Repository navigation
Change the issues type filter
Issues
- #49 · renovate opened
on Mar 19, 2026 2
is:issue state:open
is:issue state:open
Issues
Removed node's transaction log is replayed from position 0 on every resubscribe; idle peers' cursors never advance and force a base copy
Filter by Priority: P1Declarative residency policy so threads without application code shard correctly
Filter by Priority: P21Record locks: enabling replication.recordLocks moves all of a database's inbound apply onto one worker, even with no lock use
Filter by Priority: P21Replication: operations from a node principal bypass all permission checks, so a node identity is cluster-wide super_user
Filter by Priority: P3HarperFast/harper-pro#976 · openedon Oct 5, 2026 Run replication on a dedicated worker pool (replication.threads)
Filter by Priority: P1Replication places subscriptions and record-lock ownership on isolated application workers
Filter by Priority: P2Record locks: record_lock_apply_homes cannot complete a transition once a ring member is already gone
Filter by Priority: P2Forwarded replication operations leak requestId into the operation body, breaking replicated set_configuration
Filter by Priority: P2cloneNode: leaderRequest has no response deadline, so a leader that accepts and never answers hangs clone setup forever
Filter by Priority: P3Replication failover: a deferred subscription dispatch consumes its self-catchup claim, so the catch-up range can be lost entirely
Filter by Priority: P2Pace replication sender yields by elapsed time to reduce per-record CPU overhead
Filter by Priority: P2Replication should treat a peer whose database generation changed as a new incarnation, so a restored node does not silently diverge
Filter by Priority: P21Base copy tail replay can overwrite an already-copied record's blob, leaving it permanently 503 while the copy reports complete
Filter by Priority: P1Outbound replication stalls silently when a native transaction handle leaks
Filter by Priority: P2WAF shadow-rule would-block events are logged at info and dropped at the default warn log level
Filter by Priority: P2Contract: do base-copy-acquired rows propagate onward over a live subscription?
Filter by Priority: P3Legacy v4 config-validator ReDoS wedges worktree test setups
Filter by Priority: P3getReplicationSharedStatus allocates per call on the blob-retrieval request path
Filter by Priority: P3Validate replication_failedFrameReplays and test the copy-apply split exemption
Filter by Priority: P3Cluster test restart helpers swallow real errors and fail 60 s later on the wrong cause
Filter by Priority: P3Delete-echo repair tool can hang holding the database lock during read-back
Filter by Priority: P2v5.1/v5.2: harden receive-side frame-turn handling around connection close
Filter by Priority: P3Add per-origin closed timestamp floors for compact, cross-node subscription resume
Filter by Priority: P3receiveWatchdog.test.mjs leaves a fake setTimeout behind for any later default sinon.restore(), hanging the unit run
Filter by Priority: P3Deploy the replicated-deploy test fixture as harper once harper#2818 reaches this repo's core
Filter by Priority: P2