BatchService

Account_ListPoolNodeCounts

Gets the number of nodes in each state, grouped by pool.

Adds a certificate to the specified account.

Adds a collection of tasks to the specified job.

Note that each task must have a unique ID. The Batch service may not return the results for each task in the same order the tasks were submitted in this request. If the server times out or the connection is closed during the request, the request may have been partially or fully processed, or not at all. In such cases, the user should re-issue the request. Note that it is up to the user to correctly handle failures when re-issuing a request. For example, you should use the same task IDs during a retry so that if the prior operation succeeded, the retry will not create extra tasks unexpectedly. If the response contains any tasks which failed to add, a client can retry the request. In a retry, it is most efficient to resubmit only tasks that failed to add, and to omit tasks that were successfully added on the first attempt. The maximum lifetime of a task from addition to completion is 7 days. If a task has not completed within 7 days of being added it will be terminated by the Batch service and left in whatever state it was in at that time.

Adds a job schedule to the specified account.

Adds a job to the specified account.

The Batch service supports two ways to control the work done as part of a job. In the first approach, the user specifies a Job Manager task. The Batch service launches this task when it is ready to start the job. The Job Manager task controls all other tasks that run under this job, by using the Task APIs. In the second approach, the user directly controls the execution of tasks under an active job, by using the Task APIs. Also note: when naming jobs, avoid including sensitive information such as user names or secret project names. This information may appear in telemetry logs accessible to Microsoft Support engineers.

Adds a pool to the specified account.

When naming pools, avoid including sensitive information such as user names or secret project names. This information may appear in telemetry logs accessible to Microsoft Support engineers.

Adds a task to the specified job.

The maximum lifetime of a task from addition to completion is 7 days. If a task has not completed within 7 days of being added it will be terminated by the Batch service and left in whatever state it was in at that time.

Adds a user account to the specified compute node.

You can add a user account to a node only when it is in the idle or running state.

Cancels a failed deletion of a certificate from the specified account.

If you try to delete a certificate that is being used by a pool or compute node, the status of the certificate changes to deleteFailed. If you decide that you want to continue using the certificate, you can use this operation to set the status of the certificate back to active. If you intend to delete the certificate, you do not need to run this operation after the deletion failed. You must make sure that the certificate is not being used by any resources, and then you can try again to delete the certificate.

Certificate_Get

Gets information about the specified certificate.

Changes the number of compute nodes that are assigned to a pool.

You can only resize a pool when its allocation state is steady. If the pool is already resizing, the request fails with status code 409. When you resize a pool, the pool's allocation state changes from steady to resizing. You cannot resize pools which are configured for automatic scaling. If you try to do this, the Batch service returns an error 409. If you resize a pool downwards, the Batch service chooses which nodes to remove. To remove specific nodes, use the pool remove nodes API instead.

Deletes a certificate from the specified account.

You cannot delete a certificate if a resource (pool or compute node) is using it. Before you can delete a certificate, you must therefore make sure that the certificate is not associated with any existing pools, the certificate is not installed on any compute nodes (even if you remove a certificate from a pool, it is not removed from existing compute nodes in that pool until they restart), and no running tasks depend on the certificate. If you try to delete a certificate that is in use, the deletion fails. The certificate status changes to deleteFailed. You can use Cancel Delete Certificate to set the status back to active if you decide that you want to continue using the certificate.

Deletes a job schedule from the specified account.

When you delete a job schedule, this also deletes all jobs and tasks under that schedule. When tasks are deleted, all the files in their working directories on the compute nodes are also deleted (the retention period is ignored). The job schedule statistics are no longer accessible once the job schedule is deleted, though they are still counted towards account lifetime statistics.

Deletes a job.

Deleting a job also deletes all tasks that are part of that job, and all job statistics. This also overrides the retention period for task data; that is, if the job contains tasks which are still retained on compute nodes, the Batch services deletes those tasks' working directories and all their contents. When a Delete Job request is received, the Batch service sets the job to the deleting state. All update operations on a job that is in deleting state will fail with status code 409 (Conflict), with additional information indicating that the job is being deleted.

Deletes a pool from the specified account.

When you request that a pool be deleted, the following actions occur: the pool state is set to deleting; any ongoing resize operation on the pool are stopped; the Batch service starts resizing the pool to zero nodes; any tasks running on existing nodes are terminated and requeued (as if a resize pool operation had been requested with the default requeue option); finally, the pool is removed from the system. Because running tasks are requeued, the user can rerun these tasks by updating their job to target a different pool. The tasks can then run on the new pool. If you want to override the requeue behavior, then you should call resize pool explicitly to shrink the pool to zero size before deleting the pool. If you call an Update, Patch or Delete API on a pool in the deleting state, it will fail with HTTP status code 409 with error code PoolBeingDeleted.

Deletes a task from the specified job.

When a task is deleted, all of the files in its directory on the compute node where it ran are also deleted (regardless of the retention time). For multi-instance tasks, the delete task operation applies synchronously to the primary task; subtasks and their files are then deleted asynchronously in the background.

Deletes a user account from the specified compute node.

You can delete a user account to a node only when it is in the idle or running state.

Deletes the specified file from the compute node.

Deletes the specified task file from the compute node where the task ran.

Disables a job schedule.

No new jobs will be created until the job schedule is enabled again.

Disables automatic scaling for a pool.

Disables task scheduling on the specified compute node.

You can disable task scheduling on a node only if its current scheduling state is enabled.

Disables the specified job, preventing new tasks from running.

The Batch Service immediately moves the job to the disabling state. Batch then uses the disableTasks parameter to determine what to do with the currently running tasks of the job. The job remains in the disabling state until the disable operation is completed and all tasks have been dealt with according to the disableTasks option; the job then moves to the disabled state. No new tasks are started under the job until it moves back to active state. If you try to disable a job that is in any state other than active, disabling, or disabled, the request fails with status code 409.

Enables a job schedule.

Enables automatic scaling for a pool.

You cannot enable automatic scaling on a pool if a resize operation is in progress on the pool. If automatic scaling of the pool is currently disabled, you must specify a valid autoscale formula as part of the request. If automatic scaling of the pool is already enabled, you may specify a new autoscale formula and/or a new evaluation interval. You cannot call this API for the same pool more than once every 30 seconds.

Enables task scheduling on the specified compute node.

You can enable task scheduling on a node only if its current scheduling state is disabled

Enables the specified job, allowing new tasks to run.

When you call this API, the Batch service sets a disabled job to the enabling state. After the this operation is completed, the job moves to the active state, and scheduling of new tasks under the job resumes. The Batch service does not allow a task to remain in the active state for more than 7 days. Therefore, if you enable a job containing active tasks which were added more than 7 days ago, those tasks will not run.

File_GetFromComputeNode

Returns the content of the specified compute node file.

File_GetFromTask

Returns the content of the specified task file.

Gets information about the specified application.

This operation returns only applications and versions that are available for use on compute nodes; that is, that can be used in an application package reference. For administrator information about applications and versions that are not yet available to compute nodes, use the Azure portal or the Azure Resource Manager API.

Gets information about the specified compute node.

Gets information about the specified job.

Gets information about the specified task.

For multi-instance tasks, information such as affinityId, executionInfo and nodeInfo refer to the primary task. Use the list subtasks API to retrieve information about subtasks.

Gets lifetime summary statistics for all of the jobs in the specified account.

Statistics are aggregated across all jobs that have ever existed in the account, from account creation to the last update time of the statistics. The statistics may not be immediately available. The Batch service performs periodic roll-up of statistics. The typical delay is about 30 minutes.

Gets lifetime summary statistics for all of the pools in the specified account.

Statistics are aggregated across all pools that have ever existed in the account, from account creation to the last update time of the statistics. The statistics may not be immediately available. The Batch service performs periodic roll-up of statistics. The typical delay is about 30 minutes.

Gets the Remote Desktop Protocol file for the specified compute node.

Before you can access a node by using the RDP file, you must create a user account on the node. This API can only be invoked on pools created with a cloud service configuration. For pools created with a virtual machine configuration, see the GetRemoteLoginSettings API.

Gets the result of evaluating an automatic scaling formula on the pool.

This API is primarily for validating an autoscale formula, as it simply returns the result without applying the formula to the pool. The pool must have auto scaling enabled in order to evaluate a formula.

Gets the settings required for remote login to a compute node.

Before you can remotely login to a node using the remote login settings, you must create a user account on the node. This API can be invoked only on pools created with the virtual machine configuration property. For pools created with a cloud service configuration, see the GetRemoteDesktop API.

Gets the task counts for the specified job.

Task counts provide a count of the tasks by active, running or completed task state, and a count of tasks which succeeded or failed. Tasks in the preparing state are counted as running.

JobSchedule_Get

Gets information about the specified job schedule.

Lists all of the applications available in the specified account.

This operation returns only applications and versions that are available for use on compute nodes; that is, that can be used in an application package reference. For administrator information about applications and versions that are not yet available to compute nodes, use the Azure portal or the Azure Resource Manager API.

Lists all of the certificates that have been added to the specified account.

Lists all of the files in task directories on the specified compute node.

Lists all of the job schedules in the specified account.

Lists all of the jobs in the specified account.

Lists all of the pools in the specified account.

Lists all of the subtasks that are associated with the specified multi-instance task.

If the task is not a multi-instance task then this returns an empty collection.

Lists all of the tasks that are associated with the specified job.

For multi-instance tasks, information such as affinityId, executionInfo and nodeInfo refer to the primary task. Use the list subtasks API to retrieve information about subtasks.

Lists the compute nodes in the specified pool.

Lists the execution status of the Job Preparation and Job Release task for the specified job across the compute nodes where the job has run.

This API returns the Job Preparation and Job Release task status on all compute nodes that have run the Job Preparation or Job Release task. This includes nodes which have since been removed from the pool. If this API is invoked on a job which has no Job Preparation or Job Release task, the Batch service returns HTTP status code 409 (Conflict) with an error code of JobPreparationTaskNotSpecified.

Lists the files in a task's directory on its compute node.

Lists the jobs that have been created under the specified job schedule.

Lists the usage metrics, aggregated by pool across individual time intervals, for the specified account.

If you do not specify a $filter clause including a poolId, the response includes all pools that existed in the account in the time range of the returned aggregation intervals. If you do not specify a $filter clause including a startTime or endTime these filters default to the start and end times of the last aggregation interval currently available; that is, only the last aggregation interval is returned.

Pool_Get

Gets information about the specified pool.

Reactivates a task, allowing it to run again even if its retry count has been exhausted.

Reactivation makes a task eligible to be retried again up to its maximum retry count. The task's state is changed to active. As the task is no longer in the completed state, any previous exit code or failure information is no longer available after reactivation. Each time a task is reactivated, its retry count is reset to 0. Reactivation will fail for tasks that are not completed or that previously completed successfully (with an exit code of 0). Additionally, it will fail if the job has completed (or is terminating or deleting).

Reinstalls the operating system on the specified compute node.

You can reinstall the operating system on a node only if it is in an idle or running state. This API can be invoked only on pools created with the cloud service configuration property.

Restarts the specified compute node.

You can restart a node only if it is in an idle or running state.

Stops an ongoing resize operation on the pool.

This does not restore the pool to its previous state before the resize operation: it only stops any further changes being made, and the pool maintains its current state. After stopping, the pool stabilizes at the number of nodes it was at when the stop operation was done. During the stop operation, the pool allocation state changes first to stopping and then to steady. A resize operation need not be an explicit resize pool request; this API can also be used to halt the initial sizing of the pool when it is created.

Terminates a job schedule.

Terminates the specified job, marking it as completed.

When a Terminate Job request is received, the Batch service sets the job to the terminating state. The Batch service then terminates any running tasks associated with the job and runs any required job release tasks. Then the job moves into the completed state. If there are any tasks in the job in the active state, they will remain in the active state. Once a job is terminated, new tasks cannot be added and any remaining active tasks will not be scheduled.

Terminates the specified task.

When the task has been terminated, it moves to the completed state. For multi-instance tasks, the terminate task operation applies synchronously to the primary task; subtasks are then terminated asynchronously in the background.

Updates the password and expiration time of a user account on the specified compute node.

This operation replaces of all the updatable properties of the account. For example, if the expiryTime element is not specified, the current value is replaced with the default value, not left unmodified. You can update a user account on a node only when it is in the idle or running state.

Updates the properties of the specified job schedule.

This replaces only the job schedule properties specified in the request. For example, if the schedule property is not specified with this request, then the Batch service will keep the existing schedule. Changes to a job schedule only impact jobs created by the schedule after the update has taken place; currently running jobs are unaffected.

Updates the properties of the specified job schedule.

This fully replaces all the updatable properties of the job schedule. For example, if the schedule property is not specified with this request, then the Batch service will remove the existing schedule. Changes to a job schedule only impact jobs created by the schedule after the update has taken place; currently running jobs are unaffected.

Updates the properties of the specified job.

This replaces only the job properties specified in the request. For example, if the job has constraints, and a request does not specify the constraints element, then the job keeps the existing constraints.

Updates the properties of the specified job.

This fully replaces all the updatable properties of the job. For example, if the job has constraints associated with it and if constraints is not specified with this request, then the Batch service will remove the existing constraints.

Updates the properties of the specified pool.

This only replaces the pool properties specified in the request. For example, if the pool has a start task associated with it, and a request does not specify a start task element, then the pool keeps the existing start task.

Updates the properties of the specified pool.

This fully replaces all the updatable properties of the pool. For example, if the pool has a start task associated with it and if start task is not specified with this request, then the Batch service will remove the existing start task.

Upgrades the operating system of the specified pool.

During an upgrade, the Batch service upgrades each compute node in the pool. When a compute node is chosen for upgrade, any tasks running on that node are removed from the node and returned to the queue to be rerun later (or on a different compute node). The node will be unavailable until the upgrade is complete. This operation results in temporarily reduced pool capacity as nodes are taken out of service to be upgraded. Although the Batch service tries to avoid upgrading all compute nodes at the same time, it does not guarantee to do this (particularly on small pools); therefore, the pool may be temporarily unavailable to run tasks. When this operation runs, the pool state changes to upgrading. When all compute nodes have finished upgrading, the pool state returns to active. While the upgrade is in progress, the pool's currentOSVersion reflects the OS version that nodes are upgrading from, and targetOSVersion reflects the OS version that nodes are upgrading to. Once the upgrade is complete, currentOSVersion is updated to reflect the OS version now running on all nodes. This operation can only be invoked on pools created with the cloudServiceConfiguration property.