* add srv records in RCs if we have any
* add mechanism to add SRV records for plainquic exposed ports
* resign and republish rc or introset on srv record changes
* llarp::service::NameIsValid was not checking that the tld was .loki, add this check.
* make link layer initial connection timeout 5s not the session activity timeout which happens to be 60 god damn seconds.
* add lokinet_add_bootstrap_rc function for adding an rc from memory
* prevent stack overflow on error closing connection in quic
* add in memory nodedb
* refactor how convotags are set as active
* add initial stubs for endpoint statistics
* refactor time stuff to be a bit cleaner
* update lnproxy script with more arguments
* wire up last of the quic stuff
* clean up udp packet generation code
* pass EndpointBase not quic tunnel for quic stuff
* add {n,h}uint16_t::FromString
* add nuint_t::FromString
* make AlignedBuffer::IsZero non constant time call for speed
Refactors how quic packets get handled: the actual tunnels now live in
tunnel.hpp's TunnelManager which holds and manages all the quic<->tcp
tunnelling. service::Endpoint now holds a TunnelManager rather than a
quic::Server. We only need one quic server, but we need a separate quic
client instance per outgoing quic tunnel, and TunnelManager handles all
that glue now.
Adds QUIC packet handling to get to the right tunnel code. This
required multiplexing incoming quic packets, as follows:
Adds a very small quic tunnel packet header of 4 bytes:
[1, SPORT, ECN] for client->server packets, where SPORT is our
source "port" (really: just a uint16_t unique quic instance
identifier)
or
[2, DPORT, ECN] for server->client packets where the DPORT is the SPORT
from above.
(This also reworks ECN bits to get properly carried over lokinet.)
We don't need a destination/source port for the server-side because
there is only ever one quic server (and we know we're going to it when
the first byte of the header is 1).
Removes the config option for quic exposing ports; a full lokinet will
simply accept anything incoming on quic and tunnel it to the requested
port on the the local endpoint IP (this handler will come in a following
commit).
Replace ConvoTags with full addresses: we need to carry the port, as
well, which the ConvoTag can't give us, so change those to more general
SockAddrs from which we can extract both the ConvoTag *and* the port.
Add a pending connection queue along with new quic-side handlers to call
when a stream becomes available (TunnelManager uses this to wire up
pending incoming conns with quic streams as streams open up).
Completely get rid of tunnel_server/tunnel_client.cpp code; it is now
moved to tunnel.hpp.
Add listen()/forget() methods in TunnelManager for setting up quic
listening sockets (for liblokinet usage).
Add open()/close() methods in TunnelManager for spinning up new quic
clients for outgoing quic connections.
- De-lambda a lambda that is only called once, immediately after it is
defined.
- C++17 modernize key-value for loops
- Remove unused (and useless) method
All #ifndef guards on headers have been removed, I think,
in favor of #pragma once
Headers are now included as `#include "filename"` if the included file
resides in the same directory as the file including it, or any
subdirectory therein. Otherwise they are included as
`#include <project/top/dir/relative/path/filename>`
The above does not include system/os headers.
loop->call(...) is similar to the old logic->Call(...), but is smart
about the current thread: if called from within the event loop it simply
runs the argument directly, otherwise it queues it.
Similarly most of the other event loop calls are also now thread-aware:
for example, `call_later(...)` can queue the job directly when called if
in the event loop rather than having to double-queue through the even
loop (once to call, then inside the call to initiate the time).
- removes all the llarp_ev_* functions, replacing with methods/classes/functions in the llarp
namespace.
- banish ev/ev.h to the void
- Passes various things by const lvalue ref, especially shared_ptr's that don't need to be copied
(to avoid an atomic refcount increment/decrement).
- Add a llarp::UDPHandle abstract class for UDP handling
- Removes the UDP tick handler; code that needs tick can just do a separate handler on the event
loop outside the UDP socket.
- Adds an "OwnedBuffer" which owns its own memory but is implicitly convertible to a llarp_buffer_t.
This is mostly needed to take over ownership of buffers from uvw without copying them as,
currently, uvw does its own allocation (pending some open upstream issues/PRs).
- Logic:
- add `make_caller`/`call_forever`/`call_every` utility functions to abstract Call wrapping and
dependent timed tasks.
- Add inLogicThread() so that code can tell its inside the logic thread (typically for
debugging assertions).
- get rid of janky integer returns and dealing with cancellations on call_later: the other methods
added here and the event loop code remove the need for them.
- Event loop:
- redo everything with uvw instead of libuv
- rename EventLoopWakeup::Wakeup to EventLoopWakeup::Trigger to better reflect what it does.
- add EventLoopRepeater for repeated events, and replace the code that reschedules itself every
time it is called with a repeater.
- Split up `EventLoop::run()` into a non-virtual base method and abstract `run_loop()` methods;
the base method does a couple extra setup/teardown things that don't need to be in the derived class.
- udp_listen is replaced with ev->udp(...) which returns a new UDPHandle object rather that
needing gross C-style-but-not-actually-C-compatible structs.
- Remove unused register_poll_fd_(un)readable
- Use shared_ptr for EventLoopWakeup rather than returning a raw pointer; uvw lets us not have to
worry about having the event loop class maintain ownership of it.
- Add factory EventLoop::create() function to create a default (uvw-based) event loop (previously
this was one of the llarp_ev_blahblah unnamespaced functions).
- ev_libuv: this is mostly rewritten; all of the glue code/structs, in particular, are gone as
they are no longer needed with uvw.
- DNS:
- Rename DnsHandler to DnsInterceptor to better describe what it does (this is the code that
intercepts all DNS to the tun IP range for Android).
- endpoint:
- remove unused "isolated network" code
- remove distinct (but actually always the same) variables for router/endpoint logic objects
- llarp_buffer_t
- make constructors type-safe against being called with points to non-size-1 values
- tun packet reading:
- read all available packets off the device/file descriptor; previously we were reading one packet
at a time then returning to the event loop to poll again.
- ReadNextPacket() now returns a 0-size packet if the read would block (so that we can implement
the previous point).
- ReadNextPacket() now throws on I/O error
- Miscellaneous code cleanups/simplifications
* bump zmq static dep
* lokimq -> oxenmq
* llarp_nodedb -> llarp::NodeDB
* remove all crufty api parts of NodeDB
* make NodeDB rc selection api not suck
* make path builder api not suck
* propagate all above changes so that unit tests work and it all compiles
* partial tun code refactor
* take out the trash
* move vpn platform code into llarp/vpn/platform.cpp
* fix hive build
* fix win32
* fix memory leak on win32
* reduce cpu use
* make macos compile
* win32 patches:
* use wepoll for zmq
* use all cores on windows iocp read loop
* fix zmq patch for windows
* clean up cmake for win32
* add uninstall before reinstall option to win32 installer
* more ipv6 stuff
* make it compile
* fix up route poker
* remove an unneeded code block in macos wtf
* always use call to system
* fix route poker behavior on macos
* disable ipv6 on windows for now
* cpu perf improvement:
* colease calls to Router::PumpLL to 1 per event loop wakeup
* set up THEN add addresses
* emulate proactor event loop on win32
* remove excessively verbose error message
* fix issue #1499
* exclude uv_poll from win32 so that it can start up
* update logtag to include directory
* create minidump on windows if there was a crash
* make windows happy
* use dmp suffix on minidump files
* typo fix
* address feedback from jason
* use PROJECT_SOURCE_DIR instead of CMAKE_SOURCE_DIR
* quote $@ in apply-patches in case path has spaces in it
* address feedback from tom
* remove llarp/ev/pipe
* add comments for clairification
* make event loop queue size constant named
* more aggressive path building.
* do more than one in parallel path builds at a time
* correct last commit's logic
* rename numPaths -> numDesiredPaths to clarify intent
* revert string change as it will break a lot
* don't prematurly short circuit on snode builds
Co-authored-by: Thomas Winget <tewinget@gmail.com>
* Config file API/comment improvements
API improvements:
=================
Make the config API use position-independent tag parameters (Required,
Default{123}, MultiValue) rather than a sequence of bools with
overloads. For example, instead of:
conf.defineOption<int>("a", "b", false, true, 123, [] { ... });
you now write:
conf.defineOption<int>("a", "b", MultiValue, Default{123}, [] { ... });
The tags are:
- Required
- MultiValue
- Default{value}
plus new abilities (see below):
- Hidden
- RelayOnly
- ClientOnly
- Comment{"line1", "line2", "line3"}
Made option definition more powerful:
=====================================
- `Hidden` allows you to define an option that won't show up in the
generated config file if it isn't set.
- `RelayOnly`/`ClientOnly` sets up an option that is only accepted and
only shows up for relay or client configs. (If neither is specified
the option shows up in both modes).
- `Comment{...}` lets the option comments be specified as part of the
defineOption.
Comment improvements
====================
- Rewrote comments for various options to expand on details.
- Inlined all the comments with the option definitions.
- Several options that were missing comments got comments added.
- Made various options for deprecated and or internal options hidden by
default so that they don't show up in a default config file.
- show the section comment (but not option comments) *after* the
[section] tag instead of before it as it makes more sense that way
(particularly for the [bind] section which has a new long comment to
describe how it works).
Disable profiling by default
============================
We had this weird state where we use and store profiling by default but
never *load* it when starting up. This commit makes us just not use
profiling at all unless explicitly enabled.
Other misc changes:
===================
- change default worker threads to 0 (= num cpus) instead of 1, and fix
it to allow 0.
- Actually apply worker-threads option
- fixed default data-dir value erroneously having quotes around it
- reordered ifname/ifaddr/mapaddr (was previously mapaddr/ifaddr/ifname)
as mapaddr is a sort of specialization of ifaddr and so makes more
sense to come after it (particularly because it now references ifaddr
in its help message).
- removed peer-stats option (since we always require it for relays and
never use it for clients)
- removed router profiles filename option (this doesn't need to be
configurable)
- removed defunct `service-node-seed` option
- Change default logging output file to "" (which means stdout), and
also made "-" work for stdout.
* Router hive compilation fixes
* Comments for SNApp SRV settings in ini file
* Add extra blank line after section comments
* Better deprecated option handling
Allow {client,relay}-only options in {relay,client} configs to be
specified as implicitly deprecated options: they warn, and don't set
anything.
Add an explicit `Deprecated` tag and move deprecated option handling
into definition.cpp.
* Move backwards compat options into section definitions
Keep the "addBackwardsCompatibleConfigOptions" only for options in
sections that no longer exist.
* Fix INI parsing issues & C++17-ify
- don't allow inline comments because it seems they aren't allowed in
ini formats in general, and is going to cause problems if there is a
comment character in a value (e.g. an exit auth string). Additionally
it was breaking on a line such as:
# some comment; see?
because it was treating only `; see?` as the comment and then producing
an error message about the rest of the line being invalid.
- make section parsing stricter: the `[` and `]` have to be at the
beginning at end of the line now (after stripping whitespace).
- Move whitespace stripping to the top since everything in here does it.
- chop off string_view suffix/prefix rather than maintaining position
values
- fix potential infinite loop/segfault when given a line such as `]foo[`
* Make config parsing failure fatal
Load() LogError's and returns false on failure, so we weren't aborting
on config file errors.
* Formatting: allow `{}` for empty functions/structs
Instead of using two lines when empty:
{
}
* Make default dns bind 127.0.0.1 on non-Linux
* Don't show empty section; fix tests
We can conceivably have sections that only make sense for clients or
relays, and so want to completely omit that section if we have no
options for the type of config being generated.
Also fixes missing empty lines between tests.
Co-authored-by: Thomas Winget <tewinget@gmail.com>
* serialize auth codes in rpc
* format
* format
* config and route fixups
* bump lokimq
* fix compile
* discard unknown members in service keys
* format
* prepare for keytool script
* dont serialize ephemeral members in service key file
* regnerate ephemeral members in service identity on load
* add keygen script
* use nacl for generating keys
* format
* * refactor route poking out of llarp::Router and into llarp::RoutePoker
* only poke routes when we have an exit enabled as a client
* add route_poker header so it compiles
* initial relay side lns
* fix typo
* add reserved names and refactor test for dns
* lns name decryption
* all wired up (allegedly)
* refact to use service::EncryptedName for LNS responses to include nonce with ciphertext
* fully rwemove tag_lookup_job
* replace lns cache with DecayingHashTable
* check for lns name validity against the following rules:
* not localhost.loki, loki.loki, or snode.loki
* if it contains no dash then max 32 characters long, not including the .loki tld (and also assuming a leading subdomain has been stripped)
* These are from general DNS requirements, and also enforced in
registrations:
* Must be all [A-Za-z0-9-]. (A-Z will be lower-cased by the RPC call).
* cannot start or end with a -
* max 63 characters long if it does contain a dash
* cannot contain -- in the third and fourth characters unless it starts with xn--
* handle timeout in name lookup job by calling the right handler with std::nullopt
* fix up macos route poker logic
* fix typo
* use string_view
* add forgotten header
* full paths
* add debugging
* catch exception on adding route
* workarround for macos
* typofix
* typofix
* fix for macos
* fix command for macos
* because we autopoke remove explicit route poking in rpc
* probably final fix of macos route poking
* split routes instead of deleting them
* dynamic route poking
* move log statement for introset lookup and dont consider bad sessions as able to send
* send convotag reset frame when we have no session
* add exit map to rpc
* use split_any
* route poking:
* remove popen() call, replace with reading /proc/net/route for getting default route
* dynamically poke and unpoke routes on runtime
* swap intros and fix rpc endpoint for version to return what the ui expects
* use std::string::find_first_not_of instead of using a lambda
* update loki-mq submodule for tuple support
* srv record reply implementation
still need to encode srv records into intro sets / router contacts
as well as decode from them and match against queried service.proto
* inverted condition fix in config code
* SRV record struct (de-)serialization for intro sets
* parsing and using srv records from config (for/in introsets)
* adopt str utils from core and use for srv parsing
* changes to repeat requests
no longer drop repeat requests on the floor, but do not make
an *actual* request for them if one is in progress.
do not call reply hook for each reply for a request, as
each userland request is actually made into several lokinet
requests and this would result in duplicate replies.
* fetch SRVs from introsets for .loki
* make format
* dns and srv fixes, srv appears to be working
* use exceptions when fetching identity key instead of std::optional, will throw on fail
* fix up config options for endpoint auth and add better docs
* add llarp::serive::AuthType enum for controlling what kind of endpoint auth to use
* rework net code for ip ranges to be cleaner
* clean up endpoint auth code
* refactor config to validate network configs before setting up endpoints
* remove buildone from path/pathbuilder.cpp so we don't spam connection attempts
This replaces all use of std::optional's `opt.value()` with `*opt`
because macOS is great and the ghost of Steve Jobs says that actually
supporting std::optional's value() method is not for chumps before macOS
10.14. So don't use it because Apple is great.
Pretty much all of our use of it actually is done better with operator*
anyway (since operator* doesn't do a check that the optional has a
value).
Also replaced *most* of the `has_value()` calls with direct bool
context, except for one in the config section which looked really
confusing at a glance without a has_value().
macOS doing a shared library build is not working without untangling
some of the interdependencies. This commit does that, at least enough
to get macOS to compile.
This isn't the cleanest as currently implemented (we have some net/
things in `liblokinet-platform` and some in `liblokinet`, and likewise
ev/vpnio.cpp is in `liblokinet` while the rest of `ev/*` is in
`liblokinet-platform`).
llarp/config/config.cpp:
respect [network]:type option
llarp/handlers/exit.cpp:
when [network]:type is null dont init tun interface
llarp/service/context.cpp:
respect [network]:type option
change endpoint name back to "default"
llarp/tooling/router_hive.cpp:
dont use LogicCall for obtaining RCs from underlying relays, it crashes the mainloop and it's probably safe to readonly access RCs.
pybind/common.hpp:
remove typecasters as we use C++17 now
pybind/llarp/config.cpp:
remove SnappConfig
wire up NetworkConfig
pybind/llarp/handlers/pyhandler.hpp:
remove SnappConfig from constructor
pybind/llarp/handlers/pyhandler.cpp:
update constructor implementation to match header
test/hive/hive.py:
remove broke endpoint related code
wire up null endpoint option using NetworkConfig
use index at 0 for relays and clients instead of 1
dont add a python endpoint to all clients
This is an initial pass at doing explicit value checks when handling
config parsing, as opposed to using a visiting pattern. The latter
made it difficult to check for conditions such as missing required
values, multiple values, etc.
It was also generally less readable (think declarative) which further
made it difficult to get a grasp for what our actual configuration file
requirements were.
This commit reflects changes to clang-format rules. Unfortunately,
these rule changes create a massive change to the codebase, which
causes an apparent rewrite of git history.
Git blame's --ignore-rev flag can be used to ignore this commit when
attempting to `git blame` some code.
This template-ifies Router::NotifyRouterEvent() up so that it accepts
the arguments to instantiate the specified event type, forwarding them
to std::make_unique. This would allow (in the future) the function to
no-op the call and avoid memory allocation. It also slightly reduces
the amount of code required to fire an event.
This commit also simplifies some of the RouterEvent code to reduce
redundancy.
DHT PubIntroSentEvent
some helper functions added to RouterHive (C++ class) as well as RouterHive(Python class)
hive.py main() continues to be a testbed for new event types
some more internal classes in pybind
This caused some unwanted behaviour:
- on initial startup we often get two publishes in quick succession
because we're publishing and building paths at the same time
- at the 10m mark we enter a publish loop every 5 seconds because we
have paths with lifetimes < 10min that was triggering this condition,
and yet those paths will never actually be included in the introset
because they are expiring in <10m.
This should ensure that we have enough shortly after startup for initial
path builds.
The spread speed here gets slightly increased to lifetime/5 (=4min)
instead of lifetime/4 (=5min) so that our "normal" number of paths is 5
with occassional momentary drops to 4, but should always keep us >= the
new minimum of 4.
Because the path spread happens over time, this shouldn't result in a
rebuild of several paths: we'll build 4 quickly, then another at +4m,
another at +8m, etc. When the initial 4 expire, we'll be dropping from
9 to 5 established but that's still above the minimum (4) so we won't
need to reconnect to several at once, and the spread builds should keep
us at 5 all the time.
These aren't needed: CMake already knows how to follow #includes and
rebuild when headers change as long as the headers are included
*somewhere*. The extra .cpp files here just require building a bunch of
.cpp files with just header content that we just end up throw away
during linking (since the same things will also be compiled in whatever
other compilation units include the same headers).
- util::Mutex is now a std::shared_timed_mutex, which is capable of
exclusive and shared locks.
- util::Lock is still present as a std::lock_guard<util::Mutex>.
- the locking annotations are preserved, but updated to the latest
supported by clang rather than using abseil's older/deprecated ones.
- ACQUIRE_LOCK macro is gone since we don't pass mutexes by pointer into
locks anymore (WTF abseil).
- ReleasableLock is gone. Instead there are now some llarp::util helper
methods to obtain unique and/or shared locks:
- `auto lock = util::unique_lock(mutex);` gets an RAII-but-also
unlockable object (std::unique_lock<T>, with T inferred from
`mutex`).
- `auto lock = util::shared_lock(mutex);` gets an RAII shared (i.e.
"reader") lock of the mutex.
- `auto lock = util::unique_locks(mutex1, mutex2, mutex3);` can be
used to atomically lock multiple mutexes at once (returning a
tuple of the locks).
This are templated on the mutex which makes them a bit more flexible
than using a concrete type: they can be used for any type of lockable
mutex, not only util::Mutex. (Some of the code here uses them for
getting locks around a std::mutex). Until C++17, using the RAII types
is painfully verbose:
```C++
// pre-C++17 - needing to figure out the mutex type here is annoying:
std::unique_lock<util::Mutex> lock(mutex);
// pre-C++17 and even more verbose (but at least the type isn't needed):
std::unique_lock<decltype(mutex)> lock(mutex);
// our compromise:
auto lock = util::unique_lock(mutex);
// C++17:
std::unique_lock lock(mutex);
```
All of these functions will also warn (under gcc or clang) if you
discard the return value. You can also do fancy things like
`auto l = util::unique_lock(mutex, std::adopt_lock)` (which lets a
lock take over an already-locked mutex).
- metrics code is gone, which also removes a big pile of code that was
only used by metrics:
- llarp::util::Scheduler
- llarp:🧵:TimerQueue
- llarp::util::Stopwatch
