Configuration Reference

ReFrame’s behavior can be configured through its configuration file (see Configuring ReFrame for Your Site), environment variables and command-line options. An option can be specified via multiple paths (e.g., a configuration file parameter and an environment variable), in which case command-line options precede environment variables, which in turn precede configuration file options. This section provides a complete reference guide of the configuration options of ReFrame that can be set in its configuration file or specified using environment variables.

ReFrame’s configuration is in JSON syntax. The full schema describing it can be found in reframe/schemas/config.json file. The final configuration for ReFrame is validated against this schema.

The syntax we use to describe the different configuration objects follows the convention: OBJECT[.OBJECT]*.PROPERTY. Even if a configuration object contains a list of other objects, this is not reflected in the above syntax, as all objects in a certain list are homogeneous. For example, by systems.partitions.name we designate the name property of any partition object inside the partitions property of any system object inside the top level systems object. If we were to use indices, that would be rewritten as systems[i].partitions[j].name where i indexes the systems and j indexes the partitions of the i-th system. For cases, where the objects in a list are not homogeneous, e.g., the logging handlers, we surround the object type with ... For example, the logging.handlers_perflog..filelog..name syntax designates the name attribute of the filelog logging handler.

Top-level Configuration

The top-level configuration object is essentially the full configuration of ReFrame. It consists of the following properties, which we also call conventionally configuration sections:

systems
Required:

Yes

A list of system configuration objects.

environments
Required:

Yes

A list of environment configuration objects.

logging
Required:

Yes

A list of logging configuration objects.

modes
Required:

No

A list of execution mode configuration objects.

general
Required:

No

A list of general configuration objects.

autodetect_methods
Required:

No

Default:

["py::socket.gethostname"]

A list of system auto-detection methods for identifying the current system.

The list can contain two types of methods:

  1. Python methods: These are prefixed with py:: and should point to a Python callable taking zero arguments and returning a string. If the specified Python callable is not prefixed with a module, it will be looked up in the loaded configuration files starting from the last file. If the requested symbol cannot be found, a warning will be issued and the method will be ignored.

  2. Shell commands: Any string not prefixed with py:: will be treated as a shell command and will be executed during auto-detection to retrieve the hostname. The standard output of the command will be used.

If the --system option is not passed, ReFrame will try to autodetect the current system trying the methods in this list successively, until one of them succeeds. The resulting name will be matched against the hostnames patterns of each system and the system that matches first will be used as the current one.

The auto-detection methods can also be controlled through the RFM_AUTODETECT_METHODS environment variable.

New in version 4.3.

Warning

Changed in version 4.0.0: The schedulers section is removed. Scheduler options should be set per partition using the sched_options attribute.

System Configuration

systems.name
Required:

Yes

The name of this system. Only alphanumeric characters, dashes (-) and underscores (_) are allowed.

systems.descr
Required:

No

Default:

""

The description of this system.

systems.hostnames
Required:

Yes

A list of hostname regular expression patterns in Python syntax, which will be used by the framework in order to automatically select a system configuration. For the auto-selection process, see here.

systems.max_local_jobs

The maximum number of forced local build or run jobs allowed.

Forced local jobs run within the execution context of ReFrame.

Required:

No

Default:

8

New in version 3.10.0.

systems.modules_system
required:

No

default:

"nomod"

The modules system that should be used for loading environment modules on this system. Available values are the following:

  • tmod: The classic Tcl implementation of the environment modules (version 3.2).

  • tmod31: The classic Tcl implementation of the environment modules (version 3.1). A separate backend is required for Tmod 3.1, because Python bindings are different from Tmod 3.2.

  • tmod32: A synonym of tmod.

  • tmod4: The new environment modules implementation (versions older than 4.1 are not supported).

  • lmod: The Lua implementation of the environment modules.

  • spack: Spack’s built-in mechanism for managing modules.

  • nomod: This is to denote that no modules system is used by this system.

Normally, upon loading the configuration of the system ReFrame checks that a sane installation exists for the modules system requested and will issue an error if it fails to find one. The modules system sanity check is skipped when the resolve_module_conflicts is set to False. This is useful in cases where the current system does not have a modules system but the remote partitions have one and you would like ReFrame to generate the module commands.

New in version 3.4: The spack backend is added.

Changed in version 4.5.0: The modules system sanity check is skipped when the config.general.resolve_module_conflicts is not set.

systems.modules
Required:

No

Default:

[]

A list of environment module objects to be loaded always when running on this system. These modules modify the ReFrame environment. This is useful in cases where a particular module is needed, for example, to submit jobs on a specific system.

systems.env_vars
Required:

No

Default:

[]

A list of environment variables to be set always when running on this system. These variables modify the ReFrame environment. Each environment variable is specified as a two-element list containing the variable name and its value. You may reference other environment variables when defining an environment variable here. ReFrame will expand its value. Variables are set after the environment modules are loaded.

New in version 4.0.0.

systems.variables

Deprecated since version 4.0.0: Please use env_vars instead. If specified in conjunction with env_vars, it will be ignored.

systems.prefix
Required:

No

Default:

"."

Directory prefix for a ReFrame run on this system. Any directories or files produced by ReFrame will use this prefix, if not specified otherwise.

systems.stagedir
Required:

No

Default:

"${RFM_PREFIX}/stage"

Stage directory prefix for this system. This is the directory prefix, where ReFrame will create the stage directories for each individual test case.

systems.outputdir
Required:

No

Default:

"${RFM_PREFIX}/output"

Output directory prefix for this system. This is the directory prefix, where ReFrame will save information about the successful tests.

systems.resourcesdir
Required:

No

Default:

"."

Directory prefix where external test resources (e.g., large input files) are stored. You may reference this prefix from within a regression test by accessing the resourcesdir attribute of the current system.

systems.partitions
Required:

Yes

A list of system partition configuration objects. This list must have at least one element.

systems.sched_options
Required:

No

Default:

{}

Scheduler options for the local scheduler that is associated with the ReFrame’s execution context. To understand the difference between the different execution contexts, please refer to “Where each pipeline stage is executed?” For the available scheduler options, see the sched_options in the partition configuration below.

New in version 4.1.

Warning

This option is broken in 4.0.

System Partition Configuration

systems.partitions.name
Required:

Yes

The name of this partition. Only alphanumeric characters, dashes (-) and underscores (_) are allowed.

systems.partitions.descr
Required:

No

Default:

""

The description of this partition.

systems.partitions.scheduler
Required:

Yes

The job scheduler that will be used to launch jobs on this partition. Supported schedulers are the following:

  • flux: Jobs will be launched using the Flux Framework scheduler.

  • local: Jobs will be launched locally without using any job scheduler.

  • lsf: Jobs will be launched using the LSF scheduler.

  • oar: Jobs will be launched using the OAR scheduler.

  • pbs: Jobs will be launched using the PBS Pro scheduler.

  • sge: Jobs will be launched using the Sun Grid Engine scheduler.

  • slurm: Jobs will be launched using the Slurm scheduler. This backend requires job accounting to be enabled in the target system. If not, you should consider using the squeue backend below.

  • squeue: Jobs will be launched using the Slurm scheduler. This backend does not rely on job accounting to retrieve job statuses, but ReFrame does its best to query the job state as reliably as possible.

  • ssh: Jobs will be launched on a remote host using SSH.

    The remote host will be selected from the list of hosts specified in ssh_hosts. The scheduler keeps track of the hosts that it has submitted jobs to, and it will select the next available one in a round-robin fashion. For connecting to a remote host, the options specified in access will be used.

    When a job is submitted with this scheduler, its stage directory will be copied over to a unique temporary directory on the remote host, then the job will be executed and, finally, any produced artifacts will be copied back.

    The contents of the stage directory are copied to the remote host either using rsync, if available, or scp as a second choice. The same access options will be used in those operations as well. Please note, that the connection options of ssh and scp differ and ReFrame will not attempt to translate any options between the two utilities in case scp is selected for copying to the remote host. In this case, it is preferable to set up the host connection options in ~/.ssh/config and leave access blank.

    Job-scheduler command line options can be used to interact with the ssh backend. More specifically, if the --distribute option is used, a test will be generated for each host listed in ssh_hosts. You can also pin a test to a specific host if you pass the #host directive to the -J option, e.g., -J '#host=myhost'.

  • torque: Jobs will be launched using the Torque scheduler.

New in version 3.7.2: Support for the SGE scheduler is added.

New in version 3.8.2: Support for the OAR scheduler is added.

New in version 3.11.0: Support for the LSF scheduler is added.

New in version 4.4: The ssh scheduler is added.

Note

The way that multiple node jobs are submitted using the SGE scheduler can be very site-specific. For this reason, the sge scheduler backend does not try to interpret any related arguments, e.g., num_tasks, num_tasks_per_node etc. Users must specify how these resources are to be requested by setting the resources partition configuration parameter and then request them from inside a test using the extra_resources test attribute. Here is an example configuration for a system partition named foo that defines different ways for submitting MPI-only, OpenMP-only and MPI+OpenMP jobs:

{
    'name': 'foo',
    'scheduler': 'sge',
    'resources': [
        {
            'name': 'smp',
            'options': ['-pe smp {num_slots}']
        },
        {
            'name': 'mpi',
            'options': ['-pe mpi {num_slots}']
        },
        {
            'name': 'mpismp',
            'options': ['-pe mpismp {num_slots}']
        }
    ]
}

Each test then can request the different type of slots as follows:

self.extra_resouces = {
    'smp': {'num_slots': self.num_cpus_per_task},
    'mpi': {'num_slots': self.num_tasks},
    'mpismp': {'num_slots': self.num_tasks*self.num_cpus_per_task}
}

Notice that defining extra_resources allows the test to be portable to other systems that have different schedulers; the extra_resources will be simply ignored in this case and the scheduler backend will interpret the different test fields in the appropriate way.

systems.partitions.sched_options
Required:

No

Default:

{}

Scheduler-specific options for this partition. See below for the available options.

New in version 4.1.

Warning

This option is broken in 4.0.

systems.partitions.sched_options.ssh_hosts
Required:

No

Default:

[]

List of hosts in a partition that uses the ssh scheduler.

systems.partitions.sched_options.ignore_reqnodenotavail
Required:

No

Default:

false

Ignore the ReqNodeNotAvail Slurm state.

If a job associated to a test is in pending state with the Slurm reason ReqNodeNotAvail and a list of unavailable nodes is also specified, ReFrame will check the status of the nodes and, if all of them are indeed down, it will cancel the job. Sometimes, however, when Slurm’s backfill algorithm takes too long to compute, Slurm will set the pending reason to ReqNodeNotAvail and mark all system nodes as unavailable, causing ReFrame to kill the job. In such cases, you may set this parameter to true to avoid this.

This option is relevant for the Slurm backends only.

systems.partitions.sched_options.job_submit_timeout
Required:

No

Default:

60

Timeout in seconds for the job submission command.

If timeout is reached, the test issuing that command will be marked as a failure.

systems.partitions.sched_options.resubmit_on_errors
Required:

No

Default:

[]

If any of the listed errors occur, try to resubmit the job after some seconds.

As an example, you could have ReFrame trying to resubmit a job in case that the maximum submission limit per user is reached by setting this field to ["QOSMaxSubmitJobPerUserLimit"]. You can ignore multiple errors at the same time if you add more error strings in the list.

This option is relevant for the Slurm backends only.

New in version 3.4.1.

Warning

Job submission is a synchronous operation in ReFrame. If this option is set, ReFrame’s execution will block until the error conditions specified in this list are resolved. No other test would be able to proceed.

systems.partitions.sched_options.use_nodes_option
Required:

No

Default:

false

Always emit the --nodes Slurm option in the preamble of the job script.

This option is relevant for the Slurm backends only.

systems.partitions.launcher
Required:

Yes

The parallel job launcher that will be used in this partition to launch parallel programs. Available values are the following:

  • alps: Parallel programs will be launched using the Cray ALPS aprun command.

  • clush: Parallel programs will be launched using the ClusterShell clush command. This launcher uses the partition’s access property in order to determine the options to be passed to clush.

  • ibrun: Parallel programs will be launched using the ibrun command. This is a custom parallel program launcher used at TACC.

  • local: No parallel program launcher will be used. The program will be launched locally.

  • lrun: Parallel programs will be launched using LC Launcher’s lrun command.

  • lrun-gpu: Parallel programs will be launched using LC Launcher’s lrun -M "-gpu" command that enables the CUDA-aware Spectrum MPI.

  • mpirun: Parallel programs will be launched using the mpirun command.

  • mpiexec: Parallel programs will be launched using the mpiexec command.

  • pdsh: Parallel programs will be launched using the pdsh command. This launcher uses the partition’s access property in order to determine the options to be passed to pdsh.

  • srun: Parallel programs will be launched using Slurm’s srun command.

  • srunalloc: Parallel programs will be launched using Slurm’s srun command, but job allocation options will also be emitted. This can be useful when combined with the local job scheduler.

  • ssh: Parallel programs will be launched using SSH. This launcher uses the partition’s access property in order to determine the remote host and any additional options to be passed to the SSH client. The ssh command will be launched in “batch mode,” meaning that password-less access to the remote host must be configured. Here is an example configuration for the ssh launcher:

    {
        'name': 'foo'
        'scheduler': 'local',
        'launcher': 'ssh'
        'access': ['-l admin', 'remote.host'],
        'environs': ['builtin'],
    }
    
  • upcrun: Parallel programs will be launched using the UPC upcrun command.

  • upcxx-run: Parallel programs will be launched using the UPC++ upcxx-run command.

Tip

New in version 4.0.0: ReFrame also allows you to register your own custom launchers simply by defining them in the configuration. You can follow a small tutorial here.

systems.partitions.access
Required:

No

Default:

[]

A list of job scheduler options that will be passed to the generated job script for gaining access to that logical partition.

Note

For the pbs and torque backends, options accepted in the access and resources parameters may either refer to actual qsub options or may just be resources specifications to be passed to the -l option. The backend assumes a qsub option, if the options passed in these attributes start with a -.

Note

If constraints are specified in access for the Slurm backends, these will be AND’ed with any additional constraints passed either through the test job options or the -J command-line option. In other words, any constraint passed in access will always be present in the generated job script.

systems.partitions.environs
required:

No

default:

[]

A list of environment names that ReFrame will use to run regression tests on this partition. Each environment must be defined in the environments section of the configuration and the definition of the environment must be valid for this partition.

systems.partitions.container_platforms
Required:

No

Default:

[]

A list for container platform configuration objects. This will allow launching regression tests that use containers on this partition.

systems.partitions.modules
required:

No

default:

[]

A list of environment module objects to be loaded before running a regression test on this partition.

systems.partitions.time_limit
Required:

No

Default:

null

The time limit for the jobs submitted on this partition. When the value is null, no time limit is applied.

systems.partitions.env_vars
Required:

No

Default:

[]

A list of environment variables to be set before running a regression test on this partition. Each environment variable is specified as a two-element list containing the variable name and its value. You may reference other environment variables when defining an environment variable here. ReFrame will expand its value. Variables are set after the environment modules are loaded.

New in version 4.0.0.

systems.partitions.variables

Deprecated since version 4.0.0: Please use env_vars instead. If specified in conjunction with env_vars, it will be ignored.

systems.partitions.max_jobs
Required:

No

Default:

8

The maximum number of concurrent regression tests that may be active (i.e., not completed) on this partition. This option is relevant only when ReFrame executes with the asynchronous execution policy.

systems.partitions.prepare_cmds
Required:

No

Default:

[]

List of shell commands to be emitted before any environment loading commands are emitted.

New in version 3.5.0.

systems.partitions.resources
Required:

No

Default:

[]

A list of job scheduler resource specification objects.

systems.partitions.processor
Required:

No

Default:

{}

Processor information for this partition stored in a processor info object. If not set, ReFrame will try to auto-detect this information (see Auto-detecting processor information for more information).

New in version 3.5.0.

Changed in version 3.7.0: ReFrame is now able to detect the processor information automatically.

systems.partitions.devices
Required:

No

Default:

[]

A list with device info objects for this partition.

New in version 3.5.0.

systems.partitions.features
Required:

No

Default:

[]

User defined features of the partition.

These are accessible through the features attribute of the current_partition and can also be selected through the extended syntax of valid_systems. The values of this list must be alphanumeric strings starting with a non-digit character and may also contain a -.

New in version 3.11.0.

systems.partitions.extras
Required:

No

Default:

{}

User defined attributes of the partition.

These are accessible through the extras attribute of the current_partition and can also be selected through the extended syntax of valid_systems. The attributes of this object must be alphanumeric strings starting with a non-digit character and their values can be of any type.

By default, the values of the scheduler and launcher of the partition are added to the partition’s extras, if not already present.

New in version 3.5.0.

Changed in version 4.6.0: The default scheduler and launcher extras are added.

Container Platform Configuration

ReFrame can launch containerized applications, but you need to configure properly a system partition in order to do that by defining a container platform configuration.

systems.partitions.container_platforms.type
Required:

Yes

The type of the container platform. Available values are the following:

  • Apptainer: The Apptainer container runtime.

  • Docker: The Docker container runtime.

  • Sarus: The Sarus container runtime.

  • Shifter: The Shifter container runtime.

  • Singularity: The Singularity container runtime.

systems.partitions.container_platforms.default
Required:

No

If set to true, this is the default container platform of this partition. If not specified, the default container platform is assumed to be the first in the list of container_platforms.

New in version 3.12.0.

systems.partitions.container_platforms.modules
Required:

No

Default:

[]

A list of environment module objects to be loaded when running containerized tests using this container platform.

systems.partitions.container_platforms.env_vars
Required:

No

Default:

[]

List of environment variables to be set when running containerized tests using this container platform. Each environment variable is specified as a two-element list containing the variable name and its value. You may reference other environment variables when defining an environment variable here. ReFrame will expand its value. Variables are set after the environment modules are loaded.

New in version 4.0.0.

systems.partitions.container_platforms.variables

Deprecated since version 4.0.0: Please use env_vars instead. If specified in conjunction with env_vars, it will be ignored.

Custom Job Scheduler Resources

ReFrame allows you to define custom scheduler resources for each partition that you can then transparently access through the extra_resources attribute of a regression test.

systems.partitions.resources.name
required:

Yes

The name of this resources. This name will be used to request this resource in a regression test’s extra_resources.

systems.partitions.resources.options
required:

No

default:

[]

A list of options to be passed to this partition’s job scheduler. The option strings can contain placeholders of the form {placeholder_name}. These placeholders may be replaced with concrete values by a regression test through the extra_resources attribute.

For example, one could define a gpu resource for a multi-GPU system that uses Slurm as follows:

'resources': [
    {
        'name': 'gpu',
        'options': ['--gres=gpu:{num_gpus_per_node}']
    }
]

A regression test then may request this resource as follows:

self.extra_resources = {'gpu': {'num_gpus_per_node': '8'}}

And the generated job script will have the following line in its preamble:

#SBATCH --gres=gpu:8

A resource specification may also start with #PREFIX, in which case #PREFIX will replace the standard job script prefix of the backend scheduler of this partition. This is useful in cases of job schedulers like Slurm, that allow alternative prefixes for certain features. An example is the DataWarp functionality of Slurm which is supported by the #DW prefix. One could then define DataWarp related resources as follows:

'resources': [
    {
        'name': 'datawarp',
        'options': [
            '#DW jobdw capacity={capacity} access_mode={mode} type=scratch',
            '#DW stage_out source={out_src} destination={out_dst} type={stage_filetype}'
        ]
    }
]

A regression test that needs to make use of that resource, it can set its extra_resources as follows:

self.extra_resources = {
    'datawarp': {
        'capacity': '100GB',
        'mode': 'striped',
        'out_src': '$DW_JOB_STRIPED/name',
        'out_dst': '/my/file',
        'stage_filetype': 'file'
    }
}

Note

For the pbs and torque backends, options accepted in the access and resources parameters may either refer to actual qsub options or may just be resources specifications to be passed to the -l option. The backend assumes a qsub option, if the options passed in these attributes start with a -.

Environment Configuration

Environments defined in this section will be used for running regression tests. They are associated with system partitions.

environments.name
Required:

Yes

The name of this environment.

environments.modules
Required:

No

Default:

[]

A list of environment module objects to be loaded when this environment is loaded.

environments.env_vars
Required:

No

Default:

[]

A list of environment variables to be set when loading this environment. Each environment variable is specified as a two-element list containing the variable name and its value. You may reference other environment variables when defining an environment variable here. ReFrame will expand its value. Variables are set after the environment modules are loaded.

New in version 4.0.0.

environments.variables

Deprecated since version 4.0.0: Please use env_vars instead. If specified in conjunction with env_vars, it will be ignored.

environments.features
Required:

No

Default:

[]

User defined features of the environment. These are accessible through the features attribute of the current_environ and can also be selected through the extended syntax of valid_prog_environs. The values of this list must be alphanumeric strings starting with a non-digit character and may also contain a -.

New in version 3.11.0.

environments.extras
Required:

No

Default:

{}

User defined attributes of the environment. These are accessible through the extras attribute of the current_environ and can also be selected through the extended syntax of valid_prog_environs. The attributes of this object must be alphanumeric strings starting with a non-digit character and their values can be of any type.

New in version 3.9.1.

environments.prepare_cmds
Required:

No

Default:

[]

List of shell commands to be emitted before any commands that load the environment.

New in version 4.3.0.

environments.cc
Required:

No

Default:

"cc"

The C compiler to be used with this environment.

environments.cxx
Required:

No

Default:

"CC"

The C++ compiler to be used with this environment.

environments.ftn
Required:

No

Default:

"ftn"

The Fortran compiler to be used with this environment.

environments.cppflags
Required:

No

Default:

[]

A list of C preprocessor flags to be used with this environment by default.

environments.cflags
Required:

No

Default:

[]

A list of C flags to be used with this environment by default.

environments.cxxflags
Required:

No

Default:

[]

A list of C++ flags to be used with this environment by default.

environments.fflags
Required:

No

Default:

[]

A list of Fortran flags to be used with this environment by default.

environments.ldflags
Required:

No

Default:

[]

A list of linker flags to be used with this environment by default.

environments.nvcc
Required:

No

Default:

"nvcc"

The NVIDIA CUDA compiler to be used with this environment.

New in version 4.6.

environments.target_systems
Required:

No

Default:

["*"]

A list of systems or system/partitions combinations that this environment definition is valid for. A * entry denotes any system. In case of multiple definitions of an environment, the most specific to the current system partition will be used. For example, if the current system/partition combination is daint:mc, the second definition of the PrgEnv-gnu environment will be used:

'environments': [
    {
        'name': 'PrgEnv-gnu',
        'modules': ['PrgEnv-gnu']
    },
    {
        'name': 'PrgEnv-gnu',
        'modules': ['PrgEnv-gnu', 'openmpi'],
        'cc':  'mpicc',
        'cxx': 'mpicxx',
        'ftn': 'mpif90',
        'target_systems': ['daint:mc']
    }
]

However, if the current system was daint:gpu, the first definition would be selected, despite the fact that the second definition is relevant for another partition of the same system. To better understand this, ReFrame resolves definitions in a hierarchical way. It first looks for definitions for the current partition, then for the containing system and, finally, for global definitions (the * pseudo-system).

Logging Configuration

Logging in ReFrame is handled by logger objects which further delegate message to logging handlers which are eventually responsible for emitting or sending the log records to their destinations. You may define different logger objects per system but not per partition.

logging.level
Required:

No

Default:

"undefined"

The level associated with this logger object. There are the following levels in decreasing severity order:

  • critical: Catastrophic errors; the framework cannot proceed with its execution.

  • error: Normal errors; the framework may or may not proceed with its execution.

  • warning: Warning messages.

  • info: Informational messages.

  • verbose: More informational messages.

  • debug: Debug messages.

  • debug2: Further debug messages.

  • undefined: This is the lowest level; does not filter any message.

If a message is logged by the framework, its severity level will be checked by the logger and if it is higher from the logger’s level, it will be passed down to its handlers.

New in version 3.3: The debug2 and undefined levels are added.

Changed in version 3.3: The default level is now undefined.

logging.handlers
Required:

Yes

A list of logging handlers responsible for handling normal framework output.

logging.handlers_perflog
Required:

Yes

A list of logging handlers responsible for handling performance data from tests.

logging.perflog_compat
Required:

No

Default:

false

Emit a separate log record for each performance variable. Set this option to true if you want to keep compatibility with the performance logging prior to ReFrame 4.0.

logging.target_systems
Required:

No

Default:

["*"]

A list of systems or system/partitions combinations that this logging configuration is valid for. For a detailed description of this property, have a look at the target_systems definition for environments.

Common logging handler properties

All logging handlers share the following set of common attributes:

logging.handlers.type
logging.handlers_perflog.type
Required:

Yes

The type of handler. There are the following types available:

  • file: This handler sends log records to file. See here for more details.

  • filelog: This handler sends performance log records to files. See here for more details.

  • graylog: This handler sends performance log records to Graylog. See here for more details.

  • stream: This handler sends log records to a file stream. See here for more details.

  • syslog: This handler sends log records to a Syslog facility. See here for more details.

  • httpjson: This handler sends log records in JSON format using HTTP post requests. See here for more details.

logging.handlers.level
logging.handlers_perflog.level
Required:

No

Default:

"info"

The log level associated with this handler.

logging.handlers.format
logging.handlers_perflog.format
Required:

No

Default:

"%(message)s"

Log record format string.

ReFrame accepts all log record attributes from Python’s logging mechanism and adds the following attributes:

%(check_build_locally)s

The value of the build_locally attribute.

%(check_build_time_limit)s

The value of the build_time_limit attribute.

%(check_descr)s

The value of the descr attribute.

%(check_display_name)s

The value of the display_name attribute.

%(check_environ)s

The name of the test’s current_environ.

%(check_env_vars)s

The value of the env_vars attribute.

%(check_exclusive_access)s

The value of the exclusive_access attribute.

%(check_executable)s

The value of the executable attribute.

%(check_executable_opts)s

The value of the executable_opts attribute.

%(check_extra_resources)s

The value of the extra_resources attribute.

%(check_hashcode)s

The unique hash associated with this test.

%(check_info)s

Various information about this test; essentially the return value of the test’s info() function.

%(check_job_completion_time)s

Same as the (check_job_completion_time_unix)s but formatted according to datefmt.

%(check_job_completion_time_unix)s

The completion time of the associated run job (see completion_time).

%(check_job_exitcode)s

The exit code of the associated run job.

%(check_job_nodelist)s

The list of nodes that the associated run job has run on.

%(check_job_submit_time)s

The submission time of the associated run job (see submit_time).

%(check_jobid)s

The ID of the associated run job.

%(check_keep_files)s

The value of the keep_files attribute.

%(check_local)s

The value of the local attribute.

%(check_maintainers)s

The value of the maintainers attribute.

%(check_max_pending_time)s

The value of the max_pending_time attribute.

%(check_modules)s

The value of the modules attribute.

%(check_name)s

The value of the name attribute.

%(check_num_cpus_per_task)s

The value of the num_cpus_per_task attribute.

%(check_num_gpus_per_node)s

The value of the num_gpus_per_node attribute.

%(check_num_tasks)s

The value of the num_tasks attribute.

%(check_num_tasks_per_core)s

The value of the num_tasks_per_core attribute.

%(check_num_tasks_per_node)s

The value of the num_tasks_per_node attribute.

%(check_num_tasks_per_socket)s

The value of the num_tasks_per_socket attribute.

%(check_outputdir)s

The value of the outputdir attribute.

%(check_partition)s

The name of the test’s current_partition.

%(check_perfvalues)s

All the performance variables of the test combined. These will be formatted according to format_perfvars.

%(check_postbuild_cmds)s

The value of the postbuild_cmds attribute.

%(check_postrun_cmds)s

The value of the postrun_cmds attribute.

%(check_prebuild_cmds)s

The value of the prebuild_cmds attribute.

%(check_prefix)s

The value of the prefix attribute.

%(check_prerun_cmds)s

The value of the prerun_cmds attribute.

%(check_result)s

The result of the test (pass or fail).

%(check_readonly_files)s

The value of the readonly_files attribute.

%(check_short_name)s

The value of the short_name attribute.

%(check_sourcepath)s

The value of the sourcepath attribute.

%(check_sourcesdir)s

The value of the sourcesdir attribute.

%(check_stagedir)s

The value of the stagedir attribute.

%(check_strict_check)s

The value of the strict_check attribute.

%(check_system)s

The name of the test’s current_system.

%(check_tags)s

The value of the tags attribute.

%(check_time_limit)s

The value of the time_limit attribute.

%(check_unique_name)s

The value of the unique_name attribute.

%(check_use_multithreading)s

The value of the use_multithreading attribute.

%(check_valid_prog_environs)s

The value of the valid_prog_environs attribute.

%(check_valid_systems)s

The value of the valid_systems attribute.

%(check_variables)s

DEPRECATED: Please use %(check_env_vars)s instead.

%(osuser)s

The name of the OS user running ReFrame.

%(osgroup)s

The name of the OS group running ReFrame.

%(version)s

The ReFrame version.

ReFrame allows you to log any test variable, parameter or property if they are marked as “loggable”. The log record attribute will have the form %(check_NAME)s where NAME is the variable name, the parameter name or the property name that is marked as loggable.

There is also the special %(check_#ALL)s format specifier which expands to all the loggable test attributes. These include all the above specifiers and any additional loggable variables or parameters defined by the test. On expanding this specifier, ReFrame will try to guess the delimiter to use for separating the different attributes based on the existing format. If it cannot guess it, it will default to |.

Since this can lead to very long records, you may consider using it with the ignore_keys parameter to filter out some attributes that are not of interest.

New in version 3.3: Allow arbitrary test attributes to be logged.

New in version 3.4.2: Allow arbitrary job attributes to be logged.

Changed in version 3.11.0: Limit the number of attributes that can be logged. User attributes or properties must be explicitly marked as “loggable” in order to be selectable for logging.

New in version 4.0: The %(check_result)s specifier is added.

New in version 4.3: The %(check_#ALL)s special specifier is added.

logging.handlers.format_perfvars
logging.handlers_perflog.format_perfvars
Required:

No

Default:

""

Format specifier for logging the performance variables.

This defines how the %(check_perfvalues)s will be formatted. Since a test may define multiple performance variables, the formatting specified in this field will be repeated for each performance variable sequentially in the same line.

Important

The last character of this format will be interpreted as the final delimiter of the formatted performance variables to the rest of the record.

The following log record attributes are defined additionally by this format specifier:

Log record attribute

Description

%(check_perf_lower_thres)s

The lower threshold of the logged performance variable.

%(check_perf_ref)s

The reference value of the logged performance variable.

%(check_perf_unit)s

The measurement unit of the logged performance variable.

%(check_perf_upper_thres)s

The upper threshold of the logged performance variable.

%(check_perf_value)s

The actual value of the logged performance variable.

%(check_perf_var)s

The name of the logged performance variable.

Important

ReFrame versions prior to 4.0 logged a separate line for each performance variable and the %(check_perf_*)s attributes could be used directly in the format. You can re-enable this behavior by setting the config.logging.perflog_compat logging configuration parameter.

New in version 4.0.0.

logging.handlers.datefmt
logging.handlers_perflog.datefmt
Required:

No

Default:

"%FT%T"

Time format to be used for printing timestamps fields. There are two timestamp fields available: %(asctime)s and %(check_job_completion_time)s. In addition to the format directives supported by the standard library’s time.strftime() function, ReFrame allows you to use the %:z directive – a GNU date extension – that will print the time zone difference in a RFC3339 compliant way, i.e., +/-HH:MM instead of +/-HHMM.

The file log handler

This log handler handles output to normal files. The additional properties for the file handler are the following:

logging.handlers..file..name
logging.handlers_perflog..file..name
Required:

No

The name of the file where this handler will write log records. If not specified, ReFrame will create a log file prefixed with rfm- in the system’s temporary directory.

Changed in version 3.3: The name parameter is no more required and the default log file resides in the system’s temporary directory.

logging.handlers..file..append
logging.handlers_perflog..file..append
Required:

No

Default:

false

Controls whether this handler should append to its file or not.

logging.handlers..file..timestamp
logging.handlers_perflog..file..timestamp
Required:

No

Default:

false

Append a timestamp to this handler’s log file. This property may also accept a date format as described in the datefmt property. If the handler’s name property is set to filename.log and this property is set to true or to a specific timestamp format, the resulting log file will be filename_<timestamp>.log.

The filelog log handler

This handler is meant for performance logging only and logs the performance of a test in one or more files. The additional properties for the filelog handler are the following:

logging.handlers_perflog..filelog..basedir
Required:

No

Default:

"./perflogs"

The base directory of performance data log files.

logging.handlers_perflog..filelog..ignore_keys

A list of log record format specifiers that will be ignored by the special %(check_#ALL)s specifier.

New in version 4.3.

logging.handlers_perflog..filelog..prefix
Required:

Yes

This is a directory prefix (usually dynamic), appended to the basedir, where the performance logs of a test will be stored. This attribute accepts any of the check-specific formatting placeholders. This allows to create dynamic paths based on the current system, partition and/or programming environment a test executes with. For example, a value of %(check_system)s/%(check_partition)s would generate the following structure of performance log files:

{basedir}/
   system1/
       partition1/
           <test_class_name>.log
       partition2/
           <test_class_name>.log
       ...
   system2/
   ...
logging.handlers_perflog..filelog..append
Required:

No

Default:

true

Open each log file in append mode.

Changed in version 4.0.0: The filelog handler is very cautious when generating a test log file: if a change is detected in the information that is being logged, the hanlder will not append to the same file, but it will instead create a new one, saving the old file using the .h<N> suffix, where N is an integer that is increased every time a new file is being created due to such changes. Examples of changes in the logged information are when the log record format changes or a new performance metric is added, deleted or has its name changed. This behavior guarantees that each log file is consistent and it will not break existing parsers.

Changed in version 4.3: In the generated log file, the name of the test class name is used instead of the test’s short name (which included the test’s hash). This allows the results of different variants of a parameterized test to be stored in the same log file facilitating post-processing.

The graylog log handler

This handler is meant for performance logging only and sends log records to a Graylog server. The additional properties for the graylog handler are the following:

logging.handlers_perflog..graylog..address
Required:

Yes

The address of the Graylog server defined as host:port.

logging.handlers_perflog..graylog..extras
Required:

No

Default:

{}

A set of optional key/value pairs to be passed with each log record to the server. These may depend on the server configuration.

This log handler uses internally pygelf. If pygelf is not available, this log handler will be ignored. GELF is a format specification for log messages that are sent over the network. The graylog handler sends log messages in JSON format using an HTTP POST request to the specified address. More details on this log format may be found here. An example configuration of this handler for performance logging is shown here:

{
    'type': 'graylog',
    'address': 'graylog-server:12345',
    'level': 'info',
    'format': '%(message)s',
    'extras': {
        'facility': 'reframe',
        'data-version': '1.0'
    }
}

Although the format attribute is defined for this handler, it is not only the log message that will be transmitted the Graylog server. This handler transmits the whole log record, meaning that all the information will be available and indexable at the remote end.

The stream log handler

This handler sends log records to a file stream. The additional properties for the stream handler are the following:

logging.handlers..stream..name
logging.handlers_perflog..stream..name
Required:

No

Default:

"stdout"

The name of the file stream to send records to. There are only two available streams:

  • stdout: the standard output.

  • stderr: the standard error.

The syslog log handler

This handler sends log records to UNIX syslog. The additional properties for the syslog handler are the following:

logging.handlers..syslog..socktype
logging.handlers_perflog..syslog..socktype
Required:

No

Default:

"udp"

The socket type where this handler will send log records to. There are two socket types:

  • udp: A UDP datagram socket.

  • tcp: A TCP stream socket.

logging.handlers..syslog..facility
logging.handlers_perflog..syslog..facility
Required:

No

Default:

"user"

The Syslog facility where this handler will send log records to. The list of supported facilities can be found here.

logging.handlers..syslog..address
logging.handlers_perflog..syslog..address
Required:

Yes

The socket address where this handler will connect to. This can either be of the form <host>:<port> or simply a path that refers to a Unix domain socket.

The httpjson log handler

This handler sends log records in JSON format to a server using HTTP POST requests. The additional properties for the httpjson handler are the following:

logging.handlers_perflog..httpjson..url
Required:

Yes

The URL to be used in the HTTP(S) request server.

logging.handlers_perflog..httpjson..extra_headers
Required:

No

Default:

{}

A set of optional key/value pairs to be sent as HTTP message headers (e.g. API keys). These may depend on the server configuration.

New in version 4.2.

logging.handlers_perflog..httpjson..extras
Required:

No

Default:

{}

A set of optional key/value pairs to be passed with each log record to the server. These may depend on the server configuration.

logging.handlers_perflog..httpjson..ignore_keys
Required:

No

Default:

[]

These keys will be excluded from the log record that will be sent to the server.

The httpjson handler sends log messages in JSON format using an HTTP POST request to the specified URL.

An example configuration of this handler for performance logging is shown here:

{
    'type': 'httpjson',
    'url': 'http://httpjson-server:12345/rfm',
    'level': 'info',
    'extra_headers': {'Authorization': 'Token YOUR_API_TOKEN'},
    'extras': {
        'facility': 'reframe',
        'data-version': '1.0'
    },
    'ignore_keys': ['check_perfvalues']
}

This handler transmits the whole log record, meaning that all the information will be available and indexable at the remote end.

logging.handlers_perflog..httpjson..debug
Required:

No

Default:

false

If set, the httpjson handler will not attempt to send the data to the server, but it will instead dump the JSON record in the current directory. The filename has the following form: httpjson_record_<timestamp>.json.

New in version 4.1.

logging.handlers_perflog..httpjson..json_formatter

A callable for converting the log record into JSON.

The formatter’s signature is the following:

json_formatter(record: object, extras: Dict[str, str], ignore_keys: Set[str]) str
Parameters:
  • record – The prepared log record. The log record is a simple Python object with all the attributes listed in format, as well as all the default Python log record attributes. In addition to those, there is also the special __rfm_check__ attribute that contains a reference to the actual test for which the performance is being logged.

  • extras – Any extra attributes specified in extras.

  • ignore_keys – The set of keys specified in ignore_keys. ReFrame always adds the default Python log record attributes in this set.

Returns:

A string representation of the JSON record to be sent to the server or None if the record should not be sent to the server.

Note

This configuration parameter can only be used in a Python configuration file.

New in version 4.1.

Execution Mode Configuration

ReFrame allows you to define groups of command line options that are collectively called execution modes. An execution mode can then be selected from the command line with the --mode option. The options of an execution mode will be passed to ReFrame as if they were specified in the command line.

modes.name
Required:

Yes

The name of this execution mode. This can be used with the --mode command line option to invoke this mode.

modes.options
Required:

No

Default:

[]

The command-line options associated with this execution mode.

modes.target_systems
Required:

No

Default:

["*"]

A list of systems only that this execution mode is valid for. For a detailed description of this property, have a look at the target_systems definition for environments.

General Configuration

general.check_search_path
Required:

No

Default:

["${RFM_INSTALL_PREFIX}/checks/"]

A list of paths (files or directories) where ReFrame will look for regression test files. If the search path is set through the environment variable, it should be a colon separated list. If specified from command line, the search path is constructed by specifying multiple times the command line option.

general.check_search_recursive
Required:

No

Default:

false

Search directories in the search path recursively.

general.clean_stagedir
Required:

No

Default:

true

Clean stage directory of tests before populating it.

New in version 3.1.

general.colorize
Required:

No

Default:

true

Use colors in output. The command-line option sets the configuration option to false.

general.compress_report
Required:

No

Default:

false

Compress the generated run report file. See the documentation of the --compress-report option for more information.

New in version 3.12.0.

general.git_timeout
Required:

No

Default:

5

Timeout value in seconds used when checking if a git repository exists.

general.dump_pipeline_progress

Dump pipeline progress for the asynchronous execution policy in pipeline-progress.json. This option is meant for debug purposes only.

Required:

No

Default:

False

New in version 3.10.0.

general.pipeline_timeout

Timeout in seconds for advancing the pipeline in the asynchronous execution policy.

ReFrame’s asynchronous execution policy will try to advance as many tests as possible in their pipeline, but some tests may take too long to proceed (e.g., due to copying of large files) blocking the advancement of previously started tests. If this timeout value is exceeded and at least one test has progressed, ReFrame will stop processing new tests and it will try to further advance tests that have already started. See Tweaking the throughput and interactivity of test jobs in the asynchronous execution policy for more guidance on how to set this.

Required:

No

Default:

10

New in version 3.10.0.

general.perf_info_level
Required:

No

Default:

"info"

The log level at which the immediate performance info will be printed.

As soon as a performance test is finished, ReFrame will log its performance on the standard output immediately. This option controls at which verbosity level this info will appear.

For a list of available log levels, refer to the level logger configuration parameter.

New in version 4.0.0.

general.remote_detect
Required:

No

Default:

false

Try to auto-detect processor information of remote partitions as well. This may slow down the initialization of the framework, since it involves submitting auto-detection jobs to the remote partitions. For more information on how ReFrame auto-detects processor information, you may refer to Auto-detecting processor information.

New in version 3.7.0.

general.remote_workdir
Required:

No

Default:

"."

The temporary directory prefix that will be used to create a fresh ReFrame clone, in order to auto-detect the processor information of a remote partition.

New in version 3.7.0.

general.ignore_check_conflicts
Required:

No

Default:

false

Ignore test name conflicts when loading tests.

Deprecated since version 3.8.0: This option will be removed in a future version.

general.trap_job_errors
Required:

No

Default:

false

Trap command errors in the generated job scripts and let them exit immediately.

New in version 3.2.

general.keep_stage_files
Required:

No

Default:

false

Keep stage files of tests even if they succeed.

general.module_map_file
Required:

No

Default:

""

File containing module mappings.

general.module_mappings
Required:

No

Default:

[]

A list of module mappings. If specified through the environment variable, the mappings must be separated by commas. If specified from command line, multiple module mappings are defined by passing the command line option multiple times.

general.non_default_craype
Required:

No

Default:

false

Test a non-default Cray Programming Environment. This will emit some special instructions in the generated build and job scripts. See also --non-default-craype for more details.

general.purge_environment
Required:

No

Default:

false

Purge any loaded environment modules before running any tests.

general.report_file
Required:

No

Default:

"${HOME}/.reframe/reports/run-report-{sessionid}.json"

The file where ReFrame will store its report.

New in version 3.1.

Changed in version 3.2: Default value has changed to avoid generating a report file per session.

Changed in version 4.0.0: Default value was reverted back to generate a new file per run.

general.report_junit
Required:

No

Default:

null

The file where ReFrame will store its report in JUnit format. The report adheres to the XSD schema here.

New in version 3.6.0.

general.resolve_module_conflicts
Required:

No

Default:

true

ReFrame by default resolves any module conflicts and emits the right sequence of module unload and module load commands, in order to load the requested modules. This option disables this behavior if set to false.

You should avoid using this option for modules system that cannot handle module conflicts automatically, such as early Tmod verions.

Disabling the automatic module conflict resolution, however, can be useful when modules in a remote system partition are not present on the host where ReFrame runs. In order to resolve any module conflicts and generate the right load sequence of modules, ReFrame loads temporarily the requested modules and tracks any conflicts along the way. By disabling this option, ReFrame will simply emit the requested module load commands without attempting to load any module.

New in version 3.6.0.

general.save_log_files
Required:

No

Default:

false

Save any log files generated by ReFrame to its output directory

general.target_systems
Required:

No

Default:

["*"]

A list of systems or system/partitions combinations that these general options are valid for. For a detailed description of this property, have a look at the target_systems definition for environments.

general.timestamp_dirs
Required:

No

Default:

""

Append a timestamp to ReFrame directory prefixes. Valid formats are those accepted by the time.strftime() function. If specified from the command line without any argument, "%FT%T" will be used as a time format.

general.unload_modules
Required:

No

Default:

[]

A list of environment module objects to unload before executing any test. If specified using an the environment variable, a space separated list of modules is expected. If specified from the command line, multiple modules can be passed by passing the command line option multiple times.

general.use_login_shell
Required:

No

Default:

false

Use a login shell for the generated job scripts. This option will cause ReFrame to emit -l in the shebang of shell scripts. This option, if set to true, may cause ReFrame to fail, if the shell changes permanently to a different directory during its start up.

general.user_modules
Required:

No

Default:

[]

A list of environment module objects to be loaded before executing any test. If specified using an the environment variable, a space separated list of modules is expected. If specified from the command line, multiple modules can be passed by passing the command line option multiple times.

general.verbose
Required:

No

Default:

0

Set the verbosity level of the output. The higher the number, the more verbose the output will be. If set to a negative number, this will decrease the verbosity level.

Module Objects

New in version 3.3.

A module object in ReFrame’s configuration represents an environment module. It can either be a simple string or a JSON object with the following attributes:

environments.modules.name
systems.modules.name
systems.partitions.modules.name
systems.partitions.container_platforms.modules.name
Required:

Yes

The name of the module.

environments.modules.collection
systems.modules.collection
systems.partitions.modules.collection
systems.partitions.container_platforms.modules.collection
Required:

No

Default:

false

A boolean value indicating whether this module refers to a module collection. Module collections are treated differently from simple modules when loading.

environments.modules.path
systems.modules.path
systems.partitions.modules.path
systems.partitions.container_platforms.modules.path
Required:

No

Default:

null

If the module is not present in the default MODULEPATH, the module’s location can be specified here. ReFrame will make sure to set and restore the MODULEPATH accordingly for loading the module.

New in version 3.5.0.

See also

Module collections with Environment Modules and Lmod.

Processor Info

New in version 3.5.0.

A processor info object in ReFrame’s configuration is used to hold information about the processor of a system partition and is made available to the tests through the processor attribute of the current_partition.

Note

In the following the term logical CPUs refers to the smallest processing unit recognized by the OS. Depending on the microarchitecture, this can either be a core or a hardware thread in processors that support simultaneous multithreading and this feature is enabled. Therefore, properties such as num_cpus_per_core may have a value greater than one.

systems.partitions.processor.arch
Required:

No

Default:

None

The microarchitecture of the processor.

systems.partitions.processor.model
Required:

No

Default:

None

The model of the processor.

New in version 4.6.

systems.partitions.processor.platform
Required:

No

Default:

None

The hardware platform for this processor (e.g., x86_64, arm64 etc.)

New in version 4.6.

systems.partitions.processor.num_cpus
Required:

No

Default:

None

Number of logical CPUs.

systems.partitions.processor.num_cpus_per_core
Required:

No

Default:

None

Number of logical CPUs per core.

systems.partitions.processor.num_cpus_per_socket
Required:

No

Default:

None

Number of logical CPUs per socket.

systems.partitions.processor.num_sockets
Required:

No

Default:

None

Number of sockets.

systems.partitions.processor.topology
Required:

No

Default:

None

Processor topology. An example follows:

'topology': {
   'numa_nodes': ['0x000000ff'],
   'sockets': ['0x000000ff'],
   'cores': ['0x00000003', '0x0000000c',
             '0x00000030', '0x000000c0'],
   'caches': [
      {
            'type': 'L3',
            'size': 6291456,
            'linesize': 64,
            'associativity': 0,
            'num_cpus': 8,
            'cpusets': ['0x000000ff']
      },
      {
            'type': 'L2',
            'size': 262144,
            'linesize': 64,
            'associativity': 4,
            'num_cpus': 2,
            'cpusets': ['0x00000003', '0x0000000c',
                        '0x00000030', '0x000000c0']
      },
      {
            'type': 'L1',
            'size': 32768,
            'linesize': 64,
            'associativity': 0,
            'num_cpus': 2,
            'cpusets': ['0x00000003', '0x0000000c',
                        '0x00000030', '0x000000c0']
      }
   ]
}

Device Info

New in version 3.5.0.

A device info object in ReFrame’s configuration is used to hold information about a specific type of devices in a system partition and is made available to the tests through the devices attribute of the current_partition.

systems.partitions.devices.type
Required:

No

Default:

None

The type of the device, for example "gpu".

systems.partitions.devices.arch
Required:

No

Default:

None

The microarchitecture of the device.

systems.partitions.devices.model
Required:

No

Default:

None

The model of the device.

New in version 4.6.

systems.partitions.devices.num_devices
Required:

No

Default:

None

Number of devices of this type inside the system partition.