aboutsummaryrefslogtreecommitdiffstats
path: root/docs
Commit message (Collapse)AuthorAgeFilesLines
* dashboard: unpin nodes don't move on static graphtv4 days1-2/+2
|
* dashboard: don't reset map after resizetv4 days1-4/+4
|
* dashboard: don't update tooltipstv4 days1-1/+4
|
* dashboard: make via column shortertv4 days1-1/+1
|
* dashboard: ignore traffic while suspendedtv4 days1-1/+3
|
* bench tor round tripstv4 days1-1/+1
|
* predecessors go alongtv4 days1-1/+5
|
* covered neighbours have ittv4 days1-0/+8
|
* asked again countertv4 days2-1/+5
|
* neigbours by transporttv4 days2-1/+7
|
* own tree shapetv4 days2-2/+25
|
* drop full mesh stattv4 days2-4/+1
|
* bench mycelium round tripstv4 days1-2/+3
|
* delta from held headstv4 days1-2/+3
|
* offer again ungeneratedtv4 days1-1/+2
|
* stale delta offerstv4 days1-0/+3
|
* delta known to sendertv4 days1-1/+2
|
* delta propagationtv4 days1-1/+2
|
* benchmark bridge fast linktv4 days1-2/+3
|
* benchmark bridge slow linktv4 days1-1/+2
|
* benchmark bridgetv4 days1-1/+4
|
* one delta at a timetv4 days1-1/+5
|
* time blob messagestv4 days3-4/+4
|
* dashboard: hop delaystv4 days1-0/+1
|
* throughput from delivered messagetv4 days1-1/+1
|
* hop delaystv4 days2-0/+33
|
* benchmark members joiningtv4 days1-3/+3
|
* document messages that overtaketv4 days1-4/+20
|
* any message taken as deliveredtv4 days1-7/+9
|
* slow round back offtv4 days2-5/+10
|
* tree messages need no answertv4 days1-3/+10
|
* hello again when heard fromtv4 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 slottv5 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 ontv5 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 gracetv5 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 lackingtv5 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 reachedtv5 days2-6/+11
|
* add member away benchmarktv5 days1-1/+2
|
* dashboard: show reach settv5 days1-0/+1
|
* dashboard: show writer treestv5 days2-0/+44
|
* update architecture doctv5 days1-4/+7
|
* writers from authorstv6 days1-5/+7
|
* reach set for the sessiontv6 days2-3/+5
|
* tree changes had alreadytv6 days1-1/+3
|
* announcement round counterstv6 days2-1/+7
|
* docs after treestv6 days1-1/+1
|
* announce what trees misstv6 days1-1/+6
|
* parallel propagation benchmarktv6 days1-1/+3
|
* quicker propagation benchmarktv6 days1-1/+2
|
* order members by distancetv6 days1-4/+6
|