Methods for debugging your Redis Stack functions


There are two methods you can use to debug your Redis Stack functions:

  1. Make judicious use of the redis.log function, which writes to the Redis log file.
  2. Use Redis pub/sub.

Use redis.log

If you have access to the Redis log files, redis.log is a good method to use when debugging. However, there is a drawback. Redis Cloud users do not have access to the Redis log files, and it's pretty common that only system administrators have access to them on self-hosted installations. Fortunately, you can also use Redis pub/sub, which will be discussed in the next section.

You don't have to do anything special to use redis.log, as it's always available. Here is an example:

#!js api_version=1.0 name=lib

redis.registerFunction('hello', ()=> {
  redis.log('Hello log')
  return 'Hello from an external file'

After loading the library and executing the function with TFCALL, you'll see something like the following in your Redis log file:

45718:M 01 Nov 2023 07:02:40.593 * <redisgears_2> Hello log

Use Redis pub/sub

If you don't have access to your Redis database log files, you can use pub/sub. The following example demonstrates how to use the API to publish to a pub/sub channel.

#!js api_version=1.0 name=lib

const logChannel = 'tfLogChannel'

function publog(client, message) {'publish', logChannel, message)

redis.registerFunction('tflog', (client) => {
  publog(client, 'sample pub/sub log message')
  return 'sample'

In a CLI session, subscribe to the tfLogChannel channel and watch for messages.

$ redis-cli> subscribe tfLogChannel
1) "subscribe"
2) "tfLogChannel"
3) (integer) 1
Reading messages... (press Ctrl-C to quit or any key to type command)

In another CLI session, load the library, replacing what was already there, and then call the new function:> TFCALL lib.tflog 0

You'll see the following in the previous CLI session:

1) "message"
2) "tfLogChannel"
3) "sample pub/sub log message"

There is a downside to using pub/sub. Redis pub/sub provides at-most-once message delivery semantics, which means that once a message is sent, it won't be sent again. So, if a message isn't consumed, it's gone forever. This is likely okay for debugging, but for the longer term, redis.log is the better solution for log persistence.

Back to top ↑