aboutsummaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
* log why a deltatv13 days1-0/+6
|
* offered deltas lapsetv13 days2-7/+52
|
* fix: level means the peer has our headstv13 days1-6/+13
|
* fix: test links fail over cuts by nametv13 days1-5/+6
|
* dashboard: hop delaystv13 days1-3/+35
|
* throughput from delivered messagetv13 days2-2/+59
|
* hop delaystv13 days8-3/+178
|
* test links fail overtv13 days1-0/+105
|
* benchmark members joiningtv13 days1-5/+48
|
* benchmark delays by pathtv13 days3-0/+28
|
* tests count the changes senttv13 days1-5/+7
|
* fix: change publisher stops with the runtimetv13 days1-3/+8
|
* no reset for heads still on their waytv13 days1-1/+44
|
* benchmark what sync bytes aretv13 days4-3/+74
|
* changes from a peer never go backtv13 days2-0/+52
|
* older delivered messages leave us leveltv13 days2-36/+76
|
* benchmark where twice comes fromtv13 days3-6/+20
|
* resend only what is neededtv13 days2-27/+62
|
* benchmark delivered and twicetv13 days3-0/+58
|
* delivered means has all up to our headstv13 days1-13/+59
|
* any message taken as deliveredtv13 days5-40/+112
|
* slow round back offtv13 days5-5/+92
|
* tree messages need no answertv13 days5-7/+235
|
* hello again when heard fromtv13 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_treestv13 days1-0/+3
|
* unanswered pull frees the slottv13 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 ontv13 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_treestv13 days1-2/+52
|
* overdue sync asked after from thentv13 days3-49/+92
|
* pull after gracetv13 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 lackingtv13 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 aftertv13 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 incidentstv2026-11-031-6/+21
|
* unanswered members not reachedtv2026-11-035-18/+132
|
* add member away benchmarktv2026-11-031-0/+34
|
* dashboard: show reach settv2026-11-031-2/+23
|
* links: only news pass ontv2026-11-031-6/+12
|
* dashboard: show writer treestv2026-11-033-26/+189
|
* add tree teststv2026-11-032-2/+144
|
* fixup 700, 701, 702tv2026-11-034-42/+43
|
* writers from authorstv2026-11-034-27/+68
|
* reach set for the sessiontv2026-11-034-19/+13
|
* add late arrivals benchmarktv2026-11-031-3/+49
|
* tree changes had alreadytv2026-11-032-4/+21
|
* announcement round counterstv2026-11-035-9/+46
|
* docs after treestv2026-11-032-9/+11
|
* announce what trees misstv2026-11-034-35/+119
|
* parallel propagation benchmarktv2026-11-031-14/+54
|
* quicker propagation benchmarktv2026-11-031-14/+21
|
* order members by distancetv2026-11-033-37/+63
|