pcluster get-cluster-log-events - Amazon ParallelCluster
Named arguments
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

pcluster get-cluster-log-events

Retrieve the events associated with a log stream.

pcluster get-cluster-log-events [-h] 
                 --cluster-name CLUSTER_NAME
                 --log-stream-name LOG_STREAM_NAME
                [--debug]
                [--end-time END_TIME]
                [--limit LIMIT]
                [--next-token NEXT_TOKEN]
                [--query QUERY]
                [--region REGION]
                [--start-from-head START_FROM_HEAD]                                
                [--start-time START_TIME]

Named arguments

-h, --help

Shows the help text for pcluster get-cluster-log-events.

--cluster-name, -n CLUSTER_NAME

Specifies the name of the cluster.

--log-stream-name LOG_STREAM_NAME

Specifies the name of the log stream. You can use the list-cluster-log-streams command to retrieve a log stream associated with an event or events.

--debug

Enables debug logging.

--end-time END_TIME

Specifies the end of the time range, expressed in ISO 8601 format (YYYY-MM-DDThh:mm:ssZ, for example 2021-01-01T20:00:00Z). Events with a timestamp equal to or later than this time are not included.

--limit LIMIT

Specifies the maximum number of log events returned. If a value is not specified, the maximum is as many log events as can fit in a response size of 1 MB, up to 10,000 log events.

--next-token NEXT_TOKEN

The token for the next set of results.

--query QUERY

Specifies the JMESPath query to perform on the output.

--region, -r REGION

Specifies the Amazon Web Services Region to use. The Amazon Web Services Region must be specified, using the AWS_DEFAULT_REGION environment variable, the region setting in the [default] section of the ~/.aws/config file, or the --region parameter.

--start-from-head START_FROM_HEAD

If the value is true, the earliest log events are returned first. If the value is false, the most recent log events are returned first. (Defaults to false.)

--start-time START_TIME

Specifies the start of the time range, expressed in ISO 8601 format (YYYY-MM-DDThh:mm:ssZ, for example 2021-01-01T20:00:00Z). Events with a timestamp equal to this time or later than this time are included.

Example using Amazon ParallelCluster version 3.1.4:

$ pcluster get-cluster-log-events \
    -c cluster-v3 \
    -r us-east-1 \
    --log-stream-name ip-198-51-100-44.i-1234567890abcdef0.clustermgtd \
    --limit 3
{
  "nextToken": "f/36966906399261933213029082268132291405859205452101451780/s",
  "prevToken": "b/36966906399239632467830551644990755687586557090595471362/s",
  "events": [
    {
      "message": "2022-07-12 19:16:53,379 - [slurm_plugin.clustermgtd:_maintain_nodes] - INFO - Performing node maintenance actions",
      "timestamp": "2022-07-12T19:16:53.379Z"
    },
    {
      "message": "2022-07-12 19:16:53,380 - [slurm_plugin.clustermgtd:_maintain_nodes] - INFO - Following nodes are currently in replacement: (x0) []",
      "timestamp": "2022-07-12T19:16:53.380Z"
    },
    {
      "message": "2022-07-12 19:16:53,380 - [slurm_plugin.clustermgtd:_terminate_orphaned_instances] - INFO - Checking for orphaned instance",
      "timestamp": "2022-07-12T19:16:53.380Z"
    }
  ]
}