Issues

• Cluster: keys routed to another shard's queue are dropped or sent to the wrong node when a refresh moves a slot, retires its owner or demotes it

  redis/memtier_benchmark#889 · Filter by author fcostaoliveira opened on Oct 25, 2026

• Cluster --monitor-input: when a slot-map refresh retires or demotes a primary, its staged monitor lines are silently dropped and replaced (or sent to the demoted node)

  redis/memtier_benchmark#836 · Filter by author fcostaoliveira opened on Oct 24, 2026

• Cluster --monitor-input: a staged monitor command is sent to the connection it was staged on even after a newer slot map moved its slot. The MOVED that follows flushes the rest of that queue.

  redis/memtier_benchmark#835 · Filter by author fcostaoliveira opened on Oct 24, 2026
  1

• Flaky tests seen in CI (prometheus latency cross-check, CLIENT KILL timing, injector compile timeout, ASK reconnect timeout)

  redis/memtier_benchmark#740 · Filter by author fcostaoliveira opened on Oct 16, 2026
  1

• --reconnect-on-error with --requests and unlimited --max-reconnect-attempts: a connection whose setup keeps being rejected reconnects forever

  redis/memtier_benchmark#693 · Filter by author fcostaoliveira opened on Oct 16, 2026

• Cluster --read-preference secondary: an idle connection to a primary that dies is never noticed, so the slot map is never refreshed

  redis/memtier_benchmark#689 · Filter by author fcostaoliveira opened on Oct 16, 2026

• Cluster --read-preference secondary: read-only runs never refresh the slot map, so reads keep going to a replica promoted by CLUSTER FAILOVER (intended?)

  redis/memtier_benchmark#678 · Filter by author fcostaoliveira opened on Oct 15, 2026

• Cluster --reconnect-on-error: when the seed node (-s/-p) dies, the whole client stops even after its replica is promoted

  redis/memtier_benchmark#677 · Filter by author fcostaoliveira opened on Oct 15, 2026

• Cluster --read-preference: a write sent to a replica gets -MOVED to the slot's own primary, and every such reply triggers a CLUSTER SLOTS refresh

  redis/memtier_benchmark#664 · Filter by author fcostaoliveira opened on Oct 15, 2026

• Cluster mode without --retry-on-error: a -MOVED/-ASK reply is counted as a completed op (Count > commands the server ran) - intended?

  redis/memtier_benchmark#650 · Filter by author fcostaoliveira opened on Oct 14, 2026

• --transaction with --requests: run can end after a MULTI without its EXEC when --requests is not a multiple of the block length

  redis/memtier_benchmark#647 · Filter by author fcostaoliveira opened on Oct 14, 2026

• TLS cluster: worker stalls after reconnects (unsent output, handshake never completes)

  redis/memtier_benchmark#562 · Filter by author fcostaoliveira opened on Oct 12, 2026
  1

• Verify whether bufferevent_write is actually required for READONLY/CLIENT NO-TOUCH on discovered connections, or the original diagnosis was wrong

  redis/memtier_benchmark#532 · Filter by author fcostaoliveira opened on Aug 28, 2026
  1

• All CI test cells depend on a personal fork's availability (tests/test_requirements.txt -> fcostaoliveira/RLTest)

  redis/memtier_benchmark#530 · Filter by author fcostaoliveira opened on Aug 28, 2026

• Investigate whether peer_client_has_any_setup_in_progress() gating on is_ready_for_reads() (not is_conn_setup_done()) can livelock read routing

  redis/memtier_benchmark#529 · Filter by author fcostaoliveira opened on Aug 28, 2026

• Consider a generic --connection-setup-command flag instead of per-command booleans

  redis/memtier_benchmark#528 · Filter by author fcostaoliveira opened on Aug 27, 2026
  2

• Support CLIENT NO-TOUCH as a connection-setup command

  redis/memtier_benchmark#525 · Filter by author fcostaoliveira opened on Aug 27, 2026

• Customers commonly exhaust AWS burst credits on Elasticache. Need to measure network, cpu and iops against ec2 instance limits.

  redis/memtier_benchmark#514 · Filter by author alberttwong opened on Jul 16, 2026

• Pre-existing latent hang in zipfian arbitrary-command path (test_zipfian_arbitrary_command_mixed_operations) under UBSan

  redis/memtier_benchmark#482 · Filter by author fcostaoliveira opened on Jun 17, 2026

• follow-up: monitor + keyless arbitrary read-pref improvements

  redis/memtier_benchmark#457 · Filter by author fcostaoliveira opened on Jun 5, 2026

• Support multi-endpoint benchmarks for Redis Active-Active write/read testing

  redis/memtier_benchmark#452 · Filter by author alberttwong opened on Jun 3, 2026

• monitor-input: downstream c_str() re-truncates NUL bytes in arbitrary command dispatch

  redis/memtier_benchmark#439 · Filter by author fcostaoliveira opened on Jun 2, 2026

• CLI-fuzz follow-ups: 17 crash/hang classes (11 share connection-stage retry-loop root cause)

  redis/memtier_benchmark#426 · Filter by author fcostaoliveira opened on Jun 1, 2026 · 2.4

• Crash handler is not async-signal-safe (SA_ONSTACK set, sigaltstack never called) — truncates under corrupted-stack crashes

  redis/memtier_benchmark#412 · Filter by author fcostaoliveira opened on May 30, 2026 · 2.4

• Audit: latent stack-allocation foot-guns (incl. HIGH-severity unbounded sscanf in memcached parser)

  redis/memtier_benchmark#406 · Filter by author fcostaoliveira opened on May 30, 2026 · 2.4