| Commit message (Collapse) | Author | Age | Files | Lines |
| ... | |
| |/ / / / |
|
| | | | | |
|
| |/ / / |
|
| |\ \ \
| | | |
| | | |
| | | |
| | | | |
relay: Couple fixes related to channel creation
See merge request tpo/core/arti!3726
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
It was _target before because we had it in the function with a
cfg_if!(). Now don't so no need.
Signed-off-by: David Goulet <[email protected]>
|
| | | | |
| | | |
| | | |
| | | | |
Signed-off-by: David Goulet <[email protected]>
|
| | | | | |
|
| | | | |
| | | |
| | | |
| | | | |
Signed-off-by: David Goulet <[email protected]>
|
| | |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Reason is that Vec::with_capacity(len) doesn't actually allocate and the
Rustls export_keying_material() does a .is_empty() check which is under
".len() == 0".
Signed-off-by: David Goulet <[email protected]>
|
| |/ /
| |
| |
| |
| |
| |
| | |
Before this commit, we would use the RsaIdentity which is a SHA1 digest.
We do the same for the peer RSA key.
Signed-off-by: David Goulet <[email protected]>
|
| |\ \
| | |
| | |
| | |
| | | |
proto: Implement the build_certs_cell() helper
See merge request tpo/core/arti!3723
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | | |
It was all commented out until now that we have a final
RelayIdentities.
Signed-off-by: David Goulet <[email protected]>
|
| |\ \ \
| |/ /
|/| |
| | |
| | | |
relay: Sleep rotate key task with earliest valid_until
See merge request tpo/core/arti!3717
|
| | | |
| | |
| | |
| | | |
Signed-off-by: David Goulet <[email protected]>
|
| |\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
chanmgr: Use CircIdRange::Low for relay channels
Closes #1601
See merge request tpo/core/arti!3714
|
| | |/ /
| | |
| | |
| | |
| | |
| | | |
Closes #1601
Signed-off-by: David Goulet <[email protected]>
|
| |/ / |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Before:
```text
Completed handshake with Ed25519Identity { ClByyE4zT+p3dudeCVkUfVA1zp2y2sOUe7u/HHLliWU } [(RsaIdentity { $547bb37f9cd8079eaf1cfcb73fff5887d6f58138 }, [9, 111, 131, 62, 212, 144, 149, 88, 211, 87, 175, 114, 18, 111, 189, 36, 205, 21, 66, 205, 217, 176, 139, 130, 70, 211, 40, 130, 79, 29, 144, 179])]
```
After:
```text
Completed handshake with ed25519:wo9NA5RXkPAc7sB2301JBpvdk1RvSOGZEzgwuWP/msA [$feebaba2b6faf53ae1dc8dda8681ce026537dfa4]
```
|
| | |
| |
| |
| |
| | |
These were previously in a single `Option`, but now that the `Option`
was removed, I think it's nicer to make these separate fields.
|
| | |
| |
| |
| |
| |
| | |
I don't think that the `Option`s are needed anymore, since
unauthenticated channels no longer transition through the
`VerifiedChannel` state.
|
| |\ \
| | |
| | |
| | |
| | | |
chanmgr: Store our relay addresses in the builder
See merge request tpo/core/arti!3716
|
| | | |
| | |
| | |
| | |
| | |
| | | |
No need to pass from the arti relay binary our addresses when handling
an incoming channel, use the one in the channel builder that an
initiator channel uses.
|
| | | |
| | |
| | |
| | | |
Signed-off-by: David Goulet <[email protected]>
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
We need those addresses when we build a relay channel in order to send
them into our NETINFO cell.
This adds the `with_my_addrs()` on the `ChanMgrConfig` object. Next
commit will make arti-relay use it.
Signed-off-by: David Goulet <[email protected]>
|
| |/ /
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The validate_relay_target() is meant to probably have more checks in the
future hence the vagueness of it instead of being specific to the goal
of this patch.
Closes #1699
Signed-off-by: David Goulet <[email protected]>
|
| | |
| |
| |
| | |
Signed-off-by: David Goulet <[email protected]>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The arti-relay crate rotates the keys at regular interval which we need
to give to the ChanMgr to update its builder.
This function boldly replace the default factory with the new identities
including the TLS acceptor can pick up the new key.
Signed-off-by: David Goulet <[email protected]>
|
| |\ \
| | |
| | |
| | |
| | | |
tor-rtcompat: Disable rustls session resumption
See merge request tpo/core/arti!3710
|
| | | | |
|
| |\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
rtcompat: simplify async-std and smol incoming streams with unfold
Closes #2070
See merge request tpo/core/arti!3705
|
| | | | | |
|
| | | | | |
|
| |\ \ \ \
| | | | |
| | | | |
| | | | |
| | | | | |
arti-relay: Improve error/warning messages
See merge request tpo/core/arti!3711
|
| | | | | |
| | | | |
| | | | |
| | | | | |
This method was renamed previously.
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Previously, it used the sub_builder rule, which doesn't make sense
when the value is a stub place-holder to tell us whether to warn.
|
| | | | | | |
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
(There are a few places where we still do things in nonstandard
places to avoid API breakage.)
|
| | | | | | |
|
| |/ / / /
| | | |
| | | |
| | | |
| | | | |
As with previous modules, I've left some thing less conformant
with our "standard" APIs in order to keep backward compat (for now).
|
| | | | |
| | | |
| | | |
| | | | |
Signed-off-by: David Goulet <[email protected]>
|
| | | | |
| | | |
| | | |
| | | | |
Signed-off-by: David Goulet <[email protected]>
|