aboutsummaryrefslogtreecommitdiffstats
path: root/docs/peers.md
Commit message (Collapse)AuthorAgeFilesLines
* round trips in treestv3 days1-3/+6
|
* suspect pathstv3 days1-1/+4
|
* deflate blob chunkstv3 days1-2/+3
|
* deflate blob chunkstv3 days1-0/+5
|
* epoch tag in synctv3 days1-1/+4
|
* no swarm ID in synctv3 days1-1/+2
|
* announce only to those behindtv3 days1-1/+2
|
* predecessors go alongtv3 days1-1/+5
|
* covered neighbours have ittv3 days1-0/+8
|
* delta from held headstv3 days1-2/+3
|
* offer again ungeneratedtv3 days1-1/+2
|
* stale delta offerstv3 days1-0/+3
|
* delta known to sendertv3 days1-1/+2
|
* one delta at a timetv3 days1-1/+5
|
* document messages that overtaketv3 days1-4/+20
|
* any message taken as deliveredtv3 days1-7/+9
|
* slow round back offtv3 days1-4/+8
|
* tree messages need no answertv3 days1-3/+10
|
* hello again when heard fromtv3 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 slottv4 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 ontv4 days1-1/+1
| | | | | | | 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 gracetv4 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 lackingtv4 days1-2/+3
| | | | | | | 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 days1-5/+10
|
* writers from authorstv5 days1-5/+7
|
* reach set for the sessiontv5 days1-2/+4
|
* tree changes had alreadytv5 days1-1/+3
|
* announce what trees misstv5 days1-1/+6
|
* order members by distancetv5 days1-4/+6
|
* replace eager links by treestv5 days1-9/+2
|
* add writer treestv5 days1-0/+16
|
* measure 2-hop neighbourhoodtv5 days1-0/+9
|
* introduce eager linkstv5 days1-1/+9
|
* skip neighbours with our headstv5 days1-1/+4
|
* announcements answered only if heads differtv5 days1-2/+3
|
* retry new peers before probingtv5 days1-1/+3
|
* probe list memberstv5 days1-1/+3
|
* always announce new headstv5 days1-5/+4
|
* swarm: announce to whom writers can't reachtv5 days1-1/+4
|
* add documentationtv5 days1-0/+358