continuum - P2P replicated document storage service

	Commit message (Collapse)	Author	Age	Files	Lines
*	unlisted members forgotton	tv	7 days	1	-6/+12
\|
*	handshakes counted	tv	7 days	1	-0/+7
\|
*	working links recorded	tv	7 days	1	-0/+1
\|
*	heads heard in announcements	tv	7 days	2	-3/+5
\|
*	heads heard in replies	tv	7 days	2	-1/+4
\|
*	heads heard forget sent	tv	7 days	2	-1/+31
\|
*	worker sends off the pass	tv	7 days	1	-14/+43
\|
*	suspend detected	tv	7 days	1	-0/+4
\|
*	rtt table pruned	tv	7 days	1	-0/+1
\|
*	one grace wait per peer	tv	7 days	1	-1/+3
\|
*	reach sets bounded	tv	7 days	1	-5/+11
\|
*	strangers messages small	tv	7 days	1	-0/+4
\|
*	lanes comment	tv	7 days	1	-2/+5
\|
*	round trips in trees	tv	8 days	1	-1/+8
\|
*	suspect paths	tv	8 days	1	-0/+5
\|
*	deflate blob chunks	tv	8 days	1	-1/+2
\|
*	epoch tag in sync	tv	8 days	1	-3/+21
\|
*	time blob messages	tv	8 days	1	-0/+19
\|
*	slow round back off	tv	8 days	1	-0/+5
\|
*	hello again when heard from	tv	8 days	2	-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 slot	tv	9 days	1	-0/+1
\| \| \| \| \| \|	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 then	tv	9 days	1	-3/+3
\|
*	pull after grace	tv	9 days	1	-1/+1
\| \| \| \| \| \| \| \|	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 lacking	tv	9 days	2	-5/+23
\| \| \| \| \| \| \|	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 reached	tv	10 days	1	-7/+18
\|
*	replace eager links by trees	tv	10 days	1	-8/+3
\|
*	measure 2-hop neighbourhood	tv	10 days	2	-3/+11
\|
*	introduce eager links	tv	10 days	1	-3/+8
\|
*	skip neighbours with our heads	tv	10 days	1	-0/+1
\|
*	announcements answered only if heads differ	tv	10 days	3	-2/+24
\|
*	retry new peers before probing	tv	10 days	1	-1/+1
\|
*	probe list members	tv	10 days	2	-6/+7
\|
*	add per-node traffic counts	tv	10 days	1	-1/+9
\|
*	probe when reachable	tv	11 days	2	-1/+26
\|
*	probe wakes peer worker	tv	11 days	1	-2/+3
\|
*	peer worker forget records on change	tv	11 days	3	-3/+17
\|
*	peer worker sleeps until deadline, not fixed time	tv	11 days	4	-14/+80
\|
*	state: drop epoch_tx	tv	11 days	1	-1/+2
\| \| \| \|	The stamp is the only lock-free epoch label.
*	rename AutoMerge to Document	tv	12 days	1	-2/+2
\|
*	add peer clock skew warning	tv	13 days	1	-0/+1
\|
*	pong carries time	tv	13 days	2	-4/+7
\|
*	dashboard: try to get links dormant immediately	tv	13 days	1	-6/+12
\|
*	centralize call to topology_change	tv	13 days	4	-27/+23
\|
*	unify request_sync logic	tv	13 days	4	-16/+4
\|
*	unify heard_from logic	tv	13 days	2	-23/+35
\|
*	unify closing logic	tv	13 days	1	-3/+3
\|
*	unify sync_on predicate	tv	13 days	1	-7/+2
\|
*	farewell means dormant	tv	13 days	1	-1/+9
\|
*	add links messages	tv	13 days	1	-0/+6
\| \| \| \| \| \| \| \| \|	Introduce two new wire messages, LINKS and LINKS_REPLY (swarm::links). A member answers with its links: for each peer, the transport, whether it’s relayed, the RTT, and whether it’s dormant. It answers members only, at most every 5 s. This bumps wire::VERSION to 7. Every build answers, with or without the feature, so every node needs this patch before the map shows anything beyond its own node.
*	immediately probe dormant members found in hellos	tv	13 days	1	-4/+37
\| \| \| \| \| \|	Instead of waiting for its regular timer, reprobe dormant members found in hello reply messages on the next worker tick. With this a partition gets reconnected quickly as soon as the first pair reconnects.