CLI Reference¶

torc¶

torc commands

torc [OPTIONS] COMMAND [ARGS]...

Options

-c, --console-level <console_level>¶: Console log level.

-f, --file-level <file_level>¶: File log level. Set to ‘trace’ for increased verbosity.

-k, --workflow-key <workflow_key>¶: Workflow key, required for many commands. User will be prompted if it is missing unless –no-prompts is set.

-n, --no-prompts¶

Disable all user prompts.

Default:: False

-F, --output-format <output_format>¶

Output format for get/list commands. Not all commands support all formats.

Options:: text | csv | json

--timings <timings>¶

Enable tracking of function timings.

Options:: true | false

-U, --user <user>¶

Username

Default:: 'runner'

-u, --database-url <database_url>¶: Database URL. Ex: http://localhost:8529/_db/workflows/torc-service

--version¶: Show the version and exit.

Environment variables

TORC_WORKFLOW_KEY: Provide a default for -k

TORC_DATABASE_URL: Provide a default for -u

collections¶

Collections commands

torc collections [OPTIONS] COMMAND [ARGS]...

join¶

Perform a join of collections from a pre-defined configuration. Refer to show-join-configurations command for details.

Examples:

1. Show jobs and results in a table.

$ torc collections join job-results

2. Show jobs and results in JSON format.

$ torc -F json collections join job-results

3. Show results for one job. This uses the job key.

$ torc -F json collections join job-results -f return_code=0

torc collections join [OPTIONS] {compute-node-executed-jobs|compute-node-
                      utilization|job-blocks|job-needs-file|job-produces-
                      file|job-requirements|job-results|job-schedulers|job-
                      process-utilization|job-consumes-data|job-stores-data}

Options

-f, --filters <filters>¶: Filter the values according to each key=value pair on the primary collection.

-l, --limit <limit>¶: Limit the output to this number of jobs.

-s, --skip <skip>¶: Skip this number of jobs.

Arguments

NAME¶: Required argument

join-by-edge¶

Join a collection with one or more other collections connected by an edge.

Examples:

1. Show jobs and results in a table.

$ torc collections join-by-edge jobs returned

2. Show jobs and results in JSON format.

$ torc -F json collections join-by-edge jobs returned

torc collections join-by-edge [OPTIONS] COLLECTION EDGE

Options

-f, --filters <filters>¶: Filter the values according to each key=value pair on the primary collection.

--outbound, --inbound¶

Inbound or outbound edge.

Default:: True

-l, --limit <limit>¶: Limit the output to this number of jobs.

-s, --skip <skip>¶: Skip this number of jobs.

-x, --exclude-from <exclude_from>¶: Exclude this base column name on the from side. Accepts multiple

-y, --exclude-to <exclude_to>¶: Exclude this base column name on the to side. Accepts multiple

Arguments

COLLECTION¶: Required argument

EDGE¶: Required argument

list¶

List workflow collections.

torc collections list [OPTIONS]

Options

-r, --raw¶

List raw names

Default:: False

show-join-configurations¶

Show the pre-defined configurations for use in the join command.

torc collections show-join-configurations [OPTIONS]

compute-nodes¶

Compute node commands

torc compute-nodes [OPTIONS] COMMAND [ARGS]...

list¶

List all compute nodes that partipcated in a workflow.

torc compute-nodes list [OPTIONS]

list-resource-stats¶

Show resource statistics from a workflow run.

Examples:

1. List resource stats from all compute nodes in tables by resource type.

$ torc compute-nodes list-resource-stats

2. List resource stats from all compute nodes in JSON format -

one array keyed by ‘stats’.

$ torc -F json compute-nodes list-resource-stats

torc compute-nodes list-resource-stats [OPTIONS]

Options

-x, --exclude-process¶

Exclude job process stats (show compute node stats only).

Default:: False

config¶

Config commands

torc config [OPTIONS] COMMAND [ARGS]...

create¶

Create a local torc runtime configuration file.

torc config create [OPTIONS]

Options

-F, --output-format <output_format>¶

Output format for get/list commands. Not all commands support all formats.

Default:: 'text'
Options:: text | json

-d, --directory <directory>¶

Directory in which to store the config file.

Default:: PosixPath('/home/runner')

-f, --filter-workflows-by-user, --no-filter-workflows-by-user¶

Whether to filter workflows by the current user

Default:: True

-k, --workflow-key <workflow_key>¶: Workflow key. User will be prompted if it is missing unless –no-prompts is set.

--timings, --no-timings¶

Enable tracking of function timings.

Default:: False

-u, --database-url <database_url>¶: Database URL. Note the database name in this example: http://localhost:8529/_db/database_name/torc-service

--console-level <console_level>¶

Console log level.

Default:: 'info'

--file-level <file_level>¶

File log level. Set to ‘trace’ for increased verbosity.

Default:: 'debug'

events¶

Event commands

torc events [OPTIONS] COMMAND [ARGS]...

get-latest-event-timestamp¶

Return the timestamp of the latest event.

torc events get-latest-event-timestamp [OPTIONS]

list¶

List all events in a workflow.

Examples:

1. List all events.

$ torc events 91388876 list events

2. List only events with a category of job.

$ torc events 91388876 list events -f category=job

torc events list [OPTIONS]

Options

-A, --after-timestamp-ms <after_timestamp_ms>¶: Only return events that occurred after this timestamp expressed as millseconds since the epoch in UTC.

-a, --after-datetime <after_datetime>¶: Only return events that occurred after this local datetime (format = YYYY-MM-DD HH:MM:SS.ddd).

-f, --filters <filters>¶: Filter the values according to each key=value pair. Only ‘category’ is supported.

-l, --limit <limit>¶: Limit the output to this number of jobs.

-s, --skip <skip>¶: Skip this number of jobs.

monitor¶

Monitor events.

torc events monitor [OPTIONS]

Options

-c, --category <category>¶: Filter events by this category.

-d, --duration <duration>¶: Duration in seconds to monitor. Default is forever.

-p, --poll-interval <poll_interval>¶: Poll interval in seconds. Please be mindful of impacts to the database.

export¶

Export commands

torc export [OPTIONS] COMMAND [ARGS]...

sqlite¶

Export workflows stored in the database to this SQLite file. By default, export all workflows. Limit the output tables by passing specific workflow keys as positional arguments.

torc export sqlite [OPTIONS] [WORKFLOW_KEYS]...

Options

-F, --filename <filename>¶

SQLite filename

Default:: 'workflow.sqlite'

-f, --force¶

Overwrite file if it exists.

Default:: False

Arguments

WORKFLOW_KEYS¶: Optional argument(s)

files¶

File commands

torc files [OPTIONS] COMMAND [ARGS]...

add¶

Add a file to the workflow.

torc files add [OPTIONS]

Options

-n, --name <name>¶: file name

-p, --path <path>¶: Required Path of file

delete¶

Delete one or more files by key.

torc files delete [OPTIONS] [FILE_KEYS]...

Arguments

FILE_KEYS¶: Optional argument(s)

delete-all¶

Delete all files in the workflow.

torc files delete-all [OPTIONS]

list¶

List all files in a workflow.

Examples:

1. List all files in a table.

$ torc files list

2. List only files with name=file1

$ torc files list -f name=file1

3. List all files in JSON format.

$ torc -F json files list

torc files list [OPTIONS]

Options

-f, --filters <filters>¶: Filter the values according to each key=value pair.

--sort-by <sort_by>¶: Sort results by this column.

--reverse-sort¶

Reverse the sort order if –sort-by is set.

Default:: False

graphs¶

Graph commands

torc graphs [OPTIONS] COMMAND [ARGS]...

plot¶

Make a plot from an exported graph.

Example:

$ torc graphs plot job_job_dependencies

torc graphs plot [OPTIONS] [[job_job_dependencies|job_file_dependencies|job_us
                 er_data_dependencies]]...

Options

-k, --keep-dot-file¶

Keep the intermediate DOT file

Default:: False

-o, --output <output>¶

Output directory

Default:: 'output'

Arguments

NAMES¶: Optional argument(s)

plot-xgmml¶

Make a plot from an XGMML graph file exported with arangoexport.

Example:

$ torc graphs plot-xgmml export/job-blocks.xgmml

torc graphs plot-xgmml [OPTIONS] GRAPH_FILE

Options

-k, --keep-dot-file¶

Keep the intermediate DOT file

Default:: False

Arguments

GRAPH_FILE¶: Required argument

hpc¶

HPC commands

torc hpc [OPTIONS] COMMAND [ARGS]...

slurm¶

Slurm commands

torc hpc slurm [OPTIONS] COMMAND [ARGS]...

add-config¶

Add a Slurm config to the database.

torc hpc slurm add-config [OPTIONS]

Options

-N, --name <name>¶: Required Name of config

-a, --account <account>¶: Required HPC account

-g, --gres <gres>¶: Request nodes that have at least this number of GPUs. Ex: ‘gpu:2’

-m, --mem <mem>¶: Request nodes that have at least this amount of memory. Ex: ‘180G’

-n, --nodes <nodes>¶

Number of nodes to use for each job

Default:: 1

-p, --partition <partition>¶: HPC partition. Default is determined by the scheduler

-q, --qos <qos>¶

Controls priority of the jobs.

Default:: 'normal'

-t, --tmp <tmp>¶: Request nodes that have at least this amount of storage scratch space.

-w, --walltime <walltime>¶

Slurm job walltime.

Default:: '04:00:00'

-e, --extra <extra>¶: Add extra Slurm parameters, for example –extra=’–reservation=my-reservation’.

list-configs¶

Show the current Slurm configs in the database.

torc hpc slurm list-configs [OPTIONS]

modify-config¶

Modify a Slurm config in the database.

torc hpc slurm modify-config [OPTIONS] SLURM_CONFIG_KEY

Options

-N, --name <name>¶: Name of config

-a, --account <account>¶: HPC account

-g, --gres <gres>¶: Request nodes that have at least this number of GPUs. Ex: ‘gpu:2’

-m, --mem <mem>¶: Request nodes that have at least this amount of memory. Ex: ‘180G’

-n, --nodes <nodes>¶: Number of nodes to use for each job

-p, --partition <partition>¶: HPC partition. Default is determined by the scheduler

-q, --qos <qos>¶: Controls priority of the jobs.

-t, --tmp <tmp>¶: Request nodes that have at least this amount of storage scratch space.

-w, --walltime <walltime>¶: Slurm job walltime.

Arguments

SLURM_CONFIG_KEY¶: Required argument

run-jobs¶

Run workflow jobs on a Slurm compute node.

torc hpc slurm run-jobs [OPTIONS]

Options

-c, --cpu-affinity-cpus-per-job <cpu_affinity_cpus_per_job>¶: Enable CPU affinity for this number of CPUs per job.

-m, --max-parallel-jobs <max_parallel_jobs>¶: Maximum number of parallel jobs. Default is to use resource availability.

-o, --output <output>¶

Default:: 'output'

-p, --poll-interval <poll_interval>¶

Poll interval for job completions

Default:: 60

--is-subtask¶

Set to True if this is a subtask and multiple workers are running on one node.

Default:: False

-w, --wait-for-healthy-database-minutes <wait_for_healthy_database_minutes>¶

Wait this number of minutes if the database is offline.

Default:: 0

schedule-nodes¶

Schedule nodes with Slurm to run jobs.

torc hpc slurm schedule-nodes [OPTIONS]

Options

-c, --cpu-affinity-cpus-per-job <cpu_affinity_cpus_per_job>¶: Enable CPU affinity for this number of CPUs per job.

-j, --job-prefix <job_prefix>¶

Prefix for HPC job names

Default:: 'worker'

-k, --keep-submission-scripts¶

Keep Slurm submission scripts on the filesystem.

Default:: False

-m, --max-parallel-jobs <max_parallel_jobs>¶: Maximum number of parallel jobs. Default is to use resource availability.

-n, --num-hpc-jobs <num_hpc_jobs>¶: Required Number of HPC jobs to schedule

-o, --output <output>¶

Output directory for compute nodes

Default:: 'output'

-p, --poll-interval <poll_interval>¶

Poll interval for job completions

Default:: 60

-s, --scheduler-config-key <scheduler_config_key>¶: SlurmScheduler config key. Auto-selected if possible.

-S, --start-one-worker-per-node¶: Start a torc worker on each compute node. The default behavior starts a worker on the first compute node but no others. That defers control of the nodes to the user job. Setting this flag means that every compute node in the allocation will run jobs concurrently. This flag has no effect if each Slurm allocation has one compute node (default).

jobs¶

Job commands

torc jobs [OPTIONS] COMMAND [ARGS]...

add¶

Add a job to the workflow.

torc jobs add [OPTIONS]

Options

--cancel-on-blocking-job-failure, --no-cancel-on-blocking-job-failure¶

Cancel the job if a blocking job fails.

Default:: True

-c, --command <command>¶: Required Command to run

-k, --key <key>¶: Job key. Default is to auto-generate

-n, --name <name>¶: Job name

assign-resource-requirements¶

Assign resource requirements to one or more jobs.

torc jobs assign-resource-requirements [OPTIONS] RESOURCE_REQUIREMENTS_KEY
                                       [JOB_KEYS]...

Arguments

RESOURCE_REQUIREMENTS_KEY¶: Required argument

JOB_KEYS¶: Optional argument(s)

delete¶

Delete one or more jobs by key.

torc jobs delete [OPTIONS] [JOB_KEYS]...

Arguments

JOB_KEYS¶: Optional argument(s)

delete-all¶

Delete all jobs in the workflow.

torc jobs delete-all [OPTIONS]

disable¶

Set the status of one or more jobs to disabled.

torc jobs disable [OPTIONS] [JOB_KEYS]...

Arguments

JOB_KEYS¶: Optional argument(s)

list¶

List all jobs in a workflow.

Examples:

1. List all jobs in a table.

$ torc jobs list jobs

2. List only jobs with run_id=1 and status=done.

$ torc jobs list jobs -f run_id=1 -f status=done

3. List all jobs in JSON format.

$ torc -F json jobs list

torc jobs list [OPTIONS]

Options

-f, --filters <filters>¶: Filter the values according to each key=value pair.

-x, --exclude <exclude>¶: Exclude this column name. Accepts multiple

-l, --limit <limit>¶: Limit the output to this number of jobs.

-s, --skip <skip>¶: Skip this number of jobs.

--sort-by <sort_by>¶: Sort results by this column.

--reverse-sort¶

Reverse the sort order if –sort-by is set.

Default:: False

list-process-stats¶

List per-job process resource statistics from a workflow run.

Examples:

1. List stats for all jobs in a table.

$ torc jobs list-process-stats

2. List all stats in JSON format.

$ torc -F json jobs list-process-stats

torc jobs list-process-stats [OPTIONS]

Options

-l, --limit <limit>¶: Limit the output to this number of jobs.

-s, --skip <skip>¶: Skip this number of jobs.

list-user-data¶

List all user data stored or consumed for a job.

torc jobs list-user-data [OPTIONS] JOB_KEY

Options

--stores, --consumes¶

List data that is either stored by the job or consumed by the job.

Default:: True

Arguments

JOB_KEY¶: Required argument

reset-status¶

Reset the status of one or more jobs.

torc jobs reset-status [OPTIONS] [JOB_KEYS]...

Arguments

JOB_KEYS¶: Optional argument(s)

run¶

Run workflow jobs on the current system.

torc jobs run [OPTIONS]

Options

-c, --cpu-affinity-cpus-per-job <cpu_affinity_cpus_per_job>¶: Enable CPU affinity for this number of CPUs per job.

-m, --max-parallel-jobs <max_parallel_jobs>¶: Maximum number of parallel jobs. Default is to use resource availability.

-o, --output <output>¶

Default:: 'output'

-p, --poll-interval <poll_interval>¶

Poll interval for job completions

Default:: 10

-s, --scheduler-config-id <scheduler_config_id>¶: Only run jobs with this scheduler config id.

-t, --time-limit <time_limit>¶: Time limit ISO 8601 time duration format (like ‘P0DT24H’), defaults to no limit.

-w, --wait-for-healthy-database-minutes <wait_for_healthy_database_minutes>¶

Wait this number of minutes if the database is offline. Applies only to the initial connection.

Default:: 0

run-function¶

Run a function on one set of inputs stored in the workflow database. Only called by the torc worker application as part of the mapped-function workflow.

torc jobs run-function [OPTIONS]

run-postprocess¶

Run a postprocess function on the results of a mapped-function workflow.

torc jobs run-postprocess [OPTIONS]

reports¶

Report commands

torc reports [OPTIONS] COMMAND [ARGS]...

results¶

Report information about job results and log files.

torc reports results [OPTIONS] [JOB_KEYS]...

Options

-o, --output <output>¶

Default:: 'output'

-r, --run-id <run_id>¶: Enter one or more run IDs to limit output to specific runs. Default is to show all.

Arguments

JOB_KEYS¶: Optional argument(s)

resource-requirements¶

Job resource requirements commands

torc resource-requirements [OPTIONS] COMMAND [ARGS]...

add¶

Add a resource requirements definition to the workflow.

torc resource-requirements add [OPTIONS]

Options

-n, --name <name>¶: Resource requirements name

-c, --num-cpus <num_cpus>¶

Number of CPUs required by a job

Default:: 1

-m, --memory <memory>¶

Amount of memory required by a job, such as ‘20g’

Default:: '1m'

-r, --runtime <runtime>¶

ISO 8601 encoding for job runtime

Default:: 'P0DT1M'

-N, --num-nodes <num_nodes>¶

Number of compute nodes required by a job

Default:: 1

-a, --apply-to-all-jobs¶

Apply these requirements to all jobs in the workflow.

Default:: False

delete¶

Delete one or more resource requirements by key.

torc resource-requirements delete [OPTIONS] [RESOURCE_REQUIREMENT_KEYS]...

Arguments

RESOURCE_REQUIREMENT_KEYS¶: Optional argument(s)

delete-all¶

Delete all resource_requirements in the workflow.

torc resource-requirements delete-all [OPTIONS]

list¶

List all resource_requirements in a workflow.

Examples:

1. List all resource_requirements in a table.

$ torc resource_requirements list resource_requirements

2. List only resource_requirements with num_cpus=4.

$ torc resource_requirements list resource_requirements -f num_cpus=4

3. List all resource_requirements in JSON format.

$ torc -F json resource_requirements list

torc resource-requirements list [OPTIONS]

Options

-f, --filters <filters>¶: Filter the values according to each key=value pair.

-l, --limit <limit>¶: Limit the output to this number of resource_requirements.

-s, --skip <skip>¶: Skip this number of resource_requirements.

--sort-by <sort_by>¶: Sort results by this column.

--reverse-sort¶

Reverse the sort order if –sort-by is set.

Default:: False

modify¶

Modify a resource requirements definition.

torc resource-requirements modify [OPTIONS] RESOURCE_REQUIREMENTS_KEY

Options

-n, --name <name>¶: Resource requirements name

-c, --num-cpus <num_cpus>¶: Number of CPUs required by a job

-m, --memory <memory>¶: Amount of memory required by a job, such as ‘20g’

-r, --runtime <runtime>¶: ISO 8601 encoding for job runtime

-N, --num-nodes <num_nodes>¶: Number of compute nodes required by a job

Arguments

RESOURCE_REQUIREMENTS_KEY¶: Required argument

results¶

Result commands

torc results [OPTIONS] COMMAND [ARGS]...

delete¶

Delete all results for one or more workflows.

torc results delete [OPTIONS]

list¶

List all results in a workflow.

Examples:

1. List all results in a table.

$ torc results list

2. List only results with a return_code of 1.

$ torc results list -f return_code=1

3. List the latest result for each job.

$ torc results list –latest-only

4. List all results in JSON format.

$ torc -F json results 91388876 list results

torc results list [OPTIONS]

Options

-f, --filters <filters>¶: Filter the values according to each key=value pair.

-L, --latest-only¶

Limit output to the latest result for each job.

Default:: False

-l, --limit <limit>¶: Limit the output to this number of jobs.

-s, --skip <skip>¶: Skip this number of jobs.

-x, --exclude-job-names¶

Exclude job names from the output. Set this flag if you need to deserialize the objects into Result classes or to speed up the query.

Default:: False

--sort-by <sort_by>¶: Sort results by this column.

--reverse-sort¶

Reverse the sort order if –sort-by is set.

Default:: False

stats¶

Stats commands

torc stats [OPTIONS] COMMAND [ARGS]...

concatenate-process¶

Concatenate job process stats from all compute nodes into one file. output_dir must be the directory that contains the existing .sqlite files.

torc stats concatenate-process [OPTIONS] OUTPUT_DIR

Arguments

OUTPUT_DIR¶: Required argument

tui¶

Starts a terminal-based management console.

torc tui [OPTIONS]

user-data¶

User data commands

torc user-data [OPTIONS] COMMAND [ARGS]...

add¶

Add user data to the workflow. Could be a placeholder or could contain data.

Example:

1. Add a placeholder for data that will be stored by one job and consumed by another.

$ torc user-data add -n output_data_1 -s 96117190 -c 96117191 -c 96117192

2. Add a placeholder for data that will be stored by one job and consumed by another.

$ torc user-data add -n output_data_1 -d “{key1: ‘val1’, key2: ‘val2’}”

torc user-data add [OPTIONS]

Options

-d, --data <data>¶: Object encoded in a JSON5 string.

--ephemeral, --not-ephemeral¶

Whether the data is ephemeral and should be cleared on every run of the workflow.

Default:: False

-n, --name <name>¶: User data name

-s, --stores <stores>¶: Key of job that will store the data.

-c, --consumes <consumes>¶: Key of job or jobs that will consume the data. Accepts multiple.

delete¶

Delete one or more user_data objects by key.

torc user-data delete [OPTIONS] [USER_DATA_KEYS]...

Arguments

USER_DATA_KEYS¶: Optional argument(s)

delete-all¶

Delete all user_data objects in the workflow.

torc user-data delete-all [OPTIONS]

get¶

Get one user_data object by key.

torc user-data get [OPTIONS] KEY

Arguments

KEY¶: Required argument

list¶

List all user data in a workflow.

torc user-data list [OPTIONS]

Options

-f, --filters <filters>¶: Filter the values according to each key=value pair.

-l, --limit <limit>¶: Limit the output to this number of items.

-s, --skip <skip>¶: Skip this number of items.

modify¶

Modify user data.

torc user-data modify [OPTIONS] USER_DATA_KEY

Options

-n, --name <name>¶: User data name

-d, --data <data>¶: Object encoded in a JSON5 string

--ephemeral, --not-ephemeral¶: Whether the data is ephemeral and should be cleared on every run of the workflow.

Arguments

USER_DATA_KEY¶: Required argument

workflows¶

Workflow commands

torc workflows [OPTIONS] COMMAND [ARGS]...

add-jobs-from-commands-file¶

Add jobs to a workflow from a text file containing job CLI commands.

torc workflows add-jobs-from-commands-file [OPTIONS] FILENAME

Options

-c, --cpus-per-job <cpus_per_job>¶

Number of CPUs required for each job.

Default:: 1

-m, --memory-per-job <memory_per_job>¶

Amount of memory required for each job. Use ‘100m’ for 100 MB, ‘1g’ for 1 GB, etc.

Default:: '1m'

-r, --runtime-per-job <runtime_per_job>¶

Runtime required for each job in ISO8601 format. Example: P0DT1H is one hour.

Default:: 'P0DT1m'

Arguments

FILENAME¶: Required argument

cancel¶

Cancel one or more workflows.

torc workflows cancel [OPTIONS] [WORKFLOW_KEYS]...

Arguments

WORKFLOW_KEYS¶: Optional argument(s)

create¶

Create a new workflow.

torc workflows create [OPTIONS]

Options

-d, --description <description>¶: Workflow description

-k, --key <key>¶: Workflow key. Default is to auto-generate

-n, --name <name>¶: Workflow name

create-from-commands-file¶

Create a workflow from a text file containing job CLI commands.

torc workflows create-from-commands-file [OPTIONS] FILENAME

Options

-c, --cpus-per-job <cpus_per_job>¶

Number of CPUs required for each job.

Default:: 1

-d, --description <description>¶: Workflow description

-k, --key <key>¶: Workflow key. Default is to auto-generate

-m, --memory-per-job <memory_per_job>¶

Amount of memory required for each job. Use ‘100m’ for 100 MB, ‘1g’ for 1 GB, etc.

Default:: '1m'

-n, --name <name>¶: Workflow name

-r, --runtime-per-job <runtime_per_job>¶

Runtime required for each job in ISO8601 format. Example: P0DT1H is one hour.

Default:: 'P0DT1m'

Arguments

FILENAME¶: Required argument

create-from-json-file¶

Create a workflow from a JSON/JSON5 file.

torc workflows create-from-json-file [OPTIONS] FILENAME

Arguments

FILENAME¶: Required argument

delete¶

Delete one or more workflows by key.

torc workflows delete [OPTIONS] [WORKFLOW_KEYS]...

Arguments

WORKFLOW_KEYS¶: Optional argument(s)

delete-all¶

Delete all workflows stored by the user.

torc workflows delete-all [OPTIONS]

example¶

Show the example workflow.

torc workflows example [OPTIONS]

list¶

List all workflows stored by the user.

1. List all workflows for the current user in a table.

$ torc workflows list

2. List all workflows in JSON format.

$ torc -o json workflows list

3. List only archived workflows.

$ torc workflows list –only-archived

4. List all workflows for all users, including archived workflows.

$ torc workflows list –all-users –include-archived

torc workflows list [OPTIONS]

Options

-A, --only-archived¶

List only workflows that have been archived.

Default:: False

-i, --include-archived¶

Include archived workflows in the list.

Default:: False

-a, --all-users¶: List workflows for all users. Default is only for the current user.

-f, --filters <filters>¶: Filter the values according to each key=value pair.

--sort-by <sort_by>¶: Sort results by this column.

--reverse-sort¶

Reverse the sort order if –sort-by is set.

Default:: False

list-scheduler-configs¶

List the scheduler configs in the database.

torc workflows list-scheduler-configs [OPTIONS]

modify¶

Modify the workflow parameters.

torc workflows modify [OPTIONS] WORKFLOW_KEY

Options

-a, --archive <archive>¶: Set to ‘true’ to archive the workflow or ‘false’ to enable it.

-d, --description <description>¶: Workflow description

-n, --name <name>¶: Workflow name

Arguments

WORKFLOW_KEY¶: Required argument

process-auto-tune-resource-requirements-results¶

Process the results of the first round of auto-tuning resource requirements.

torc workflows process-auto-tune-resource-requirements-results 
    [OPTIONS]

reset-status¶

Reset the status of the workflow(s) and all jobs.

torc workflows reset-status [OPTIONS] [WORKFLOW_KEYS]...

Options

-f, --failed-only¶

Only reset the status of failed jobs.

Default:: False

-r, --restart¶

Send the ‘workflows restart’ command after resetting status.

Default:: False

Arguments

WORKFLOW_KEYS¶: Optional argument(s)

restart¶

Restart the workflow defined in the database specified by the URL. Resets all jobs with a status of canceled, submitted, submitted_pending, and terminated. Does not affect jobs with a status of done unless an input file has changed.

torc workflows restart [OPTIONS]

Options

-d, --dry-run¶

Perform a dry run. Show status changes but do not change any database values.

Default:: False

-i, --ignore-missing-data¶

Ignore checks for missing files and user data documents.

Default:: False

--only-uninitialized¶

Only initialize jobs with a status of uninitialized.

Default:: False

set-compute-node-parameters¶

Set parameters that control how the torc worker app behaves on compute nodes. Run ‘torc workflows show-config’ to see the current values.

torc workflows set-compute-node-parameters [OPTIONS]

Options

-e, --expiration-buffer <expiration_buffer>¶: Set the number of seconds before the expiration time at which torc will terminate jobs.

-h, --wait-for-healthy-db <wait_for_healthy_db>¶: Set the number of minutes that torc will tolerate an offline database.

-i, --ignore-workflow-completion <ignore_workflow_completion>¶: Set to ‘true’ to cause torc to ignore workflow completions and hold onto compute node allocations indefinitely. Useful for debugging failed jobs. Set to ‘false’ to revert to the default behavior.

-w, --wait-for-new-jobs <wait_for_new_jobs>¶: Set the number of seconds that torc will wait for new jobs before exiting. Does not apply if the workflow is complete.

show¶

Show the workflow.

torc workflows show [OPTIONS]

Options

--sanitize, --no-santize¶

Remove all database fields from workflow objects.

Default:: True

show-config¶

Show the workflow config.

torc workflows show-config [OPTIONS]

show-status¶

Show the workflow status.

torc workflows show-status [OPTIONS]

start¶

Start the workflow defined in the database specified by the URL.

torc workflows start [OPTIONS]

Options

-a, --auto-tune-resource-requirements¶

Setup the workflow such that only one job from each resource group is run in the first round. Upon completion torc will look at actual resource utilization of those jobs and apply the results to the resource requirements definitions. When jobs finish, please call ‘torc workflows process_auto_tune_resource_requirements_results’ to update the requirements.

Default:: False

-i, --ignore-missing-data¶

Ignore checks for missing files and user data documents.

Default:: False

template¶

Show the workflow template.

torc workflows template [OPTIONS]

CLI Reference¶

torc¶

collections¶

join¶

join-by-edge¶

list¶

show-join-configurations¶

compute-nodes¶

list¶

list-resource-stats¶

config¶

create¶

events¶

get-latest-event-timestamp¶

list¶

monitor¶

export¶

sqlite¶

files¶

add¶

delete¶

delete-all¶

list¶

graphs¶

plot¶

plot-xgmml¶

hpc¶

slurm¶

add-config¶

list-configs¶

modify-config¶

recommend-nodes¶

run-jobs¶

schedule-nodes¶

jobs¶

add¶

assign-resource-requirements¶

delete¶

delete-all¶

disable¶

list¶

list-process-stats¶

list-user-data¶

reset-status¶

run¶

run-function¶

run-postprocess¶

reports¶

results¶

resource-requirements¶

add¶

delete¶

delete-all¶

list¶

modify¶

results¶

delete¶

list¶

stats¶

concatenate-process¶

tui¶

user-data¶

add¶

delete¶

delete-all¶

get¶

list¶

modify¶

workflows¶

add-jobs-from-commands-file¶

cancel¶

create¶

create-from-commands-file¶

create-from-json-file¶

delete¶

delete-all¶

example¶

list¶

list-scheduler-configs¶

modify¶