ooni-probe-cli/internal/tunnel/tor_test.go

361 lines
9.8 KiB
Go
Raw Normal View History

package tunnel
import (
"context"
"errors"
"fmt"
"net/url"
"os"
"path/filepath"
"strings"
"syscall"
"testing"
"github.com/cretz/bine/control"
"github.com/cretz/bine/tor"
)
// torCloser is used to mock a running tor process, which
// we abstract as a io.Closer in tor.go.
type torCloser struct {
counter int
}
// Close implements io.Closer.Close.
func (c *torCloser) Close() error {
c.counter++
return errors.New("mocked mocked mocked")
}
func TestTorTunnelNonNil(t *testing.T) {
closer := new(torCloser)
proxy := &url.URL{Scheme: "x", Host: "10.0.0.1:443"}
tun := &torTunnel{
bootstrapTime: 128,
instance: closer,
proxy: proxy,
}
if tun.BootstrapTime() != 128 {
t.Fatal("not the bootstrap time we expected")
}
if tun.SOCKS5ProxyURL() != proxy {
t.Fatal("not the url we expected")
}
tun.Stop()
if closer.counter != 1 {
t.Fatal("something went wrong while stopping the tunnel")
}
}
func TestTorWithCancelledContext(t *testing.T) {
ctx, cancel := context.WithCancel(context.Background())
cancel() // fail immediately
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
})
if !errors.Is(err, context.Canceled) {
t.Fatal("not the error we expected")
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestTorWithEmptyTunnelDir(t *testing.T) {
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "",
})
if !errors.Is(err, ErrEmptyTunnelDir) {
t.Fatal("not the error we expected")
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestTorBinaryNotFoundFailure(t *testing.T) {
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TorBinary: "/nonexistent/directory/tor",
TunnelDir: "testdata",
})
if !errors.Is(err, syscall.ENOENT) {
t.Fatal("not the error we expected", err)
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestTorStartFailure(t *testing.T) {
expected := errors.New("mocked error")
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
testExecabsLookPath: func(name string) (string, error) {
return "/usr/local/bin/tor", nil
},
testTorStart: func(ctx context.Context, conf *tor.StartConf) (*tor.Tor, error) {
return nil, expected
},
})
if !errors.Is(err, expected) {
t.Fatal("not the error we expected")
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
func TestTorGetProtocolInfoFailure(t *testing.T) {
expected := errors.New("mocked error")
ctx := context.Background()
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
testExecabsLookPath: func(name string) (string, error) {
return "/usr/local/bin/tor", nil
},
testTorStart: func(ctx context.Context, conf *tor.StartConf) (*tor.Tor, error) {
return &tor.Tor{}, nil
},
testTorProtocolInfo: func(tor *tor.Tor) (*control.ProtocolInfo, error) {
return nil, expected
},
})
if !errors.Is(err, expected) {
t.Fatal("not the error we expected")
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestTorEnableNetworkFailure(t *testing.T) {
expected := errors.New("mocked error")
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
testExecabsLookPath: func(name string) (string, error) {
return "/usr/local/bin/tor", nil
},
testTorStart: func(ctx context.Context, conf *tor.StartConf) (*tor.Tor, error) {
return &tor.Tor{}, nil
},
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
testTorProtocolInfo: func(tor *tor.Tor) (*control.ProtocolInfo, error) {
return &control.ProtocolInfo{}, nil
},
testTorEnableNetwork: func(ctx context.Context, tor *tor.Tor, wait bool) error {
return expected
},
})
if !errors.Is(err, expected) {
t.Fatal("not the error we expected")
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestTorGetInfoFailure(t *testing.T) {
expected := errors.New("mocked error")
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
testExecabsLookPath: func(name string) (string, error) {
return "/usr/local/bin/tor", nil
},
testTorStart: func(ctx context.Context, conf *tor.StartConf) (*tor.Tor, error) {
return &tor.Tor{}, nil
},
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
testTorProtocolInfo: func(tor *tor.Tor) (*control.ProtocolInfo, error) {
return &control.ProtocolInfo{}, nil
},
testTorEnableNetwork: func(ctx context.Context, tor *tor.Tor, wait bool) error {
return nil
},
testTorGetInfo: func(ctrl *control.Conn, keys ...string) ([]*control.KeyVal, error) {
return nil, expected
},
})
if !errors.Is(err, expected) {
t.Fatal("not the error we expected")
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestTorGetInfoInvalidNumberOfKeys(t *testing.T) {
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
testExecabsLookPath: func(name string) (string, error) {
return "/usr/local/bin/tor", nil
},
testTorStart: func(ctx context.Context, conf *tor.StartConf) (*tor.Tor, error) {
return &tor.Tor{}, nil
},
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
testTorProtocolInfo: func(tor *tor.Tor) (*control.ProtocolInfo, error) {
return &control.ProtocolInfo{}, nil
},
testTorEnableNetwork: func(ctx context.Context, tor *tor.Tor, wait bool) error {
return nil
},
testTorGetInfo: func(ctrl *control.Conn, keys ...string) ([]*control.KeyVal, error) {
return nil, nil
},
})
if !errors.Is(err, ErrTorUnableToGetSOCKSProxyAddress) {
t.Fatal("not the error we expected", err)
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestTorGetInfoInvalidKey(t *testing.T) {
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
testExecabsLookPath: func(name string) (string, error) {
return "/usr/local/bin/tor", nil
},
testTorStart: func(ctx context.Context, conf *tor.StartConf) (*tor.Tor, error) {
return &tor.Tor{}, nil
},
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
testTorProtocolInfo: func(tor *tor.Tor) (*control.ProtocolInfo, error) {
return &control.ProtocolInfo{}, nil
},
testTorEnableNetwork: func(ctx context.Context, tor *tor.Tor, wait bool) error {
return nil
},
testTorGetInfo: func(ctrl *control.Conn, keys ...string) ([]*control.KeyVal, error) {
return []*control.KeyVal{{}}, nil
},
})
if !errors.Is(err, ErrTorUnableToGetSOCKSProxyAddress) {
t.Fatal("not the error we expected")
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestTorGetInfoInvalidProxyType(t *testing.T) {
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
testExecabsLookPath: func(name string) (string, error) {
return "/usr/local/bin/tor", nil
},
testTorStart: func(ctx context.Context, conf *tor.StartConf) (*tor.Tor, error) {
return &tor.Tor{}, nil
},
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
testTorProtocolInfo: func(tor *tor.Tor) (*control.ProtocolInfo, error) {
return &control.ProtocolInfo{}, nil
},
testTorEnableNetwork: func(ctx context.Context, tor *tor.Tor, wait bool) error {
return nil
},
testTorGetInfo: func(ctrl *control.Conn, keys ...string) ([]*control.KeyVal, error) {
return []*control.KeyVal{{Key: "net/listeners/socks", Val: "127.0.0.1:9050"}}, nil
},
})
if err != nil {
t.Fatal(err)
}
if tun == nil {
t.Fatal("expected non-nil tunnel here")
}
}
func TestTorUnsupportedProxy(t *testing.T) {
ctx := context.Background()
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
tun, _, err := torStart(ctx, &Config{
Session: &MockableSession{},
TunnelDir: "testdata",
testExecabsLookPath: func(name string) (string, error) {
return "/usr/local/bin/tor", nil
},
testTorStart: func(ctx context.Context, conf *tor.StartConf) (*tor.Tor, error) {
return &tor.Tor{}, nil
},
feat(torsf): collect tor logs, select rendezvous method, count bytes (#683) This diff contains significant improvements over the previous implementation of the torsf experiment. We add support for configuring different rendezvous methods after the convo at https://github.com/ooni/probe/issues/2004. In doing that, I've tried to use a terminology that is consistent with the names being actually used by tor developers. In terms of what to do next, this diff basically instruments torsf to always rendezvous using domain fronting. Yet, it's also possible to change the rendezvous method from the command line, when using miniooni, which allows to experiment a bit more. In the same vein, by default we use a persistent tor datadir, but it's also possible to use a temporary datadir using the cmdline. Here's how a generic invocation of `torsf` looks like: ```bash ./miniooni -O DisablePersistentDatadir=true \ -O RendezvousMethod=amp \ -O DisableProgress=true \ torsf ``` (The default is `DisablePersistentDatadir=false` and `RendezvousMethod=domain_fronting`.) With this implementation, we can start measuring whether snowflake and tor together can boostrap, which seems the most important thing to focus on at the beginning. Understanding why the bootstrap most often does not converge with a temporary datadir on Android devices remains instead an open problem for now. (I'll also update the relevant issues or create new issues after commit this.) We also address some methodology improvements that were proposed in https://github.com/ooni/probe/issues/1686. Namely: 1. we record the tor version; 2. we include the bootstrap percentage by reading the logs; 3. we set the anomaly key correctly; 4. we measure the bytes send and received (by `tor` not by `snowflake`, since doing it for snowflake seems more complex at this stage). What remains to be done is the possibility of including Snowflake events into the measurement, which is not possible until the new improvements at common/event in snowflake.git are included into a tagged version of snowflake itself. (I'll make sure to mention this aspect to @cohosh in https://github.com/ooni/probe/issues/2004.)
2022-02-07 17:05:36 +01:00
testTorProtocolInfo: func(tor *tor.Tor) (*control.ProtocolInfo, error) {
return &control.ProtocolInfo{}, nil
},
testTorEnableNetwork: func(ctx context.Context, tor *tor.Tor, wait bool) error {
return nil
},
testTorGetInfo: func(ctrl *control.Conn, keys ...string) ([]*control.KeyVal, error) {
return []*control.KeyVal{{Key: "net/listeners/socks", Val: "unix:/foo/bar"}}, nil
},
})
if !errors.Is(err, ErrTorReturnedUnsupportedProxy) {
t.Fatal("not the error we expected")
}
if tun != nil {
t.Fatal("expected nil tunnel here")
}
}
func TestMaybeCleanupTunnelDir(t *testing.T) {
fakeTunDir := filepath.Join("testdata", "fake-tun-dir")
if err := os.RemoveAll(fakeTunDir); err != nil {
t.Fatal(err)
}
if err := os.MkdirAll(fakeTunDir, 0700); err != nil {
t.Fatal(err)
}
fakeData := []byte("deadbeef\n")
logfile := filepath.Join(fakeTunDir, "tor.log")
if err := os.WriteFile(logfile, fakeData, 0600); err != nil {
t.Fatal(err)
}
for idx := 0; idx < 3; idx++ {
filename := filepath.Join(fakeTunDir, fmt.Sprintf("torrc-%d", idx))
if err := os.WriteFile(filename, fakeData, 0600); err != nil {
t.Fatal(err)
}
filename = filepath.Join(fakeTunDir, fmt.Sprintf("control-port-%d", idx))
if err := os.WriteFile(filename, fakeData, 0600); err != nil {
t.Fatal(err)
}
filename = filepath.Join(fakeTunDir, fmt.Sprintf("antani-%d", idx))
if err := os.WriteFile(filename, fakeData, 0600); err != nil {
t.Fatal(err)
}
}
files, err := filepath.Glob(filepath.Join(fakeTunDir, "*"))
if err != nil {
t.Fatal(err)
}
if len(files) != 10 {
t.Fatal("unexpected number of files")
}
maybeCleanupTunnelDir(fakeTunDir, logfile)
files, err = filepath.Glob(filepath.Join(fakeTunDir, "*"))
if err != nil {
t.Fatal(err)
}
if len(files) != 3 {
t.Fatal("unexpected number of files")
}
expectPrefix := filepath.Join(fakeTunDir, "antani-")
for _, file := range files {
if !strings.HasPrefix(file, expectPrefix) {
t.Fatal("unexpected file name: ", file)
}
}
}