instructure/canvas-lms · error · CanvasCache::Redis::UnsupportedRedisMethod

Redis method `# ` is not supported by Twemproxy, and so…

Error message

Redis method `#{command}` is not supported by Twemproxy, and so shouldn't be used in Canvas

What it means

canvas_cache's Twemproxy adapter blocks Redis commands that Twemproxy cannot route (per UNSUPPORTED_METHODS, e.g. KEYS, SCAN-like multi-node ops, transactions) by raising UnsupportedRedisMethod. Because Canvas runs Redis behind Twemproxy in production, unsupported commands would silently misbehave, so they're rejected with a clear message instead.

Solutions

Rewrite the code to use supported commands (e.g. SCAN via per-key patterns, pipelined gets) instead of blocked ones
Use a direct (non-Twemproxy) Redis connection for administrative/maintenance commands
Set RedisClient.dangerous_redis_methods_allowed? / run in GuardRail :deploy environment only if the command is genuinely needed and safe
Check UNSUPPORTED_METHODS in the gem to confirm which commands are blocked and pick alternatives

Example fix

// before
keys = redis.keys('course:cities:*')
// after
keys = redis.scan_each(match: 'course:cities:*').to_a # supported variant

Defensive patterns

Strategy: validation

Validate before calling

UNSAFE = %w[KEYS SCAN MULTI EXEC].freeze
raise UnsupportedRedisMethod, cmd if UNSAFE.include?(cmd.to_s.upcase)

Try / catch

begin
  redis.keys(pattern)
rescue CanvasCache::Redis::UnsupportedRedisMethod => e
  Rails.logger.warn("twemproxy: #{e.message}")
  fallback_supported_scan(pattern)
end

Prevention

Review canvas_cache UNSUPPORTED_METHODS before using exotic Redis commands
Prefer pipelined simple commands over multi-key/transactional ops
Test redis-dependent code against a Twemproxy-proxied instance

When it happens

Trigger: Calling an unsupported Redis command (e.g. redis.keys('*'), multi/exec, script ops in some setups) through a redis-client configured with the Twemproxy adapter; check_command is invoked by call and call_pipelined.

Common situations: Debugging in console against production-style proxy config; gems/libraries issuing commands that assume direct Redis access; using methods that work locally but not behind Twemproxy after deploying.

Related errors

AI-assisted analysis of instructure/canvas-lms@1c9f0bb801 (2026-09-15). Data as JSON: /api/errors/45e0861d33b1bcd3. Report an issue: GitHub.

Appendix: source

Thrown at gems/canvas_cache/lib/redis_client/twemproxy.rb:103

      slowlog
      sync
    ].freeze

    def call(command, _config)
      check_command(command.first.to_s)
      super
    end

    def call_pipelined(commands, _config)
      commands.each do |command|
        check_command(command.first.to_s)
      end
      super
    end

    def check_command(command)
      if UNSUPPORTED_METHODS.include?(command)
        raise CanvasCache::Redis::UnsupportedRedisMethod,
              "Redis method `#{command}` is not supported by Twemproxy, and so shouldn't be used in Canvas"
      end
      if ALLOWED_UNSUPPORTED.include?(command) && !RedisClient.dangerous_redis_methods_allowed? && GuardRail.environment != :deploy
        raise CanvasCache::Redis::UnsupportedRedisMethod,
              "Redis method `#{command}` is potentially dangerous, and should only be called from console, and only if you fully understand the consequences. If you're sure, retry after wrapping in `RedisClient.with_dangerous_redis_methods`"
      end
    end
  end
end

View on GitHub (pinned to 1c9f0bb801)