Commit Graph

661 Commits (a51576d1ea1e0969013d4b3e796adfa6886802ab)

Author SHA1 Message Date
Jeff a76acd4956 fix wire protocol race condition
only send close packet once, before we were sending a close after we got a close causing excess log spam.
include handshake phase when checking for connection timeouts.

when we change our rc make sure to put it into nodedb too when we are a service node to prevent weirdness in dht lookups.
3 years ago
Jeff eeb93343c0 rpc fixes for lokinet gui
only add stats for services when we have them
3 years ago
audric 41405be612 summary status: no services means not running 3 years ago
Audric Ackermann ad6d206aa6
add version and uptime to summary status endpoint 3 years ago
Audric Ackermann d9467f4dee fix numPaths value in Router::ExtractSummaryStatus 3 years ago
Jeff fc444741f1 move constant to new header
create llarp/constants/time.hpp for time/duration constants
3 years ago
Jeff 5fac6c84d8 detect timeskip and thaw network when we think it happened. 3 years ago
majestrate 27619f5fb4
Merge pull request #1838 from majestrate/sd_notify_extra_info-2022-01-13
redo systemd status line
3 years ago
Jeff b3d9cd463f route poker: allow not blackholing routes.
allow runtime configuration to disable ip blackholing.
3 years ago
Jeff 205584acdf redo systemd status line to include number of paths and endpoint count. optionally warn about low path success. 3 years ago
majestrate 2c44ffe85b
Update llarp/router/route_poker.cpp
Co-authored-by: Jason Rhinelander <jason@imaginary.ca>
3 years ago
Jeff 27ba3e044d
prevent segfault in route poker if Init() is not called 3 years ago
Jeff Becker 14ffdb6639
configurable route poker
this allows you to use exit nodes without forcing routes over the interface, useful for using lokinet with an exit and selectively routing over the lokinet interface using an external socks proxy or binding to device explicitly.

* make route poker configurable, defaults to enabled but allows disabling it on runtime if desired
* add config option [network]:auto-routing to enable/disable route poker
3 years ago
Jeff 388fc53380
match io loop event order on windows/apple to match linux.
on win32/apple reading packets from the interface does not count as an io operation.
manually trigger pump on win32/apple to pretend that it is an io event.
add platform quark function MaybeWakeUpperLayers on vpn::Interface to manaully wake up the other components on platforms that need that (ones on which packet io is not done via io events).
on non linux platforms, use uv_prepare_t instead of uv_check_t as the former triggers before blocking for io, instead of after. this better matches linux's order of operations in libuv.
3 years ago
Jason Rhinelander 09c2b31b25 Move paths pumping into PumpLL 3 years ago
Jason Rhinelander b3a8160c89 Pump hidden context sooner
This has to happen before pumping links because it is putting things on
links.
3 years ago
Jason Rhinelander 9844d358e6 Fix packet delay creep
- Make the main PumpLL also pump hidden services, rather than using
  separate wakers in each TunEndpoint.  It seems there is some
  interactions that just one or the other is not enough.

- Eliminate TunEndpoint send queue -- it isn't needed as we can just
  send directly.
3 years ago
Jason Rhinelander 78cc466bf2 Rename PumpLL -> TriggerPump
And rename the actual pump implementation back to PumpLL.
3 years ago
Jason Rhinelander 031ea7aa37 Get rid of external event loop direct wakeups
If something needs to wake up the event loop it should be using an
async, as we are now with PumpLL(); but we had various code triggering a
wakeup, expecting that PumpLL gets called on every wakeup, which isn't
true anymore.
3 years ago
Jason Rhinelander 74215fc44c Fix link layer delivery race condition (fix random ping delays)
We trigger a pump immediately, but this is racey because we add to our
plaintext data in a worker thread; if the worker thread runs after the
pump then it ends up leaving plaintext to be handled, but there's no
wakeup until the next one.

This was the cause of seeing a random +1s and bunching added to ping
responses sometimes: it wasn't until the *next* ping goes through the
network that the plaintext queue gets processed, at which point it
flushes the old one and often the new one together.

The fix here gets rid of the map of sessions needing wakeups and instead
adds an atomic flag to all of them to let us figure out which ones
need to be flushed.
3 years ago
Jason Rhinelander 9113a6b36b Triggered pumping
This redoes how/when we pump so that we should only be calling the
idempotent PumpLL() when we actually have (or may have) something to
pump.
3 years ago
Jason Rhinelander bfc6d35b33 Simplifications & C++17
- Modernize some iterator loops
- Simplify a couple places by using `if (init; ...)`
- Replace various std::binds with lambdas
3 years ago
Jason Rhinelander cf5c4e7284 outbound_message_handler: skip instead of aborting on removed paths 3 years ago
Jason Rhinelander 633431be66 Make outbound message queue PumpLL again if it doesn't send all 3 years ago
Jeff Becker 8a9025e234 only pump path context in idempotent pumpll 3 years ago
Jeff Becker bb86996acf make event loop pump function non idempotent 3 years ago
Jeff Becker 8744c93944 prevent 100% cpu usage on service nodes
call_soon wakes up the mainloop and is often reentrant, we dont want to
ever way up the event loop but we also want to always defer the call so
we always use the workers
3 years ago
Jeff Becker 22d4b88edc
make PumpLL idempotent to reduce cpu use a bit 3 years ago
Sean e11a94c95c
RPC call for summary get_status (#1742)
* RPC call for summary get_status

* lint

* update with review notes

* further review points

* uint64_t
3 years ago
Jeff Becker 5c457ff486
refactor logging to use std::source_location
* use std::source_location instead of godawful macros in logging
* remove unused/absolutely haram af json logstream
* fix bug in android logger where it doesn't respect eLogNone
3 years ago
Jeff Becker d1246947f7
SystemD DNS fix:
when dns is port 53 call SetLinKDNS otherwise call SetLinkDNSEx as on older versions of systemd-resolved SetLinkDNSEx is not available.
3 years ago
Jeff Becker 9ad63140f2
use std::sample to grab MaxGossipPeers when populating std::unordered_set 3 years ago
Jeff Becker a739e7b532
try not to prefer outbound vs inbound link sessions
before when we get the list of router ids for gossip it was highly bias towards outbound sessions.
instead now we get a full list of link session router ids in random order, truncate them to be at most MaxGossipPeers number of keys, and then put them into an unordered set
3 years ago
Jeff Becker d91ce53da1
limit RC gossip to 20 peers max 3 years ago
Jeff Becker 64cd2990bc
remove old routing table maniuplation code 3 years ago
Jeff Becker 0871862452
initial routing table refactor
* move routing table manipulation to vpn platform
* add initial linux implementation of vpn platform route manipulation
3 years ago
Jeff Becker e96ec156ea
add / remove route blackhole so we dont leak if we crash 3 years ago
Jeff Becker 14cc115489
dont use constexpr 3 years ago
Jeff Becker 069d9487b7
* throw exception on invalid SessionResult when transforming to SendStatus
* add case for printing unknown SessionResult
3 years ago
Jeff Becker 18cb59a1b5
* make tranform function for SessionResult to SendStatus
* add case for SessionResult::EstablishFail
* clean up outbound message handler to use transform function instead of many private member functions
3 years ago
Jeff Becker 75451d7124
* add establish fail enum
* dont call outbound session hooks for inbound sessions
3 years ago
Jeff Becker e3281cd026
add string representation to SessionResult 3 years ago
Jeff Becker 1d1d97b0ef
disable peer status entirely 3 years ago
Jeff Becker 2578983a09
dont give peer stats in rpc 3 years ago
Jeff Becker 99379c5def
establish outbound sessions when we have no outbound session.
when we have an inbound session we still want to make an outbound session.
3 years ago
Jeff Becker cf0349c259
if we have an inbound session don't fail with NoLink 3 years ago
Jeff Becker f4fa83703f
refactor to use RouterID instead of rc.pubkey 3 years ago
Thomas Winget 8515fe09d4 fix minor oversight/inefficiency with already established connections 3 years ago
Jeff Becker a24b82119b
fix #1655
* make it so that we don't set up unbound resolver when we have no resolvers provided by config
* clean up dns codepath and make it use llarp::SockAddr instead of llarp::IpAddress
3 years ago
Jeff Becker 5074dd5f2b
re-enable multithreading on clients but not on service nodes 3 years ago