-deb.py && ./misc/build-deb.py" then it hung indefinitely. (I wonder what it was doing.) ] [docs: Shawn Willden grants permission for his contributions under GPL2+|TGPPL1+ zooko@zooko.com**20090921164651 Ignore-this: ef1912010d07ff2ffd9678e7abfd0d57 ] [docs: Csaba Henk granted permission to license fuse.py under the same terms as Tahoe-LAFS itself zooko@zooko.com**20090921154659 Ignore-this: c61ba48dcb7206a89a57ca18a0450c53 ] [setup: mark setup.py as having utf-8 encoding in it zooko@zooko.com**20090920180343 Ignore-this: 9d3850733700a44ba7291e9c5e36bb91 ] [doc: licensing cleanups zooko@zooko.com**20090920171631 Ignore-this: 7654f2854bf3c13e6f4d4597633a6630 Use nice utf-8 © instead of "(c)". Remove licensing statements on utility modules that have been assigned to allmydata.com by their original authors. (Nattraverso was not assigned to allmydata.com -- it was LGPL'ed -- but I checked and src/allmydata/util/iputil.py was completely rewritten and doesn't contain any line of code from nattraverso.) Add notes to misc/debian/copyright about licensing on files that aren't just allmydata.com-licensed. ] [build-deb.py: run darcsver early, otherwise we get the wrong version later on Brian Warner **20090918033620 Ignore-this: 6635c5b85e84f8aed0d8390490c5392a ] [new approach for debian packaging, sharing pieces across distributions. Still experimental, still only works for sid. warner@lothar.com**20090818190527 Ignore-this: a75eb63db9106b3269badbfcdd7f5ce1 ] [new experimental deb-packaging rules. Only works for sid so far. Brian Warner **20090818014052 Ignore-this: 3a26ad188668098f8f3cc10a7c0c2f27 ] [setup.py: read _version.py and pass to setup(version=), so more commands work Brian Warner **20090818010057 Ignore-this: b290eb50216938e19f72db211f82147e like "setup.py --version" and "setup.py --fullname" ] [test/check_speed.py: fix shbang line Brian Warner **20090818005948 Ignore-this: 7f3a37caf349c4c4de704d0feb561f8d ] [setup: remove bundled version of darcsver-1.2.1 zooko@zooko.com**20090816233432 Ignore-this: 5357f26d2803db2d39159125dddb963a That version of darcsver emits a scary error message when the darcs executable or the _darcs subdirectory is not found. This error is hidden (unless the --loud option is passed) in darcsver >= 1.3.1. Fixes #788. ] [de-Service-ify Helper, pass in storage_broker and secret_holder directly. Brian Warner **20090815201737 Ignore-this: 86b8ac0f90f77a1036cd604dd1304d8b This makes it more obvious that the Helper currently generates leases with the Helper's own secrets, rather than getting values from the client, which is arguably a bug that will likely be resolved with the Accounting project. ] [immutable.Downloader: pass StorageBroker to constructor, stop being a Service Brian Warner **20090815192543 Ignore-this: af5ab12dbf75377640a670c689838479 child of the client, access with client.downloader instead of client.getServiceNamed("downloader"). The single "Downloader" instance is scheduled for demolition anyways, to be replaced by individual filenode.download calls. ] [tests: double the timeout on test_runner.RunNode.test_introducer since feisty hit a timeout zooko@zooko.com**20090815160512 Ignore-this: ca7358bce4bdabe8eea75dedc39c0e67 I'm not sure if this is an actual timing issue (feisty is running on an overloaded VM if I recall correctly), or it there is a deeper bug. ] [stop making History be a Service, it wasn't necessary Brian Warner **20090815114415 Ignore-this: b60449231557f1934a751c7effa93cfe ] [Overhaul IFilesystemNode handling, to simplify tests and use POLA internally. Brian Warner **20090815112846 Ignore-this: 1db1b9c149a60a310228aba04c5c8e5f * stop using IURI as an adapter * pass cap strings around instead of URI instances * move filenode/dirnode creation duties from Client to new NodeMaker class * move other Client duties to KeyGenerator, SecretHolder, History classes * stop passing Client reference to dirnode/filenode constructors - pass less-powerful references instead, like StorageBroker or Uploader * always create DirectoryNodes by wrapping a filenode (mutable for now) * remove some specialized mock classes from unit tests Detailed list of changes (done one at a time, then merged together) always pass a string to create_node_from_uri(), not an IURI instance always pass a string to IFilesystemNode constructors, not an IURI instance stop using IURI() as an adapter, switch on cap prefix in create_node_from_uri() client.py: move SecretHolder code out to a separate class test_web.py: hush pyflakes client.py: move NodeMaker functionality out into a separate object LiteralFileNode: stop storing a Client reference immutable Checker: remove Client reference, it only needs a SecretHolder immutable Upload: remove Client reference, leave SecretHolder and StorageBroker immutable Repairer: replace Client reference with StorageBroker and SecretHolder immutable FileNode: remove Client reference mutable.Publish: stop passing Client mutable.ServermapUpdater: get StorageBroker in constructor, not by peeking into Client reference MutableChecker: reference StorageBroker and History directly, not through Client mutable.FileNode: removed unused indirection to checker classes mutable.FileNode: remove Client reference client.py: move RSA key generation into a separate class, so it can be passed to the nodemaker move create_mutable_file() into NodeMaker test_dirnode.py: stop using FakeClient mockups, use NoNetworkGrid instead. This simplifies the code, but takes longer to run (17s instead of 6s). This should come down later when other cleanups make it possible to use simpler (non-RSA) fake mutable files for dirnode tests. test_mutable.py: clean up basedir names client.py: move create_empty_dirnode() into NodeMaker dirnode.py: get rid of DirectoryNode.create remove DirectoryNode.init_from_uri, refactor NodeMaker for customization, simplify test_web's mock Client to match stop passing Client to DirectoryNode, make DirectoryNode.create_with_mutablefile the normal DirectoryNode constructor, start removing client from NodeMaker remove Client from NodeMaker move helper status into History, pass History to web.Status instead of Client test_mutable.py: fix minor typo ] [docs: edits for docs/running.html from Sam Mason zooko@zooko.com**20090809201416 Ignore-this: 2207e80449943ebd4ed50cea57c43143 ] [docs: install.html: instruct Debian users to use this document and not to go find the DownloadDebianPackages page, ignore the warning at the top of it, and try it zooko@zooko.com**20090804123840 Ignore-this: 49da654f19d377ffc5a1eff0c820e026 http://allmydata.org/pipermail/tahoe-dev/2009-August/002507.html ] [docs: relnotes.txt: reflow to 63 chars wide because google groups and some web forms seem to wrap to that zooko@zooko.com**20090802135016 Ignore-this: 53b1493a0491bc30fb2935fad283caeb ] [docs: about.html: fix English usage noticed by Amber zooko@zooko.com**20090802050533 Ignore-this: 89965c4650f9bd100a615c401181a956 ] [docs: fix mis-spelled word in about.html zooko@zooko.com**20090802050320 Ignore-this: fdfd0397bc7cef9edfde425dddeb67e5 ] [TAG allmydata-tahoe-1.5.0 zooko@zooko.com**20090802031303 Ignore-this: 94e5558e7225c39a86aae666ea00f166 ] Patch bundle hash: 97102a87b48059967bad705fa0fe9c405e6d292a