index
:
continuum
master
P2P replicated document storage service
summary
refs
log
tree
commit
diff
stats
log msg
author
committer
range
path:
root
/
src
/
state.rs
Commit message (
Expand
)
Author
Age
Files
Lines
*
add node aliases
tv
19 hours
1
-0
/
+2
*
add throughput stats
tv
19 hours
1
-1
/
+14
*
add propagation stats
tv
19 hours
1
-1
/
+6
*
add per peer incident stats
tv
19 hours
1
-2
/
+38
*
add swarm stats
tv
19 hours
1
-0
/
+2
*
add traffic counter
tv
20 hours
1
-0
/
+2
*
assess swarm stability
tv
20 hours
1
-0
/
+2
*
record RTT sampling instant
tv
20 hours
1
-0
/
+6
*
prefer closer sources to catch up
tv
21 hours
1
-0
/
+6
*
record RTT from keepalive messages
tv
21 hours
1
-6
/
+9
*
pull missing changes
tv
21 hours
1
-0
/
+2
*
find reachable neighbours
tv
21 hours
1
-0
/
+30
*
sync with neighbours
tv
21 hours
1
-0
/
+18
*
use nonce in handshake
tv
22 hours
1
-0
/
+3
*
consolidate backoff code
tv
22 hours
1
-20
/
+25
*
retry using fresh messages
tv
23 hours
1
-11
/
+52
*
heal from noticed divergence
tv
23 hours
1
-6
/
+27
*
heal after reply timeout
tv
25 hours
1
-0
/
+3
*
drop stall breaker; resend needed changes
tv
25 hours
1
-17
/
+0
*
both sides may open sync after reset
tv
26 hours
1
-121
/
+26
*
perform handshake in one round trip
tv
27 hours
1
-5
/
+0
*
drop pre-membership sync messages
tv
27 hours
1
-4
/
+0
*
bump number of heads in keepalive messages
tv
27 hours
1
-0
/
+4
*
probe sync source during catch-up
tv
27 hours
1
-3
/
+14
*
enrich keepalive content
tv
27 hours
1
-5
/
+9
*
use only ready peers as sync source
tv
28 hours
1
-0
/
+4
*
catch up with a single source only
tv
28 hours
1
-0
/
+15
*
introduce announce opening variant
tv
29 hours
1
-6
/
+14
*
record the cause that started a sync session
tv
29 hours
1
-6
/
+59
*
merge sync_reset_at and epoch_unannounced into opening
tv
29 hours
1
-47
/
+64
*
introduce sync session
tv
30 hours
1
-81
/
+90
*
merge membership and authenticate into standing
tv
30 hours
1
-13
/
+27
*
announce epochs by sync instead of ping
tv
31 hours
1
-4
/
+29
*
pick pair leader by hash instead of key order
tv
36 hours
1
-2
/
+2
*
ping only from one side
tv
36 hours
1
-2
/
+20
*
open sync only from one side
tv
43 hours
1
-0
/
+24
*
don't resend sync messages to up-to-date peers
tv
2 days
1
-0
/
+6
*
introduce pre-shared swarm key
tv
2 days
1
-0
/
+9
*
rename cluster to swarm everywhere!
tv
3 days
1
-59
/
+47
*
resend sync messages that haven't been replied
tv
3 days
1
-0
/
+67
*
add heads to reads
tv
4 days
1
-5
/
+39
*
rename compaction to remake
tv
4 days
1
-6
/
+6
*
reorganize all the modules
tv
4 days
1
-4
/
+4
*
serve stale reads during rejoin
tv
5 days
1
-0
/
+69
*
take_keepalive_due: consider added_at
tv
5 days
1
-0
/
+1
*
replace snapshots by quarantine & replay
tv
5 days
1
-11
/
+31
*
fix catch-up / compaction in general
tv
5 days
1
-2
/
+19
*
fix catch-up
tv
5 days
1
-4
/
+24
*
add epoch catch-up / snapshots
tv
5 days
1
-2
/
+28
*
add epoch switch
tv
5 days
1
-1
/
+67
[next]