2021-02-02 12:05:47 +01:00
|
|
|
// Package geolocate implements IP lookup, resolver lookup, and geolocation.
|
|
|
|
package geolocate
|
|
|
|
|
|
|
|
import (
|
|
|
|
"context"
|
|
|
|
"fmt"
|
|
|
|
|
|
|
|
"github.com/ooni/probe-cli/v3/internal/engine/model"
|
fix(geolocate): no proxy when discovering our IP address (#251)
* fix(geolocate): no proxy when discovering our IP address
The use case of --proxy is that you cannot contact the OONI
backend otherwise. It is wrong, though, using the proxy when
discovering our IP address. The measurement won't use the
proxy anyway. Therefore, we need to use the IP address that
is performing the measurement. Not the one of the proxy.
What's more, stun is not using a proxy. Therefore, it does
not make much sense that http IP resolvers use a proxy. This
leads to inconsistencies. So, here's anothe reason why this
patch is a good thing (TM).
Finally, because knowing the IP address enables us to sanitize
the data, it's important we discover the correct IP.
Now, up until this point, the `--proxy` option has mostly
been a developers toy. But, users have asked us to have the
possibility of configuring a proxy.
This explains why I have been looking into making `--proxy`
right for a couple of hours now.
See https://github.com/ooni/probe/issues/1382
* fix(session): properly configure the IP lookupper
2021-03-10 12:01:08 +01:00
|
|
|
"github.com/ooni/probe-cli/v3/internal/engine/netx"
|
refactor: flatten and separate (#353)
* refactor(atomicx): move outside the engine package
After merging probe-engine into probe-cli, my impression is that we have
too much unnecessary nesting of packages in this repository.
The idea of this commit and of a bunch of following commits will instead
be to reduce the nesting and simplify the structure.
While there, improve the documentation.
* fix: always use the atomicx package
For consistency, never use sync/atomic and always use ./internal/atomicx
so we can just grep and make sure we're not risking to crash if we make
a subtle mistake on a 32 bit platform.
While there, mention in the contributing guidelines that we want to
always prefer the ./internal/atomicx package over sync/atomic.
* fix(atomicx): remove unnecessary constructor
We don't need a constructor here. The default constructed `&Int64{}`
instance is already usable and the constructor does not add anything to
what we are doing, rather it just creates extra confusion.
* cleanup(atomicx): we are not using Float64
Because atomicx.Float64 is unused, we can safely zap it.
* cleanup(atomicx): simplify impl and improve tests
We can simplify the implementation by using defer and by letting
the Load() method call Add(0).
We can improve tests by making many goroutines updated the
atomic int64 value concurrently.
* refactor(fsx): can live in the ./internal pkg
Let us reduce the amount of nesting. While there, ensure that the
package only exports the bare minimum, and improve the documentation
of the tests, to ease reading the code.
* refactor: move runtimex to ./internal
* refactor: move shellx into the ./internal package
While there, remove unnecessary dependency between packages.
While there, specify in the contributing guidelines that
one should use x/sys/execabs instead of os/exec.
* refactor: move ooapi into the ./internal pkg
* refactor(humanize): move to ./internal and better docs
* refactor: move platform to ./internal
* refactor(randx): move to ./internal
* refactor(multierror): move into the ./internal pkg
* refactor(kvstore): all kvstores in ./internal
Rather than having part of the kvstore inside ./internal/engine/kvstore
and part in ./internal/engine/kvstore.go, let us put every piece of code
that is kvstore related into the ./internal/kvstore package.
* fix(kvstore): always return ErrNoSuchKey on Get() error
It should help to use the kvstore everywhere removing all the
copies that are lingering around the tree.
* sessionresolver: make KVStore mandatory
Simplifies implementation. While there, use the ./internal/kvstore
package rather than having our private implementation.
* fix(ooapi): use the ./internal/kvstore package
* fix(platform): better documentation
2021-06-04 10:34:18 +02:00
|
|
|
"github.com/ooni/probe-cli/v3/internal/runtimex"
|
2021-02-04 11:00:27 +01:00
|
|
|
"github.com/ooni/probe-cli/v3/internal/version"
|
2021-02-02 12:05:47 +01:00
|
|
|
)
|
|
|
|
|
|
|
|
const (
|
|
|
|
// DefaultProbeASN is the default probe ASN as number.
|
|
|
|
DefaultProbeASN uint = 0
|
|
|
|
|
|
|
|
// DefaultProbeCC is the default probe CC.
|
|
|
|
DefaultProbeCC = "ZZ"
|
|
|
|
|
|
|
|
// DefaultProbeIP is the default probe IP.
|
|
|
|
DefaultProbeIP = model.DefaultProbeIP
|
|
|
|
|
|
|
|
// DefaultProbeNetworkName is the default probe network name.
|
|
|
|
DefaultProbeNetworkName = ""
|
|
|
|
|
|
|
|
// DefaultResolverASN is the default resolver ASN.
|
|
|
|
DefaultResolverASN uint = 0
|
|
|
|
|
|
|
|
// DefaultResolverIP is the default resolver IP.
|
|
|
|
DefaultResolverIP = "127.0.0.2"
|
|
|
|
|
|
|
|
// DefaultResolverNetworkName is the default resolver network name.
|
|
|
|
DefaultResolverNetworkName = ""
|
|
|
|
)
|
|
|
|
|
|
|
|
var (
|
|
|
|
// DefaultProbeASNString is the default probe ASN as a string.
|
|
|
|
DefaultProbeASNString = fmt.Sprintf("AS%d", DefaultProbeASN)
|
|
|
|
|
|
|
|
// DefaultResolverASNString is the default resolver ASN as a string.
|
|
|
|
DefaultResolverASNString = fmt.Sprintf("AS%d", DefaultResolverASN)
|
|
|
|
)
|
|
|
|
|
|
|
|
// Logger is the definition of Logger used by this package.
|
|
|
|
type Logger interface {
|
fix(geolocate): no proxy when discovering our IP address (#251)
* fix(geolocate): no proxy when discovering our IP address
The use case of --proxy is that you cannot contact the OONI
backend otherwise. It is wrong, though, using the proxy when
discovering our IP address. The measurement won't use the
proxy anyway. Therefore, we need to use the IP address that
is performing the measurement. Not the one of the proxy.
What's more, stun is not using a proxy. Therefore, it does
not make much sense that http IP resolvers use a proxy. This
leads to inconsistencies. So, here's anothe reason why this
patch is a good thing (TM).
Finally, because knowing the IP address enables us to sanitize
the data, it's important we discover the correct IP.
Now, up until this point, the `--proxy` option has mostly
been a developers toy. But, users have asked us to have the
possibility of configuring a proxy.
This explains why I have been looking into making `--proxy`
right for a couple of hours now.
See https://github.com/ooni/probe/issues/1382
* fix(session): properly configure the IP lookupper
2021-03-10 12:01:08 +01:00
|
|
|
Debug(msg string)
|
2021-02-02 12:05:47 +01:00
|
|
|
Debugf(format string, v ...interface{})
|
fix(geolocate): no proxy when discovering our IP address (#251)
* fix(geolocate): no proxy when discovering our IP address
The use case of --proxy is that you cannot contact the OONI
backend otherwise. It is wrong, though, using the proxy when
discovering our IP address. The measurement won't use the
proxy anyway. Therefore, we need to use the IP address that
is performing the measurement. Not the one of the proxy.
What's more, stun is not using a proxy. Therefore, it does
not make much sense that http IP resolvers use a proxy. This
leads to inconsistencies. So, here's anothe reason why this
patch is a good thing (TM).
Finally, because knowing the IP address enables us to sanitize
the data, it's important we discover the correct IP.
Now, up until this point, the `--proxy` option has mostly
been a developers toy. But, users have asked us to have the
possibility of configuring a proxy.
This explains why I have been looking into making `--proxy`
right for a couple of hours now.
See https://github.com/ooni/probe/issues/1382
* fix(session): properly configure the IP lookupper
2021-03-10 12:01:08 +01:00
|
|
|
Info(msg string)
|
|
|
|
Infof(format string, v ...interface{})
|
|
|
|
Warn(msg string)
|
|
|
|
Warnf(format string, v ...interface{})
|
2021-02-02 12:05:47 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
// Results contains geolocate results
|
|
|
|
type Results struct {
|
|
|
|
// ASN is the autonomous system number
|
|
|
|
ASN uint
|
|
|
|
|
|
|
|
// CountryCode is the country code
|
|
|
|
CountryCode string
|
|
|
|
|
2021-04-07 18:48:02 +02:00
|
|
|
// didResolverLookup indicates whether we did a resolver lookup.
|
|
|
|
didResolverLookup bool
|
2021-02-02 12:05:47 +01:00
|
|
|
|
|
|
|
// NetworkName is the network name
|
|
|
|
NetworkName string
|
|
|
|
|
|
|
|
// IP is the probe IP
|
|
|
|
ProbeIP string
|
|
|
|
|
|
|
|
// ResolverASN is the resolver ASN
|
|
|
|
ResolverASN uint
|
|
|
|
|
|
|
|
// ResolverIP is the resolver IP
|
|
|
|
ResolverIP string
|
|
|
|
|
|
|
|
// ResolverNetworkName is the resolver network name
|
|
|
|
ResolverNetworkName string
|
|
|
|
}
|
|
|
|
|
|
|
|
// ASNString returns the ASN as a string
|
|
|
|
func (r *Results) ASNString() string {
|
|
|
|
return fmt.Sprintf("AS%d", r.ASN)
|
|
|
|
}
|
|
|
|
|
|
|
|
type probeIPLookupper interface {
|
|
|
|
LookupProbeIP(ctx context.Context) (addr string, err error)
|
|
|
|
}
|
|
|
|
|
|
|
|
type asnLookupper interface {
|
2021-04-01 16:57:31 +02:00
|
|
|
LookupASN(ip string) (asn uint, network string, err error)
|
2021-02-02 12:05:47 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
type countryLookupper interface {
|
2021-04-01 16:57:31 +02:00
|
|
|
LookupCC(ip string) (cc string, err error)
|
2021-02-02 12:05:47 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
type resolverIPLookupper interface {
|
|
|
|
LookupResolverIP(ctx context.Context) (addr string, err error)
|
|
|
|
}
|
|
|
|
|
fix(geolocate): no proxy when discovering our IP address (#251)
* fix(geolocate): no proxy when discovering our IP address
The use case of --proxy is that you cannot contact the OONI
backend otherwise. It is wrong, though, using the proxy when
discovering our IP address. The measurement won't use the
proxy anyway. Therefore, we need to use the IP address that
is performing the measurement. Not the one of the proxy.
What's more, stun is not using a proxy. Therefore, it does
not make much sense that http IP resolvers use a proxy. This
leads to inconsistencies. So, here's anothe reason why this
patch is a good thing (TM).
Finally, because knowing the IP address enables us to sanitize
the data, it's important we discover the correct IP.
Now, up until this point, the `--proxy` option has mostly
been a developers toy. But, users have asked us to have the
possibility of configuring a proxy.
This explains why I have been looking into making `--proxy`
right for a couple of hours now.
See https://github.com/ooni/probe/issues/1382
* fix(session): properly configure the IP lookupper
2021-03-10 12:01:08 +01:00
|
|
|
// Resolver is a DNS resolver.
|
|
|
|
type Resolver interface {
|
|
|
|
LookupHost(ctx context.Context, domain string) ([]string, error)
|
|
|
|
Network() string
|
|
|
|
Address() string
|
|
|
|
}
|
|
|
|
|
2021-02-02 12:05:47 +01:00
|
|
|
// Config contains configuration for a geolocate Task.
|
|
|
|
type Config struct {
|
fix(geolocate): no proxy when discovering our IP address (#251)
* fix(geolocate): no proxy when discovering our IP address
The use case of --proxy is that you cannot contact the OONI
backend otherwise. It is wrong, though, using the proxy when
discovering our IP address. The measurement won't use the
proxy anyway. Therefore, we need to use the IP address that
is performing the measurement. Not the one of the proxy.
What's more, stun is not using a proxy. Therefore, it does
not make much sense that http IP resolvers use a proxy. This
leads to inconsistencies. So, here's anothe reason why this
patch is a good thing (TM).
Finally, because knowing the IP address enables us to sanitize
the data, it's important we discover the correct IP.
Now, up until this point, the `--proxy` option has mostly
been a developers toy. But, users have asked us to have the
possibility of configuring a proxy.
This explains why I have been looking into making `--proxy`
right for a couple of hours now.
See https://github.com/ooni/probe/issues/1382
* fix(session): properly configure the IP lookupper
2021-03-10 12:01:08 +01:00
|
|
|
// Resolver is the resolver we should use when
|
|
|
|
// making requests for discovering the IP. When
|
|
|
|
// this field is not set, we use the stdlib.
|
|
|
|
Resolver Resolver
|
2021-02-02 12:05:47 +01:00
|
|
|
|
|
|
|
// Logger is the logger to use. If not set, then we will
|
|
|
|
// use a logger that discards all messages.
|
|
|
|
Logger Logger
|
|
|
|
|
|
|
|
// UserAgent is the user agent to use. If not set, then
|
|
|
|
// we will use a default user agent.
|
|
|
|
UserAgent string
|
|
|
|
}
|
|
|
|
|
|
|
|
// Must ensures that NewTask is successful.
|
|
|
|
func Must(task *Task, err error) *Task {
|
|
|
|
runtimex.PanicOnError(err, "NewTask failed")
|
|
|
|
return task
|
|
|
|
}
|
|
|
|
|
|
|
|
// NewTask creates a new instance of Task from config.
|
|
|
|
func NewTask(config Config) (*Task, error) {
|
|
|
|
if config.Logger == nil {
|
|
|
|
config.Logger = model.DiscardLogger
|
|
|
|
}
|
|
|
|
if config.UserAgent == "" {
|
|
|
|
config.UserAgent = fmt.Sprintf("ooniprobe-engine/%s", version.Version)
|
|
|
|
}
|
fix(geolocate): no proxy when discovering our IP address (#251)
* fix(geolocate): no proxy when discovering our IP address
The use case of --proxy is that you cannot contact the OONI
backend otherwise. It is wrong, though, using the proxy when
discovering our IP address. The measurement won't use the
proxy anyway. Therefore, we need to use the IP address that
is performing the measurement. Not the one of the proxy.
What's more, stun is not using a proxy. Therefore, it does
not make much sense that http IP resolvers use a proxy. This
leads to inconsistencies. So, here's anothe reason why this
patch is a good thing (TM).
Finally, because knowing the IP address enables us to sanitize
the data, it's important we discover the correct IP.
Now, up until this point, the `--proxy` option has mostly
been a developers toy. But, users have asked us to have the
possibility of configuring a proxy.
This explains why I have been looking into making `--proxy`
right for a couple of hours now.
See https://github.com/ooni/probe/issues/1382
* fix(session): properly configure the IP lookupper
2021-03-10 12:01:08 +01:00
|
|
|
if config.Resolver == nil {
|
|
|
|
config.Resolver = netx.NewResolver(
|
|
|
|
netx.Config{Logger: config.Logger})
|
|
|
|
}
|
2021-02-02 12:05:47 +01:00
|
|
|
return &Task{
|
|
|
|
countryLookupper: mmdbLookupper{},
|
2021-04-07 18:48:02 +02:00
|
|
|
probeIPLookupper: ipLookupClient(config),
|
2021-02-02 12:05:47 +01:00
|
|
|
probeASNLookupper: mmdbLookupper{},
|
|
|
|
resolverASNLookupper: mmdbLookupper{},
|
|
|
|
resolverIPLookupper: resolverLookupClient{},
|
|
|
|
}, nil
|
|
|
|
}
|
|
|
|
|
|
|
|
// Task performs a geolocation. You must create a new
|
|
|
|
// instance of Task using the NewTask factory.
|
|
|
|
type Task struct {
|
|
|
|
countryLookupper countryLookupper
|
|
|
|
probeIPLookupper probeIPLookupper
|
|
|
|
probeASNLookupper asnLookupper
|
|
|
|
resolverASNLookupper asnLookupper
|
|
|
|
resolverIPLookupper resolverIPLookupper
|
|
|
|
}
|
|
|
|
|
|
|
|
// Run runs the task.
|
|
|
|
func (op Task) Run(ctx context.Context) (*Results, error) {
|
|
|
|
var err error
|
|
|
|
out := &Results{
|
|
|
|
ASN: DefaultProbeASN,
|
|
|
|
CountryCode: DefaultProbeCC,
|
|
|
|
NetworkName: DefaultProbeNetworkName,
|
|
|
|
ProbeIP: DefaultProbeIP,
|
|
|
|
ResolverASN: DefaultResolverASN,
|
|
|
|
ResolverIP: DefaultResolverIP,
|
|
|
|
ResolverNetworkName: DefaultResolverNetworkName,
|
|
|
|
}
|
|
|
|
ip, err := op.probeIPLookupper.LookupProbeIP(ctx)
|
|
|
|
if err != nil {
|
|
|
|
return out, fmt.Errorf("lookupProbeIP failed: %w", err)
|
|
|
|
}
|
|
|
|
out.ProbeIP = ip
|
2021-04-01 16:57:31 +02:00
|
|
|
asn, networkName, err := op.probeASNLookupper.LookupASN(out.ProbeIP)
|
2021-02-02 12:05:47 +01:00
|
|
|
if err != nil {
|
|
|
|
return out, fmt.Errorf("lookupASN failed: %w", err)
|
|
|
|
}
|
|
|
|
out.ASN = asn
|
|
|
|
out.NetworkName = networkName
|
2021-04-01 16:57:31 +02:00
|
|
|
cc, err := op.countryLookupper.LookupCC(out.ProbeIP)
|
2021-02-02 12:05:47 +01:00
|
|
|
if err != nil {
|
|
|
|
return out, fmt.Errorf("lookupProbeCC failed: %w", err)
|
|
|
|
}
|
|
|
|
out.CountryCode = cc
|
2021-04-07 18:48:02 +02:00
|
|
|
out.didResolverLookup = true
|
|
|
|
// Note: ignoring the result of lookupResolverIP and lookupASN
|
|
|
|
// here is intentional. We don't want this (~minor) failure
|
|
|
|
// to influence the result of the overall lookup. Another design
|
|
|
|
// here could be that of retrying the operation N times?
|
|
|
|
resolverIP, err := op.resolverIPLookupper.LookupResolverIP(ctx)
|
|
|
|
if err != nil {
|
|
|
|
return out, nil // intentional
|
|
|
|
}
|
|
|
|
out.ResolverIP = resolverIP
|
|
|
|
resolverASN, resolverNetworkName, err := op.resolverASNLookupper.LookupASN(
|
|
|
|
out.ResolverIP,
|
|
|
|
)
|
|
|
|
if err != nil {
|
|
|
|
return out, nil // intentional
|
2021-02-02 12:05:47 +01:00
|
|
|
}
|
2021-04-07 18:48:02 +02:00
|
|
|
out.ResolverASN = resolverASN
|
|
|
|
out.ResolverNetworkName = resolverNetworkName
|
2021-02-02 12:05:47 +01:00
|
|
|
return out, nil
|
|
|
|
}
|