aboutsummaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
* pick pair leader by hash instead of key ordertv11 days5-4/+83
|
* ping only from one sidetv11 days2-3/+22
|
* allow changing log filtering while node runstv11 days4-1/+97
| | | | This also changes the default to tracing-subscriber.
* consistently use the words node, peer, and seedtv11 days7-12/+18
|
* open sync only from one sidetv11 days2-3/+45
| | | | | | | | | If both sides open, the openings cross, and an empty node's answer to the other opening (empty heads) makes the other side send the whole document a second time. Now after a sync reset, only the side with the smaller public key opens the sync; the other answers it, or opens itself if nothing arrives.
* don't resend sync messages to up-to-date peerstv11 days3-9/+34
| | | | | | | | | | | When an sync messages arrives and automerge has nothing to say about it, then no reply is generated. With c9279e4 it was assumed that every sync message gets an answer, so missing replies caused unnecessary resends. Now a reply is only awaited if the peer's last reported heads differ from ours. If the peer's heads become equal through a third peer without replying, the waiting will also abort if a ping or pong showss the same heads digest on both sides.
* introduce pre-shared swarm keytv11 days7-1/+316
|
* rename cluster to swarm everywhere!tv13 days14-177/+162
|
* make remake switch delay configurabletv13 days3-2/+44
|
* install signal handlers early & handle SIGTERMtv13 days2-17/+58
|
* handle SIGTERMtv13 days1-0/+11
|
* let linux signal supervised mycelium when we dietv13 days1-8/+22
|
* resend sync messages that haven't been repliedtv13 days3-6/+90
|
* add the replace kind of remaketv13 days4-4/+256
| | | | | | | | | | | | | | POST /admin/remakes accepts specs with replace kind, e.g.: { "kind": "replace", "baseHeads": ["..."], "spec": { "document": { "accounts": { ... } }, "rules": [ { "from": "users", "to": "accounts" }, { "from": "users/admin", "drop": true }, { "from": "", "to": "" } ] } }
* allow choosing the base in remakestv13 days2-5/+54
| | | | | | | | | POST /admin/remakes accepts "baseHeads": ["<hex>", ...], for example the heads of an earlier GET /store?heads=true. Every head must be in the coordinating node's document, otherwise the request gets a 400. Without it, the base is the current heads. This works for any kind.
* add heads to readstv13 days2-9/+67
| | | | | | | | | | GET /store?heads=true and GET /store/{key}?heads=true add the header x-continuum-heads (sorted, comma-separated hex). The heads come from the same moment as the read. - It's opt-in because reading the heads takes the write lock. Reads without the flag don't change. - Stale reads during a rejoin never carry heads.
* turn compaction into a kind of remaketv13 days9-72/+589
|
* add store::copy_path_to(src, from, dst, to)tv13 days1-4/+14
|
* rename compaction to remaketv13 days16-175/+179
|
* reorganize all the modulestv13 days27-44/+55
|
* serve stale reads during rejointv2026-10-205-14/+118
|
* bootstrap: mark node ready only after epoch checktv2026-10-201-1/+3
|
* take_keepalive_due: consider added_attv2026-10-201-0/+1
| | | | This prevents an immediate ping during handshake.
* cleanup loggingtv2026-10-205-56/+69
|
* replace snapshots by quarantine & replaytv2026-10-2013-600/+789
|
* fix catch-up / compaction in generaltv2026-10-205-20/+51
|
* fix catch-uptv2026-10-207-94/+427
|
* add epoch catch-up / snapshotstv2026-10-2013-34/+831
|
* add force param to POST /admin/compactiontv2026-10-201-5/+21
|
* fix compactiontv2026-10-202-26/+97
| | | | | - introduce DELETE /admin/compaction to abort local compaction proposal - continue coordination if a POST /admin/compaction?wait=true terminates
* add epoch switchtv2026-10-2013-12/+1813
|
* add compaction coordinatortv2026-10-205-5/+148
|
* add compactiontv2026-10-204-0/+595
|
* cargo clippytv2026-10-191-1/+1
|
* cargo fmttv2026-10-196-21/+37
|
* add epochstv2026-10-1910-43/+431
|
* initial committv2026-10-1914-0/+4224