| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| ... | |
| | | |
| | |
| | |
| | | |
It seems this config is for tokio-console and not the RPC.
|
| | | | |
|
| |\ \ \
| | | |
| | | |
| | | |
| | | | |
Validate Host header for non-CONNECT requests to HTTP CONNECT port
See merge request tpo/core/arti!3429
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This validation makes it harder for an adversarial webpage to probe
for the version of arti and its capabilities.
See torspec!437.
|
| |\ \ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Implement a usage-based timeout for strongly isolated circuits (prop368)
Closes #2237
See merge request tpo/core/arti!3430
|
| | | | | | |
|
| | | | | | |
|
| | | | | |
| | | | |
| | | | |
| | | | | |
Part of prop368.
|
| | |_|/ /
|/| | |
| | | |
| | | |
| | | | |
These appear to be old comments. Now that these features are stable, we
consider them safe to use.
|
| |\ \ \ \
| |/ / /
|/| | |
| | | |
| | | |
| | | |
| | | | |
arti/arti-client/tor-hsservice: Support disabling onion services in the config
Closes #2133
See merge request tpo/core/arti!3253
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Second attempt at semver in progress. Based on the feedback from wesley,
it seems neither of these crates had any breaking changes.
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
If I understand correctly, the breaking changes propagate like this.
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
There is no longer a hard error anywhere for trying to launch a service
which is disabled in the config. Instead, it always means returning
`Ok(None)`.
The axum and hyper examples were updated again as a consequence.
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The "enabled" config option is back to using a regular `bool`. When
unset, it defaults to true, and the service runs as if it had been set.
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
I didn't realize we had tests that made use of the example config. The
pre-commit and pre-push hooks I had didn't trigger them, so my
modification to the example config silently got through and caused
`cfg::test::onion_services` to fail in CI.
The issue was that I had put my example for
`onion_services."allium-cepa".enabled` with a default of `true`. The
correct default is actually `"auto"`, so I switched it and improved the
description of the options.
I've fixed my hooks. Which, to be fair, are just the hooks provided in
`./maint/hooks/`. Maybe those should be improved?
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The "enabled" config option now uses `tor_config::BoolOrAuto`. When
unset (which defaults to "Auto"), the service will run with a warning.
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
When a service is disabled in the config, `arti::onion_proxy::Proxy`
handles it gracefully, and
`arti_client::client::TorClient::launch_onion_service_with_hsid` and
`arti_client::client::TorClient::launch_onion_service` both hard error.
The axum and hyper examples were updated again as a consequence.
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Hidden services disabled in the config are now handled more gracefully
by arti/arti-client, before the hard error occurs.
The axum and hyper examples were updated as a consequence.
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | | |
Signed-off-by: hashcatHitman <[email protected]>
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This has the meta attributes with optional values feature and also
hygiene rework for modules.
The breaking changes don't break arti.
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This is the pattern we used elsewhere with the "rpc" option.
IMO, this kind of thing is another argument in favor if
arti#1704 (redoing config types using derive-deftly.)
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This is controlled by a new option, `logging.tokio_console.enabled`.
It requires building with `--cfg tokio_unstable`. See documentation
and comments for more information.
|
| |\ \ \ \
| | | | |
| | | | |
| | | | |
| | | | | |
Bump derive-deftly to 1.4.0
See merge request tpo/core/arti!3448
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This will let us use the new modules feature.
There are no breaking changes to beta features in 1.4.0.
|
| |/ / / /
| | | |
| | | |
| | | | |
Run maint/add_warning
|
| | |_|/
|/| | |
|
| |\ \ \
| |_|/
|/| |
| | |
| | | |
Replace copy_interactive with futures-copy.
See merge request tpo/core/arti!3416
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This change lets us avoid spawning extra tasks (due to one-direction
nature of copy_interactive), and avoid some lock contention (due to
use of AsyncReadExt::split).
Addresses part of #786.
|
| |/ /
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
```text
warning: use of deprecated associated function `assert_cmd::Command::cargo_bin`: incompatible with a custom cargo build-dir, see instead `cargo::cargo_bin_cmd!`
--> crates/arti/tests/cli_tests/hsc.rs:26:28
|
26 | let mut cmd = Command::cargo_bin("arti").unwrap();
| ^^^^^^^^^
|
= note: `#[warn(deprecated)]` on by default
```
|
| | |
| |
| |
| |
| |
| |
| | |
```bash
readarray -t publish < <(cargo metadata --format-version 1 | jq -r '.packages[] | select((.id | startswith("path+file:///")) and (.rust_version != null) and (.publish == null or .publish == true or .publish != [])) | .name')
for package in "${publish[@]}"; do echo "$package:"; cargo set-version --bump minor -p "$package"; done
```
|
| |\ \
| | |
| | |
| | |
| | |
| | |
| | | |
arti: Revise strings that implied that we were a SOCKS-only proxy.
Closes #2225
See merge request tpo/core/arti!3398
|
| | | | |
|
| | | |
| | |
| | |
| | | |
Closes #2225
|
| | | | |
|
| |/ /
| |
| |
| |
| |
| |
| | |
Apparently the http_connect refactoring made these no longer
trigger.
Closes #2226
|
| | |
| |
| |
| |
| |
| | |
We can't remove them all, since X-Tor-Stream-Isolation is recognized
by C-Tor. I've added a non-deprecated Tor-Stream-Isolation
that works indepenently.
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
