Read this on the main serverless docs site

AWS - Logs

Lets you watch the logs of a specific function or agent.

bash

# Lambda function logs
serverless logs -f hello

# Agent logs
serverless logs -a myAgent

# Optionally tail the logs with --tail or -t
serverless logs -f hello -t
serverless logs -a myAgent -t

This command returns as many log events as can fit in 1MB (up to 10,000 log events). You can use the --filter option to ensure the logs you're looking for are included.

Options

--function or -f The function you want to fetch the logs for. Required when not using --agent.
--agent or -a The agent name (for AgentCore Runtime agents). Required when not using --function. Cannot be used together with --function.
--stage or -s The stage you want to view the function logs for. If not provided, the plugin will use the default stage listed in serverless.yml. If that doesn't exist either it'll just fetch the logs from the dev stage.
--region or -r The region you want to view the function logs for. If not provided, the plugin will use the default region listed in serverless.yml. If that doesn't exist either it'll just fetch the logs from the us-east-1 region.
--aws-profile The AWS profile you want to use.
--startTime A specific unit in time to start fetching logs from (ie: 2010-10-20 or 1469705761). Without this, the default time is the past 10 minutes. Here's a list of the supported string formats:

bash

30m                   # since 30 minutes ago
2h                    # since 2 hours ago
3d                    # since 3 days ago

2013-02-08            # A calendar date part
2013-W06-5            # A week date part
2013-039              # An ordinal date part

20130208              # Basic (short) full date
2013W065              # Basic (short) week, weekday
2013W06               # Basic (short) week only
2013050               # Basic (short) ordinal date

2013-02-08T09         # An hour time part separated by a T
20130208T080910,123   # Short date and time up to ms, separated by comma
20130208T080910.123   # Short date and time up to ms
20130208T080910       # Short date and time up to seconds
20130208T0809         # Short date and time up to minutes
20130208T08           # Short date and time, hours only

--filter You can specify a filter string to filter the log output. This is useful if you want to to get the error logs for example.
--tail or -t You can optionally tail the logs and keep listening for new logs in your terminal session by passing this option.
--interval or -i If you choose to tail the output, you can control the interval at which the framework polls the logs with this option. The default is 1000ms.

Examples

AWS

Note: There's a small lag between invoking the function and actually having the log event registered in CloudWatch. So it takes a few seconds for the logs to show up right after invoking the function.

bash

serverless logs -f hello

This will fetch the logs from last 10 minutes as startTime was not given.

bash

serverless logs -f hello --startTime 5h

This will fetch the logs that happened in the past 5 hours.

bash

serverless logs -f hello --startTime 1469694264

This will fetch the logs that happened starting at epoch 1469694264.

bash

serverless logs -f hello -t

Serverless will tail the CloudWatch log output and print new log messages coming in starting from 10 seconds ago.

bash

serverless logs -f hello --filter serverless

This will fetch only the logs that contain the string serverless

Agents

Note: Agent logs come from CloudWatch log groups created by Bedrock AgentCore. The agent must have been invoked at least once for logs to be available.

bash

serverless logs -a myAgent

This will fetch agent logs from the last 10 minutes.

bash

serverless logs -a myAgent --startTime 1h

This will fetch agent logs from the past hour.

bash

serverless logs -a myAgent -t

Tail the agent's CloudWatch log output and print new log messages as they arrive.

bash

serverless logs -a myAgent --filter "ERROR"

This will fetch only the agent logs that contain the string ERROR.