aboutsummaryrefslogtreecommitdiffstats
path: root/docs/peers.md
Commit message (Collapse)AuthorAgeFilesLines
* heal is lasttv5 days1-0/+6
|
* offer answerstv5 days1-1/+2
|
* round trips in treestv6 days1-3/+6
|
* suspect pathstv6 days1-1/+4
|
* deflate blob chunkstv6 days1-2/+3
|
* deflate blob chunkstv6 days1-0/+5
|
* epoch tag in synctv6 days1-1/+4
|
* no swarm ID in synctv6 days1-1/+2
|
* announce only to those behindtv6 days1-1/+2
|
* predecessors go alongtv6 days1-1/+5
|
* covered neighbours have ittv6 days1-0/+8
|
* delta from held headstv6 days1-2/+3
|
* offer again ungeneratedtv6 days1-1/+2
|
* stale delta offerstv6 days1-0/+3
|
* delta known to sendertv6 days1-1/+2
|
* one delta at a timetv6 days1-1/+5
|
* document messages that overtaketv6 days1-4/+20
|
* any message taken as deliveredtv6 days1-7/+9
|
* slow round back offtv6 days1-4/+8
|
* tree messages need no answertv6 days1-3/+10
|
* hello again when heard fromtv6 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 slottv7 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 ontv7 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 gracetv7 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 lackingtv7 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 reachedtv8 days1-5/+10
|
* writers from authorstv8 days1-5/+7
|
* reach set for the sessiontv8 days1-2/+4
|
* tree changes had alreadytv8 days1-1/+3
|
* announce what trees misstv8 days1-1/+6
|
* order members by distancetv8 days1-4/+6
|
* replace eager links by treestv8 days1-9/+2
|
* add writer treestv8 days1-0/+16
|
* measure 2-hop neighbourhoodtv8 days1-0/+9
|
* introduce eager linkstv8 days1-1/+9
|
* skip neighbours with our headstv8 days1-1/+4
|
* announcements answered only if heads differtv8 days1-2/+3
|
* retry new peers before probingtv8 days1-1/+3
|
* probe list memberstv8 days1-1/+3
|
* always announce new headstv8 days1-5/+4
|
* swarm: announce to whom writers can't reachtv8 days1-1/+4
|
* add documentationtv8 days1-0/+358