aboutsummaryrefslogtreecommitdiffstats
path: root/src/swarm
Commit message (Collapse)AuthorAgeFilesLines
* rename PING/PONG to STATUS/STATUS_REPLYtv13 days9-70/+72
|
* build topology per transporttv14 days1-4/+128
|
* use libp2p mDNStv14 days1-0/+11
|
* add support for several transportstv14 days1-1/+7
|
* admit nodeId and addresses in the APItv14 days1-1/+8
|
* add identity layertv14 days1-2/+6
|
* introduce addressestv14 days4-16/+100
|
* introduce node keystv14 days1-6/+6
|
* talk about "transport" unless mycelium is meanttv14 days9-32/+37
|
* add smaller topology testtv2026-10-261-19/+35
|
* reject and warn about too large delta blobstv2026-10-262-0/+9
|
* add delta blobstv2026-10-263-19/+312
|
* make clippy happy againtv2026-10-265-35/+35
|
* warn about dropped blob requeststv2026-10-261-0/+28
|
* add sync and relay statstv2026-10-263-1/+44
|
* introduce bounded blob downloadstv2026-10-261-13/+41
|
* warn if swarm key is readable by otherstv2026-10-261-0/+22
|
* add farewell pingtv2026-10-262-2/+22
|
* decode sync messages lazilytv2026-10-261-1/+3
| | | | | | | Sync messages are now being decoded only where they are applied: after membership has been checked, in the sync lane's blocking task. Debug now shows only the size instead of the whole message.
* keepalive sends head samples alongside digesttv2026-10-261-21/+22
| | | | | | When there are more than 32 heads that can be send in a single keepalive message, the digest is accompanied by 32 heads, a different window each time, so after a few keepalives every head has been listed.
* add pull queuetv2026-10-263-34/+84
|
* send sync messages off workertv2026-10-261-1/+23
|
* read stamps sync and lock freetv2026-10-264-13/+20
|
* add sync lanestv2026-10-263-2/+70
|
* compute member pair scores cheaplytv2026-10-261-22/+65
|
* unreachability doesn't expiretv2026-10-262-9/+3
|
* forget records only oncetv2026-10-262-9/+37
|
* fix race between alive check and forgettingtv2026-10-262-8/+23
|
* forgotten nodes cancel their own recordstv2026-10-261-15/+14
|
* add session IDstv2026-10-261-15/+23
| | | | | This prevents the regenerate flag from being lost across reset_sync during an in-flight send.
* number handshakestv2026-10-262-7/+32
| | | | This prevents late hellos from resetting the session.
* keep silent strangers for a daytv2026-10-262-3/+40
|
* send bounded peer liststv2026-10-262-7/+38
|
* wrap unix timestamps in newtypetv2026-10-263-3/+3
|
* allow humand-friendly durations and sizes in CLItv2026-10-262-2/+2
|
* SyncSessions owns Automerge's sync statetv2026-10-262-80/+12
|
* unify keepalivetv2026-10-261-15/+23
|
* split peer moduletv2026-10-269-915/+995
|
* unify peers snapshot logictv2026-10-261-11/+3
|
* unify lock logictv2026-10-2610-70/+31
|
* move protocol paths to dedicated moduletv2026-10-263-7/+7
|
* refresh peers after stalltv2026-10-252-1/+19
|
* add random swarms testtv2026-10-251-0/+62
|
* add clock moduletv2026-10-253-6/+11
| | | | | The Clock module is the system clock that every unix timestamp new goes through. It comes with an offset that only tests can set.
* avoid stalled sourcetv2026-10-251-2/+24
| | | | | | | | | | | | | 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_limittv2026-10-251-6/+3
|
* recompute neighbours whenever topology changestv2026-10-254-5/+40
|
* run peer worker when becoming readytv2026-10-251-0/+8
|
* newly added neighbour get a sync attempttv2026-10-251-3/+12
|
* add simulation testtv2026-10-251-1/+4
|