aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAge
* Use log_fn_ratelim in a few places.Nick Mathewson2012-12-26
|
* Wrapper function for the common rate-limited-log pattern.Nick Mathewson2012-12-26
|
* Rate-limit "No circuits are opened" message to once-per-hourNick Mathewson2012-12-26
| | | | | | | | mr-4 reports on #7799 that he was seeing it several times per second, which suggests that things had gone very wrong. This isn't a real fix, but it should make Tor usable till we can figure out the real issue.
* Fix a possibly-unused-var warning. Thank you, GCC.Nick Mathewson2012-12-25
|
* Fix compilation warning: must not format u64 as long.Nick Mathewson2012-12-25
|
* Fix whitespaceNick Mathewson2012-12-25
|
* Merge remote-tracking branch 'mikeperry/209-path-bias-changes'Nick Mathewson2012-12-25
|\
| * Nick's Code review #3 part 2.Mike Perry2012-12-18
| |
| * Changes from Nick's code review 'part 1'Mike Perry2012-12-18
| | | | | | | | I think this is actually his third code review of this branch so far.
| * Tags on relay cells can result in certain reason codes.Mike Perry2012-12-11
| | | | | | | | | | Close the circuit (it's probably junk anyways), and make sure we don't probe it/count it as a success.
| * Woops, this log message triggers with the 2-hop bias commit.Mike Perry2012-12-11
| |
| * Add a missing comment.Mike Perry2012-12-10
| |
| * More log message and space fixups.Mike Perry2012-12-10
| |
| * Clean up some XXX comments.Mike Perry2012-12-09
| |
| * Space fixes.Mike Perry2012-12-09
| |
| * Fix a rather serious use-count state bug.Mike Perry2012-12-09
| | | | | | | | | | We need to use the success count or the use count depending on the consensus parameter.
| * Convert to doubles for all pathbias state.Mike Perry2012-12-09
| | | | | | | | Let's hope this solves the rounding error issue..
| * Also shorten circuit_successes to circ_successes.Mike Perry2012-12-09
| | | | | | | | | | | | For consistency and great justice. Ok, mostly consistency.
| * Rename first_hop to circ_attempt.Mike Perry2012-12-09
| | | | | | | | | | Since we've generalized what we can count from (first or second hop), we should generalize the variable and constant naming too.
| * Change from first hop accounting to 2nd hop accountingMike Perry2012-12-09
| | | | | | | | | | | | | | This has several advantages, including more resilience to ambient failure. I still need to rename all the first_hop vars tho.. Saving that for a separate commit.
| * Add intro+rend cannibalize param..Mike Perry2012-12-09
| |
| * Flag cannibalized circs as used (non-ideal).Mike Perry2012-12-09
| | | | | | | | Also add some comments.
| * Allow any valid 'end' cell to mean a circuit was used successfully.Mike Perry2012-12-08
| | | | | | | | Also improve some log messages.
| * Sadly, we can't safely count client intro circ successMike Perry2012-12-08
| |
| * Fix some hidden service edge cases.Mike Perry2012-12-08
| |
| * Refactor path use bias code into own function.Mike Perry2012-12-07
| | | | | | | | Also, improve and log some failure cases.
| * Actually, both nacks and acks indicate a valid pathMike Perry2012-12-07
| |
| * Note more potential issues.Mike Perry2012-12-07
| |
| * Note a strange case for SOCKS streams.Mike Perry2012-12-07
| |
| * Fix another crash bug.Mike Perry2012-12-07
| |
| * Fix a crash bug and pass down a remote reason code.Mike Perry2012-12-07
| | | | | | | | Unexpected channel closures count as remote circ failures.
| * Document that care needs to be taken with any_streams_attached.Mike Perry2012-12-07
| |
| * Remove n_chan codepaths for determinining guard.Mike Perry2012-12-07
| | | | | | | | Cpath is apparently good enough.
| * Move a pathbias function that depends on entryguard_t.Mike Perry2012-12-07
| |
| * Fix a crash due to NULL circ->n_chan.Mike Perry2012-12-07
| | | | | | | | | | Is this redundant? Can we always rely on circ->cpath->extend_info being present for origin circuits?
| * Prop209: Rend circuits weren't ever marked dirty.Mike Perry2012-12-07
| |
| * Prop 209: Add in hidserv path bias counts for usage.Mike Perry2012-12-07
| |
| * Prop 209: Add path bias counts for timeouts and other mechanisms.Mike Perry2012-12-07
| | | | | | | | | | | | | | Turns out there's more than one way to block a tagged circuit. This seems to successfully handle all of the normal exit circuits. Hidden services need additional tweaks, still.
| * Be explicit about units for timeout.Mike Perry2012-12-07
| |
| * Fix an assert crash and an incorrectly placed return.Mike Perry2012-12-07
| |
| * Update with code review changes from Nick.Mike Perry2012-12-07
| |
| * Fix spaces.Mike Perry2012-12-07
| |
| * Refactor pathbias functions to use pathbias_should_count.Mike Perry2012-12-07
| |
| * Update Path Bias log messages to match Proposal 209.Mike Perry2012-12-07
| |
| * Add log message checks for different rates.Mike Perry2012-12-07
| | | | | | | | May want to squash this forward or back..
| * Update pathbias parameters to match Proposal 209.Mike Perry2012-12-07
| | | | | | | | Needs manpage update and testing still..
| * Add the ability to count circuit timeouts for guards.Mike Perry2012-12-07
| | | | | | | | This is purely for informational reasons for debugging.
* | Add configuration options for directory guardsNick Mathewson2012-12-25
| | | | | | | | | | In addition to all the other ways to make directory gurads not go, you can now set UseEntryGuardsAsDirGuards to 0.
* | Directory guard implementation.Nick Mathewson2012-12-25
| | | | | | | | Implements proposal 207; ticket 6526.
* | Remember which of our guards are directory cachesNick Mathewson2012-12-25
| |
generated by cgit v1.2.3 (git 2.49.0) at 2025-04-28 20:14:19 +0000