aboutsummaryrefslogtreecommitdiffstats
path: root/src/swarm
Commit message (Collapse)AuthorAgeFilesLines
* changes from a peer never go backtv6 days1-0/+4
|
* older delivered messages leave us leveltv6 days1-1/+3
|
* benchmark where twice comes fromtv6 days2-4/+11
|
* resend only what is neededtv6 days1-1/+1
|
* benchmark delivered and twicetv6 days2-0/+30
|
* any message taken as deliveredtv6 days2-13/+23
|
* slow round back offtv6 days2-5/+86
|
* tree messages need no answertv6 days3-6/+40
|
* hello again when heard fromtv6 days2-2/+19
| | | | | | 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 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.
* overdue sync asked after from thentv7 days2-4/+4
|
* pull after gracetv7 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 lackingtv7 days4-10/+34
| | | | | | | 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 reachedtv7 days2-14/+43
|
* links: only news pass ontv8 days1-6/+12
|
* dashboard: show writer treestv8 days1-26/+70
|
* writers from authorstv8 days3-27/+46
|
* reach set for the sessiontv8 days2-16/+10
|
* tree changes had alreadytv8 days1-4/+14
|
* announcement round counterstv8 days2-8/+23
|
* docs after treestv8 days1-5/+6
|
* announce what trees misstv8 days2-35/+115
|
* order members by distancetv8 days2-26/+59
|
* replace eager links by treestv8 days6-137/+35
|
* add writer treestv8 days5-27/+266
|
* add writer tree benchmarktv8 days1-0/+148
|
* measure 2-hop neighbourhoodtv8 days7-8/+136
|
* dashboard: show eager links on the maptv8 days1-5/+19
|
* add eager link counterstv8 days2-0/+21
|
* introduce eager linkstv8 days3-21/+100
|
* skip neighbours with our headstv8 days2-15/+37
|
* announcements answered only if heads differtv8 days4-7/+30
|
* retry new peers before probingtv8 days1-1/+1
|
* probe list memberstv8 days3-6/+17
|
* links: add failovertv8 days1-14/+129
|
* swarm: links by whom we reachtv8 days2-12/+25
|
* always announce new headstv8 days2-45/+9
|
* links: when asking for a lot relays, do chunktv8 days1-17/+72
|
* add budget for relayed LINKStv8 days1-1/+29
|
* add links relay testtv8 days1-0/+33
|
* swarm: relay links messagestv8 days2-24/+434
|
* swarm: announce to whom writers can't reachtv8 days2-28/+75
|
* dashboard: double-click opens peer dashboardtv8 days1-4/+61
|
* add per-node traffic countstv8 days1-1/+9
|
* probe when reachabletv9 days2-1/+26
|
* probe wakes peer workertv9 days1-2/+3
|
* document: introduce patch logtv9 days2-5/+5
|
* large messages over slow links are deltastv9 days2-7/+40
|
* links: wait for subscriber instead of tickertv9 days1-6/+16
|
* peer worker forget records on changetv9 days4-4/+28
|