Age | Commit message (Collapse) | Author | |
---|---|---|---|
2018-06-17 | Merge remote-tracking branch 'juga/bug26301' | Nick Mathewson | |
2018-06-04 | Change interval of time for bandwidth changes | juga0 | |
Paragraph on events required to update descriptor were not reflecting changes made in #23856 where the bandwidth changes need to happen in a time interval of 3h instead of 20min. | |||
2018-06-04 | Include uptime in bandwidth changes event | juga0 | |
The descriptor would be updated on bandwidth changes only when the uptime is less than 24h. | |||
2018-05-31 | Fix typo | juga0 | |
2018-05-31 | Rephrase bandwidth-file line description | juga0 | |
also change version of Tor where 1st appeared. | |||
2018-05-30 | Add bandwidth-file line to vote documents | juga0 | |
2018-04-16 | Change the oldest-allowed consensus method to 25. | Nick Mathewson | |
2018-04-16 | dir-spec: Add consensus param circ_max_cell_queue_size | David Goulet | |
Signed-off-by: David Goulet <dgoulet@torproject.org> | |||
2018-04-03 | specify that Guard needs the V2Dir flag too | Roger Dingledine | |
part of the change made by ticket 22310, where dir auths no longer vote in favor of the Guard flag for relays that don't advertise directory support. | |||
2018-04-03 | Merge remote-tracking branch 'teor/bug25284' | Nick Mathewson | |
2018-04-02 | Merge remote-tracking branch 'filippo/master' | Nick Mathewson | |
2018-03-30 | Tor has never supported versions in the hidden-service-dir descriptor line | teor | |
Instead, we use the HSDir "proto" versions, or default to version 2. Closes 25284. | |||
2018-03-19 | Fix an ambiguity in the grammar for Document. | Nick Mathewson | |
Previously, a BEGIN/END block might be interpreted as starting with the keyword "-----BEGIN". This change specifically disallows that, to match the behavior of the existing Tor code. Fixes bug 25188. Bugfix from "witchof0x20". | |||
2018-03-03 | Merge branch 'ticket25095_01_squashed' | Nick Mathewson | |
2018-03-03 | tor-spec: Document DoS mitigation consensus param | David Goulet | |
Closes #25095 Signed-off-by: David Goulet <dgoulet@torproject.org> | |||
2018-02-15 | Describe the protocol versions lists more accurately. | Nick Mathewson | |
2018-02-14 | Document new consensus params for bug24769. | Mike Perry | |
2018-02-09 | Add a graf about fingerprinting anonymous directory connections. | Nick Mathewson | |
2018-02-09 | Merge branch 'document_prop278' | Nick Mathewson | |
2018-01-31 | clarify [will become mandatory] | Nick Mathewson | |
2018-01-25 | Note that "protocols" will become mandatory. | Nick Mathewson | |
Closes ticket 24023. | |||
2018-01-24 | Merge prop278 (compression) into dir-spec. | Nick Mathewson | |
2018-01-19 | dir-spec: Add consensus parameter hs_service_max_rdv_failures | David Goulet | |
Signed-off-by: David Goulet <dgoulet@torproject.org> | |||
2017-12-04 | Merge remote-tracking branch 'teor/bug23826-23828-v2' | Nick Mathewson | |
2017-11-17 | dir-spec: remove 5.4.4. Warning about a router's status. | Filippo Valsorda | |
It is obsolete as authorities do not assign nicknames anymore. | |||
2017-11-17 | dir-spec, rend-spec-v3: typos and simplifications | Filippo Valsorda | |
2017-11-13 | dir-spec: Document consensus param HSV3MaxDescriptorSize | David Goulet | |
Signed-off-by: David Goulet <dgoulet@torproject.org> | |||
2017-11-03 | fixup! dir-spec: Document HS v3 consensus params | George Kadianakis | |
2017-11-01 | dir-spec: Document HS v3 consensus params | David Goulet | |
Signed-off-by: David Goulet <dgoulet@torproject.org> | |||
2017-10-26 | Add consensus method 28, which removes IPv6 addresses from microdescriptors | teor | |
Documents #23828. | |||
2017-10-26 | Add consensus method 27, which adds IPv6 addresses to the micro consensus | teor | |
Documents #23826. | |||
2017-10-26 | Document current authority IPv6 reachability behaviour in dir-spec | teor | |
This is the tor-spec part of #23870. | |||
2017-10-26 | Correct the multiplicity of the "p" line in microdescriptors | teor | |
These are all "at most once" items -- saying "zero times" implies that we should reject md documents that include them, which isn't true. Fixes 24007. | |||
2017-10-25 | Correct the multiplicity of the "a" lines. | Nick Mathewson | |
These are all "any number" items -- saying "at most once" implies that we should reject consensus or md documents that include them more then once, which isn't true. | |||
2017-10-25 | Document the current "or-address" and "a" line implementation | teor | |
Preparation for #20916. | |||
2017-10-25 | Add descriptions for consensus methods 19 to 21 | teor | |
And clarify consensus method 18. | |||
2017-10-25 | Fix some dir-spec.txt typos and formatting | teor | |
2017-10-24 | Merge remote-tracking branch 'arma/ticket23637-spec' | Nick Mathewson | |
2017-10-10 | Describe behavior of unrecognized Method. | Nick Mathewson | |
2017-10-10 | Clarify a bit harder | Nick Mathewson | |
2017-10-10 | Merge remote-tracking branch 'isis/ticket18329_02' | Nick Mathewson | |
2017-09-24 | assign Exit flag differently (from ticket 23637) | Roger Dingledine | |
2017-09-20 | Close proposal 162 as "as implemented as it will get. | Nick Mathewson | |
2017-09-20 | Replace MAY with SHOULD for SHA256-signatures on microdesc consensuses. | Nick Mathewson | |
2017-09-12 | Update "protocols" description to say it is obsolete | Nick Mathewson | |
Closes ticket 23445. | |||
2017-09-07 | Remove section 5.4.5. | Nick Mathewson | |
This section made sense with the v2 networkstatus algorithm, but we haven't used that one for years and years. | |||
2017-08-23 | document the term "reasonably live"; clarify "live" | Nick Mathewson | |
2017-08-07 | dir-spec: Attempt to better document ECC key formats and sign bits. | Isis Lovecruft | |
2017-08-03 | dir-spec: Specify linewrapping behaviour for base64-encoded data. | Isis Lovecruft | |
* THANKS TO Mr. George Costanza for pointing out the ambiguity. | |||
2017-07-03 | Standardize use of 'NUM' for non-negative integers in extrainfo | Damian Johnson | |
Fields used 'NUM', 'num', 'N', and in one case had a 'nul' typo. I'm not entirely clear what formally indicates 'non-negative integer' but at least we can standardize this. Back in May I reached out to Mike to confirm padding-counts are and will always be integers, so specifying that. |