Job Information

job-summary-<job name>

**Table 1** Fields in job-summary-<*job name*> ConfigMap
Parameter	Description	Value
hccl.json	Communication information about the processor used by a job. The value can be escaped to the JSON format. The fields are described as follows: status: whether RankTable has been generated. initializing: Devices are being allocated to a job, and RankTable is not generated. complete: After RankTable is generated, the status changes to complete immediately, and fields such as server_list are displayed synchronously. server_list: device allocation information device: NPU allocation information, including the NPU IP address, and rank ID. server_id: AI server ID, which is globally unique. server_name: server name server_sn: server SN. Ensure that the server SN exists. If it does not exist, contact Huawei technical support. server_count: number of servers used by a job version: version information	String
job_id	Kubernetes ID of a job	String
operator	add: After a job adding command is received, the task status is changed to add. delete: After a task deletion command is received, the task status is changed to delete.	String
deleteTime	Time when a job is deleted.	String
sharedTorIp	Information about the shared switch used by a job.	String
masterAddr	MASTER_ADDR value specified during PyTorch training.	String
total	Number of ConfigMaps.	Integer
time	Time when a job starts.	String
framework	Framework used by a job.	String
job_status	Job status: Pending Running Complete Failed	String
job_name	Job name.	String
cm_index	Index of the current ConfigMap.	String

current-job-statistic

This file records statistics about the current job in the cluster and stores the information in /var/log/mindx-dl/clusterd/event_job.log. Due to the capacity limit of Kubernetes ConfigMap, a maximum of about 10,000 cluster jobs can be collected. When the size of a log file reaches 20 MB, the log file is automatically dumped. A maximum of five dump logs can be saved, and the dump logs can be retained for a maximum of 40 days.

Parameter	Description
data	-
- ID	Job ID allocated by the Kubernetes cluster.
- customID	User-defined job ID. If the content is empty, the job ID is not displayed.
- cardNum	Number of cards used by a job. If the content is empty, the card quantity is not displayed.
- podFirstRunTime	Time when all pods of a job are running for the first time. If the content is empty, the time is not displayed.
- stopTime	Time when all pods of a job are complete or are forcibly deleted. If the content is empty, the time is not displayed.
- podLastRunTime	Last time when all pods of a job were restored to running. If the content is empty, the time is not displayed.
- podLastFaultTime	Last time when some or all pods of a job failed. If the content is empty, the time is not displayed.
- podFaultTimes	Number of times that pods are rescheduled due to job faults. If the number is 0, the rescheduling times are not displayed.
totalJob	Total number of jobs in the current cluster.

Parent topic: ClusterD