aboutsummaryrefslogtreecommitdiffstats
path: root/docs
Commit message (Collapse)AuthorAgeFilesLines
...
* dashboard: don't reset map after resizetv12 days1-4/+4
|
* dashboard: don't update tooltipstv12 days1-1/+4
|
* dashboard: make via column shortertv12 days1-1/+1
|
* dashboard: ignore traffic while suspendedtv12 days1-1/+3
|
* bench tor round tripstv12 days1-1/+1
|
* predecessors go alongtv12 days1-1/+5
|
* covered neighbours have ittv12 days1-0/+8
|
* asked again countertv12 days2-1/+5
|
* neigbours by transporttv12 days2-1/+7
|
* own tree shapetv12 days2-2/+25
|
* drop full mesh stattv12 days2-4/+1
|
* bench mycelium round tripstv13 days1-2/+3
|
* delta from held headstv13 days1-2/+3
|
* offer again ungeneratedtv13 days1-1/+2
|
* stale delta offerstv13 days1-0/+3
|
* delta known to sendertv13 days1-1/+2
|
* delta propagationtv13 days1-1/+2
|
* benchmark bridge fast linktv13 days1-2/+3
|
* benchmark bridge slow linktv13 days1-1/+2
|
* benchmark bridgetv13 days1-1/+4
|
* one delta at a timetv13 days1-1/+5
|
* time blob messagestv13 days3-4/+4
|
* dashboard: hop delaystv13 days1-0/+1
|
* throughput from delivered messagetv13 days1-1/+1
|
* hop delaystv13 days2-0/+33
|
* benchmark members joiningtv13 days1-3/+3
|
* document messages that overtaketv13 days1-4/+20
|
* any message taken as deliveredtv13 days1-7/+9
|
* slow round back offtv13 days2-5/+10
|
* tree messages need no answertv13 days1-3/+10
|
* hello again when heard fromtv13 days1-1/+4
| | | | | | 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.
* unanswered pull frees the slottv13 days1-3/+4
| | | | | | 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 days2-2/+2
| | | | | | | 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.
* pull after gracetv13 days1-1/+3
| | | | | | | | 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 days2-3/+4
| | | | | | | 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 members not reachedtv2026-11-032-6/+11
|
* add member away benchmarktv2026-11-031-1/+2
|
* dashboard: show reach settv2026-11-031-0/+1
|
* dashboard: show writer treestv2026-11-032-0/+44
|
* update architecture doctv2026-11-031-4/+7
|
* writers from authorstv2026-11-031-5/+7
|
* reach set for the sessiontv2026-11-032-3/+5
|
* tree changes had alreadytv2026-11-031-1/+3
|
* announcement round counterstv2026-11-032-1/+7
|
* docs after treestv2026-11-031-1/+1
|
* announce what trees misstv2026-11-031-1/+6
|
* parallel propagation benchmarktv2026-11-031-1/+3
|
* quicker propagation benchmarktv2026-11-031-1/+2
|
* order members by distancetv2026-11-031-4/+6
|
* replace eager links by treestv2026-11-035-32/+13
|