continuum - P2P replicated document storage service

	Commit message (Collapse)	Author	Age	Files	Lines
*	offer again ungenerated	tv	8 days	2	-32/+139
\|
*	stale delta offers	tv	8 days	2	-0/+79
\|
*	blob window grows	tv	8 days	1	-5/+10
\|
*	delta known to sender	tv	8 days	2	-6/+14
\|
*	delta propagation	tv	8 days	2	-2/+18
\|
*	benchmark bridge fast link	tv	8 days	1	-17/+22
\|
*	sim links push back	tv	8 days	1	-18/+33
\|
*	benchmark bridge slow link	tv	8 days	1	-8/+19
\|
*	benchmark bridge	tv	8 days	1	-0/+120
\|
*	sim link rates	tv	8 days	1	-1/+49
\|
*	one delta at a time	tv	8 days	2	-9/+53
\|
*	time blob messages	tv	8 days	8	-6/+55
\|
*	log why a delta	tv	8 days	1	-0/+6
\|
*	offered deltas lapse	tv	8 days	2	-7/+52
\|
*	fix: level means the peer has our heads	tv	8 days	1	-6/+13
\|
*	fix: test links fail over cuts by name	tv	8 days	1	-5/+6
\|
*	dashboard: hop delays	tv	8 days	1	-3/+35
\|
*	throughput from delivered message	tv	8 days	2	-2/+59
\|
*	hop delays	tv	8 days	8	-3/+178
\|
*	test links fail over	tv	8 days	1	-0/+105
\|
*	benchmark members joining	tv	8 days	1	-5/+48
\|
*	benchmark delays by path	tv	9 days	3	-0/+28
\|
*	tests count the changes sent	tv	9 days	1	-5/+7
\|
*	fix: change publisher stops with the runtime	tv	9 days	1	-3/+8
\|
*	no reset for heads still on their way	tv	9 days	1	-1/+44
\|
*	benchmark what sync bytes are	tv	9 days	4	-3/+74
\|
*	changes from a peer never go back	tv	9 days	2	-0/+52
\|
*	older delivered messages leave us level	tv	9 days	2	-36/+76
\|
*	benchmark where twice comes from	tv	9 days	3	-6/+20
\|
*	resend only what is needed	tv	9 days	2	-27/+62
\|
*	benchmark delivered and twice	tv	9 days	3	-0/+58
\|
*	delivered means has all up to our heads	tv	9 days	1	-13/+59
\|
*	any message taken as delivered	tv	9 days	5	-40/+112
\|
*	slow round back off	tv	9 days	5	-5/+92
\|
*	tree messages need no answer	tv	9 days	5	-7/+235
\|
*	hello again when heard from	tv	9 days	3	-2/+71
\| \| \| \| \| \|	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.
*	bump pulled_changes_go_along_their_writers_trees	tv	9 days	1	-0/+3
\|
*	unanswered pull frees the slot	tv	9 days	2	-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.
*	no heal while both move on	tv	9 days	1	-3/+49
\| \| \| \| \| \| \|	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.
*	bump pulled_changes_go_along_their_writers_trees	tv	9 days	1	-2/+52
\|
*	overdue sync asked after from then	tv	9 days	3	-49/+92
\|
*	pull after grace	tv	9 days	2	-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 lacking	tv	9 days	5	-12/+36
\| \| \| \| \| \| \|	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 sync asked after	tv	9 days	1	-6/+28
\| \| \| \| \|	An overdue sync answer is now followed by a PING too, and the peer stays reached once it answers the PING.
*	simtest: benchmark incidents	tv	10 days	1	-6/+21
\|
*	unanswered members not reached	tv	10 days	5	-18/+132
\|
*	add member away benchmark	tv	10 days	1	-0/+34
\|
*	dashboard: show reach set	tv	10 days	1	-2/+23
\|
*	links: only news pass on	tv	10 days	1	-6/+12
\|
*	dashboard: show writer trees	tv	10 days	3	-26/+189
\|