Commit Graph

487 Commits

Author SHA1 Message Date
Jeff Becker
f4971a88fd
use lokimq workers instead of llarp:🧵:ThreadPool 2020-07-02 11:07:34 -04:00
Jeff Becker
1e8368b636
make it compile 2020-07-02 11:07:33 -04:00
Jeff Becker
61c5808765
exit authentication (initial) 2020-07-02 11:07:33 -04:00
Jeff Becker
800668348a
add regression test for key backup bug 2020-06-08 08:42:10 -04:00
Jeff Becker
ac26037c34
when in 1 hop mode we where hitting a case where we were failing to build paths
this was because we were trying to diversify the endpoints used which is not
appicable in 1 hop mode because first hops are sticky.
2020-06-02 07:14:13 -04:00
Jeff Becker
c6885ec285
remove Endpoint::GetExitRouters() 2020-06-01 09:35:21 -04:00
Jeff Becker
6af498092b
exit traffic via loki addresses 2020-05-23 16:07:19 -04:00
Jason Rhinelander
ebd2142114 Don't use std::optional::value() because f u macos
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().
2020-05-20 19:18:28 -03:00
Jason Rhinelander
26ecf23ad8 Silence various warnings 2020-05-17 21:28:17 -03:00
Stephen Shelton
ed31cdaacf
Cosmetic fixes 2020-05-14 11:51:27 -06:00
Stephen Shelton
5f0cbb4625
Don't forget to load your keys 2020-05-04 11:07:22 -04:00
Stephen Shelton
9ba0f117c7
Revert to previous "if hops > 0" logic in Configure() 2020-05-04 11:07:21 -04:00
Stephen Shelton
960dc37c1f
Refactor TunEndpoint::Configure() (partially) 2020-05-04 11:07:21 -04:00
Stephen Shelton
f2a26adcaa
Move all [endpoint] options to [network] 2020-05-04 11:07:21 -04:00
Stephen Shelton
dd9ab0f1d5
Remove ability to have multiple endpoints/snodes from config 2020-05-04 11:07:21 -04:00
Stephen Shelton
b7451b7632
Rip out tag, prefetch tags/addrs from Endpoint 2020-05-04 11:06:47 -04:00
Stephen Shelton
df01770466
Rename SnappConfig -> EndpointConfig 2020-05-04 11:05:51 -04:00
Stephen Shelton
93b35c92a7
Eradicate service::Config 2020-05-04 11:05:51 -04:00
Jason Rhinelander
1697bf90fe C++17
Compiles with C++17, replaces ghc::filesystem with std::filesystem,
nonstd::optional with std::optional, and llarp::string_view with
std::string_view.
2020-05-01 17:43:27 -03:00
Jason Rhinelander
3c6aa002b8 Fix errors and warnings compiling on mac 2020-04-10 14:26:50 -03:00
Stephen Shelton
273270916e
The Great Wall of Blame
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.
2020-04-07 12:38:56 -06:00
Jeff
d3091cf9fc
Merge pull request #1167 from tewinget/tooling
RouterHive initial PR
2020-04-07 13:13:02 -04:00
Stephen Shelton
ea835405c5
Refactor NotifyRouterEvent() to forward args, event code cleanup
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.
2020-03-12 11:50:46 -06:00
Stephen Shelton
6664202868
Add FindRouterSentEvent and FindRouterReceivedEvent to RouterHive 2020-03-10 12:12:11 -06:00
Jeff
f3d23d6fb6
strip rr from dns reply for .snode and .loki (#1172)
fix dns replies so that extra RR for edns are not included.
2020-03-09 17:47:27 -03:00
Thomas Winget
84a1d7dbcc clang format....... 2020-03-06 20:20:11 -05:00
Thomas Winget
695784b2b6 more hive things
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
2020-03-04 00:57:26 -05:00
Jason Rhinelander
c87fd66386 Fix lookup filter to filter second, not first, lookup 2020-03-04 01:32:20 -04:00
Jeff Becker
877443d95c more introspection code 2020-03-03 19:57:09 -05:00
Jeff Becker
d4ccf895e3
Merge branch 'fix-up-introset-lookup-fails-2020-03-02' of ssh://github.com/majestrate/loki-network into fix-up-introset-lookup-fails-2020-03-02 2020-03-02 12:02:58 -05:00
Jeff Becker
79746e3719
remote lookup limiting of 500ms per address 2020-03-02 12:02:21 -05:00
Stephen Shelton
9be8050854
Make format 2020-03-02 09:57:37 -07:00
Stephen Shelton
886a981997
Consolidate introset publishing constants 2020-03-02 09:57:37 -07:00
Jeff Becker
a09bb89d15
dont put multiple outbound contexts 2020-03-02 11:56:47 -05:00
Jeff Becker
158a9018f3
rename variable 2020-03-02 11:18:26 -05:00
Jeff Becker
9ff31f6402
spelling is hard 2020-03-02 11:17:50 -05:00
Jeff Becker
8f4362f092
fix up introset lookups 2020-03-02 11:12:29 -05:00
Jason Rhinelander
806d43c34e
Merge pull request #1147 from notlesh/diagnose-pubintro-round2-2020-02-28
Diagnose pubintro round2
2020-03-01 20:26:41 -04:00
Jason Rhinelander
b60adc909d Make IntroSet publish confirmed less verbose
Essentially just rate limit the confirmation message to one message per
second.
2020-03-01 14:04:37 -04:00
Jason Rhinelander
cfee824a79 Don't republish when we have extra paths
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.
2020-03-01 13:56:04 -04:00
Jason Rhinelander
823c17206f Add min intro set paths, slightly increase spread speed
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.
2020-03-01 12:37:43 -04:00
Jason Rhinelander
27b8743107 Add warns for introset publish failures 2020-03-01 11:48:43 -04:00
Stephen Shelton
62014de91c
Handle PubIntro relayOrder logic on client-side 2020-02-28 16:45:04 -07:00
Stephen Shelton
61d19179f7
Remove dead code 2020-02-28 16:32:00 -07:00
Jeff Becker
ce335dc47d
remove convotags on dead outbound context 2020-02-27 11:54:54 -05:00
Jeff
19a751c41b
Merge pull request #1127 from notlesh/spread-snapp-path-builds-evenly-2020-02-24
Spread snapp path builds evenly
2020-02-25 16:25:47 -05:00
Jeff
4f29405e9e
Merge pull request #1115 from majestrate/handle-multiple-responses-per-lookup-2020-02-20
handle multiple responses per hidden service lookup
2020-02-25 16:25:19 -05:00
Stephen Shelton
abe4015986
Use constants for path build timing 2020-02-25 13:52:59 -07:00
Stephen Shelton
230037b9f3
Increase default path timeout from 10min to 20min 2020-02-25 13:42:14 -07:00
Stephen Shelton
f0374eb2b7
Use existing 'lastBuild' var for spacing path builds 2020-02-25 13:39:28 -07:00