aboutsummaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
* benchmark bridgetv8 days1-0/+120
|
* sim link ratestv8 days1-1/+49
|
* one delta at a timetv8 days2-9/+53
|
* time blob messagestv8 days8-6/+55
|
* log why a deltatv8 days1-0/+6
|
* offered deltas lapsetv8 days2-7/+52
|
* fix: level means the peer has our headstv8 days1-6/+13
|
* fix: test links fail over cuts by nametv8 days1-5/+6
|
* dashboard: hop delaystv8 days1-3/+35
|
* throughput from delivered messagetv8 days2-2/+59
|
* hop delaystv8 days8-3/+178
|
* test links fail overtv8 days1-0/+105
|
* benchmark members joiningtv8 days1-5/+48
|
* benchmark delays by pathtv8 days3-0/+28
|
* tests count the changes senttv8 days1-5/+7
|
* fix: change publisher stops with the runtimetv8 days1-3/+8
|
* no reset for heads still on their waytv8 days1-1/+44
|
* benchmark what sync bytes aretv8 days4-3/+74
|
* changes from a peer never go backtv8 days2-0/+52
|
* older delivered messages leave us leveltv8 days2-36/+76
|
* benchmark where twice comes fromtv8 days3-6/+20
|
* resend only what is neededtv8 days2-27/+62
|
* benchmark delivered and twicetv8 days3-0/+58
|
* delivered means has all up to our headstv8 days1-13/+59
|
* any message taken as deliveredtv8 days5-40/+112
|
* slow round back offtv8 days5-5/+92
|
* tree messages need no answertv8 days5-7/+235
|
* hello again when heard fromtv8 days3-2/+71
| | | | | | While we wait for a peer's HELLO_REPLY, anything else that comes from it makes our HELLO go again at once, at most once a second, instead of when the backoff says.
* bump pulled_changes_go_along_their_writers_treestv9 days1-0/+3
|
* unanswered pull frees the slottv9 days2-1/+23
| | | | | | When the answer to a pull is overdue (the sync reply timeout: 1 s in the simtests, 10 s in production), the pull gives up the slot, and the next waiting neighbour is pulled at once.
* no heal while both move ontv9 days1-3/+49
| | | | | | | The heal clock now also restarts when our heads and the peer's have both changed since the last STATUS that showed them differing. While writes go on, two neighbours's heads differ because of changes still on their way, and that alone no longer leads to a heal.
* bump pulled_changes_go_along_their_writers_treestv9 days1-2/+52
|
* overdue sync asked after from thentv9 days3-49/+92
|
* pull after gracetv9 days2-4/+33
| | | | | | | | Before pulling, a node waits one measured round trip to the announcer (1 s at most) and pulls only what is still missing. If the round trip isn't known yet, it pulls at once, as now. On the clean benchmark links the round trip is a few milliseconds, so those scenarios shouldn't change.
* announcements answered when lackingtv9 days5-12/+36
| | | | | | | A neighbour now answers an announcement only if it lacks some of the announced heads. Heads that merely differ are no longer enough, since they do while writes go on. The check reads the document, so it runs off the message consumer, as the pull check already does.
* unanswered sync asked aftertv9 days1-6/+28
| | | | | An overdue sync answer is now followed by a PING too, and the peer stays reached once it answers the PING.
* simtest: benchmark incidentstv9 days1-6/+21
|
* unanswered members not reachedtv10 days5-18/+132
|
* add member away benchmarktv10 days1-0/+34
|
* dashboard: show reach settv10 days1-2/+23
|
* links: only news pass ontv10 days1-6/+12
|
* dashboard: show writer treestv10 days3-26/+189
|
* add tree teststv10 days2-2/+144
|
* fixup 700, 701, 702tv10 days4-42/+43
|
* writers from authorstv10 days4-27/+68
|
* reach set for the sessiontv10 days4-19/+13
|
* add late arrivals benchmarktv10 days1-3/+49
|
* tree changes had alreadytv10 days2-4/+21
|
* announcement round counterstv10 days5-9/+46
|
* docs after treestv10 days2-9/+11
|