WATCH

Monitors changes to keys to determine the execution of a transaction.

Browse all commands · Upstream source documentation

Reference field	Value
Canonical name	WATCH
Catalog group	Transaction commands
Upstream since	2.2.0 — upstream Redis/module version, not an ffkv release
Pinned catalog	Redis 8.10.0

Syntax

WATCH key [key ...]

ffkv behavior and limits

Use one persistent connection for MULTI/EXEC and WATCH. Runtime command errors do not roll back earlier effects. Blocking list/stream commands do not wait inside EXEC. A disconnected client cannot infer whether a submitted write committed.

Source: README.md:389–419, 125; tests/bdd/features/transactions.feature; tests/bdd/features/transactions_runtime.feature.

All requests share the 8 MiB framing/1,024-argument ingress limit; normal RESP replies have 64 MiB wire/logical budgets with bounded nesting/value counts. Management and Lua origins impose smaller budgets. These are not a process RSS cap. Catalog inclusion is not exhaustive option or wire-format parity. See limits and compatibility.

Upstream metadata

The following fields come from the pinned Redis metadata, not measured ffkv performance or an assertion that ffkv enforces every upstream flag. Running COMMAND INFO/COMMAND DOCS describe the current engine.

Upstream Redis complexity: O(1) for every key.

Upstream arity: -2 (includes the command name; a negative number is a minimum).

Command flags: NOSCRIPT, LOADING, STALE, FAST, ALLOW_BUSY.

ACL categories: TRANSACTION.

Argument structure

Argument	Type	Constraints / metadata
key	key	multiple; key_spec_index: 0

Key specifications

Upstream key discovery descriptors (indexes are relative to the full command vector):

[
  {
    "flags": [
      "RO"
    ],
    "begin_search": {
      "index": {
        "pos": 1
      }
    },
    "find_keys": {
      "range": {
        "lastkey": -1,
        "step": 1,
        "limit": 0
      }
    }
  }
]

Additional pinned upstream metadata

{
  "group": "transactions",
  "since": "2.2.0"
}

Metadata source · Local snapshot: src/command_metadata.json, key watch.

BDD coverage and source examples

These are source observations, not a test run performed by this documentation build. A tag identifies scenario intent; an extracted vector identifies a command invocation, which may intentionally produce an error. Scenarios may require setup, multiple connections, persistence, cluster nodes or native-module fixtures.

• An applied bounded increment invalidates a watching transaction — tests/bdd/features/strings.feature:1092; invocation observed.
• A competing write aborts WATCH without applying queued changes — tests/bdd/features/transactions.feature:61; command tag and invocation observed.
• UNWATCH permits a later transaction after the watched value changes — tests/bdd/features/transactions.feature:113; invocation observed.
• DISCARD removes the queue and watches but not committed values — tests/bdd/features/transactions.feature:168; invocation observed.

Extracted command vectors

The following JSON arrays are the test harness’s argument vectors, not shell commands. Numeric arguments are converted by the harness; objects of the form {"base64":"…"} represent decoded binary bytes, not a literal JSON command argument. Preserve scenario setup and connection ordering. See the CLI guide for shell quoting and binary stdin. No reply is invented or inferred.

A competing write aborts WATCH without applying queued changes — tests/bdd/features/transactions.feature:67 (client “buyer” executes:)

[
  "WATCH",
  "stock"
]

Next source assertion: the reply equals:

"OK"

An applied bounded increment invalidates a watching transaction — tests/bdd/features/strings.feature:1098 (client “watcher” executes:)

[
  "WATCH",
  "counter"
]

Next source assertion: the reply equals:

"OK"