continuum - P2P replicated document storage service

	Commit message (Collapse)	Author	Age	Files	Lines
*	split state moduleHEAD master	tv	14 hours	7	-1297/+1362
\|
*	make test RNG configurable via env	tv	14 hours	1	-3/+12
\|
*	add replay test	tv	14 hours	1	-0/+65
\|
*	add heads summary test	tv	14 hours	1	-0/+37
\|
*	add random swarms test	tv	14 hours	4	-32/+149
\|
*	add soak test	tv	14 hours	2	-0/+102
\|
*	add clock module	tv	14 hours	18	-75/+164
\| \| \| \| \|	The Clock module is the system clock that every unix timestamp new goes through. It comes with an offset that only tests can set.
*	add stalled source test	tv	15 hours	2	-3/+40
\|
*	avoid stalled source	tv	15 hours	3	-2/+28
\| \| \| \| \| \| \| \| \| \| \| \| \|	A source given up for not answering our sync messages could become the source again straight away. Whoever asks first after the reset takes the role, and that's often the same peer, which would then stall for another delivery limit. Now, a stalled source isn't chosen again for one more delivery_limit. If it's the only candidate, catching up waits that long and then tries it again. Before, a retry with that peer couldn't have helped any sooner anyway: a source is only given up after a whole delivery limit without a sync message, and that clock restarts when it is chosen again.
*	MYCELIUM_DELIVERY_LIMIT -> Timing::delivery_limit	tv	15 hours	3	-7/+16
\|
*	add convergence over lossy-links test	tv	15 hours	1	-0/+33
\|
*	make sync reply timeout part of timing	tv	15 hours	4	-16/+42
\| \| \| \|	This enables upcoming lossy-links tests.
*	recompute neighbours whenever topology changes	tv	15 hours	6	-5/+47
\|
*	run peer worker when becoming ready	tv	15 hours	1	-0/+8
\|
*	newly added neighbour get a sync attempt	tv	16 hours	1	-3/+12
\|
*	add simulation test	tv	16 hours	7	-8/+976
\|
*	build node in separate module	tv	16 hours	2	-110/+199
\|
*	fix warning about swarm key (fixup)	tv	17 hours	1	-1/+1
\|
*	require catch-up source to make progress	tv	17 hours	3	-1/+98
\|
*	late messages don't flip peer's epoch back	tv	17 hours	7	-25/+129
\|
*	bump abort margin & allow aborting active remake	tv	17 hours	3	-42/+179
\|
*	build_genesis -> remake_genesis (fixup)	tv	17 hours	3	-3/+3
\|
*	only keep list of actor IDs within one epoch (fixup)	tv	17 hours	3	-2/+14
\|
*	allow to remake genesis	tv	17 hours	5	-1/+75
\|
*	allow to persist document	tv	17 hours	9	-84/+670
\|
*	nodes keep list of every actor ID they've used	tv	18 hours	6	-20/+56
\|
*	explain that changes get replayed by all nodes (fixup)	tv	18 hours	1	-2/+2
\|
*	add template importer to config	tv	24 hours	1	-6/+107
\|
*	add force module usable by API and CLI	tv	24 hours	5	-54/+121
\|
*	add files module to consistently track stats	tv	24 hours	8	-143/+268
\|
*	warn about unresolvable seeds once; log success	tv	24 hours	1	-7/+22
\|
*	genesis conflicts seeds or recover-from	tv	24 hours	1	-16/+19
\| \| \| \| \|	genesis together with seeds or recover-from is now an error wherever the values come from, and not settled by precedence.
*	add disk I/O stats	tv	24 hours	6	-27/+138
\|
*	add possibility to start with a document	tv	25 hours	4	-13/+163
\| \| \| \|	The --recover-from option allows a node start with a given document.
*	add state file	tv	26 hours	7	-4/+266
\| \| \| \| \|	This allows storing the swarm ID and additionally discovered seeds, which prevents a formerly genesis node to start a new document.
*	put swarm ID into the document	tv	26 hours	4	-6/+27
\|
*	add genesis flag and never give up	tv	26 hours	3	-76/+112
\| \| \| \| \| \| \| \| \|	The --genesis flag has been introduced to start with a new document. This prevents restarts to accidentally start with a new document and accept reads/writes, causing a second swarm to appear. Additionally, bootstrapping doesn't timeout anymore as that's of no use for nobody.
*	compact remake fixes base at the commit	tv	39 hours	5	-40/+138
\| \| \| \| \| \| \|	Instead of fixing base heads at when proposed, compact fixes them when committing. This shrinks the window where writes can be lost if their author restarts to the switch delay; during replace and import remakes writes can still be lost during the voting period.
*	fix wording around swarm key admission	tv	40 hours	2	-3/+3
\|
*	allow superseding expired remakes	tv	40 hours	4	-26/+279
\|
*	remake coordinators resume after restart	tv	40 hours	3	-5/+49
\|
*	don't commit expired remakes	tv	41 hours	1	-58/+93
\|
*	add votes register	tv	41 hours	8	-124/+507
\|
*	bound stable-hub degree	tv	42 hours	1	-4/+46
\| \| \| \| \| \|	Each stable member can be preferred at most by 4*k members. The budget is spent in key order, which is the same on every node, so the graph stays deterministic, symmetric and connected.
*	coalesce announcements	tv	42 hours	3	-2/+60
\|
*	cache heads	tv	42 hours	7	-24/+61
\| \| \| \|	To reduce the number of write locks, heads are now cached.
*	corroberate newer epochs	tv	42 hours	1	-11/+53
\|
*	follow and download from ready members only	tv	42 hours	3	-3/+29
\|
*	being caught up needs some heads	tv	43 hours	1	-5/+9
\| \| \| \| \| \| \| \| \| \| \|	A sync message from a ready peer only completes the initial sync if that peer has at least one head. A peer with an empty document can no longer make a joiner ready on an empty store: a misconfigured second swarm, or a genesis node in its first seconds. Effect on a brand-new swarm: a joiner now waits until the genesis node has registered itself in the member list, which happens within about 5 s of it becoming ready.
*	only protocol-writers may modify _continuum	tv	43 hours	6	-14/+63
\|