gwcelery.conf module

GWCelery application configuration.

This module defines configuration variables and default values, including both generic options for Celery as well as options that control the behavior of specific GWCelery tasks.

To override the configuration, define the CELERY_CONFIG_MODULE environment variable to the fully qualified name of any Python module that can be located in sys.path, including any of the following presets:

gwcelery.conf.expose_to_public = False

Set to True if events meeting the public alert threshold really should be exposed to the public.

gwcelery.conf.lvalert_host = 'lvalert-playground.cgca.uwm.edu'

LVAlert host.

gwcelery.conf.gracedb_host = 'gracedb-playground.ligo.org'

GraceDB host.

gwcelery.conf.voevent_broadcaster_address = ':5342'

The VOEvent broker will bind to this address to send GCNs. This should be a string of the form host:port. If host is empty, then listen on all available interfaces.

gwcelery.conf.voevent_broadcaster_whitelist = []

List of hosts from which the broker will accept connections. If empty, then completely disable the broker’s broadcast capability.

gwcelery.conf.voevent_receiver_address = '68.169.57.253:8099'

The VOEvent listener will connect to this address to receive GCNs. For options, see GCN’s list of available VOEvent servers. If this is an empty string, then completely disable the GCN listener.

gwcelery.conf.superevent_d_t_start = {'gstlal': 1.0, 'mbtaonline': 1.0, 'pycbc': 1.0, 'spiir': 1.0}

Pipeline based lower extent of superevent segments. For cwb and lib this is decided from extra attributes.

gwcelery.conf.superevent_d_t_end = {'gstlal': 1.0, 'mbtaonline': 1.0, 'pycbc': 1.0, 'spiir': 1.0}

Pipeline based upper extent of superevent segments For cwb and lib this is decided from extra attributes.

gwcelery.conf.superevent_query_d_t_start = 100.0

Lower extent of superevents query

gwcelery.conf.superevent_query_d_t_end = 100.0

Upper extent of superevents query

gwcelery.conf.superevent_default_d_t_start = 1.0

Default lower extent of superevent segments

gwcelery.conf.superevent_default_d_t_end = 1.0

Default upper extent for superevent segments

gwcelery.conf.superevent_far_threshold = 0.0002777777777777778

Maximum false alarm rate to consider events superevents.

gwcelery.conf.preliminary_alert_far_threshold = {'burst': 3.1709791983764586e-08, 'cbc': 1.9290123456790122e-07, 'test': 3.8580246913580245e-07}

Group specific maximum false alarm rate to consider sending preliminary alerts.

gwcelery.conf.preliminary_alert_trials_factor = {'burst': 3.0, 'cbc': 4.0}

Trials factor corresponding to trigger categories. For CBC and Burst, trials factor is the number of pipelines. CBC pipelines are gstlal, pycbc, mbtaonline and spiir. Burst searches are cwb.allsky, cwb.bbh and cwb.imbh.

gwcelery.conf.orchestrator_timeout = 300.0

The orchestrator will wait this many seconds from the time of the creation of a new superevent to the time that annotations begin, in order to let the superevent manager’s decision on the preferred event stabilize.

gwcelery.conf.pe_timeout = 345.0

The orchestrator will wait this many seconds from the time of the creation of a new superevent to the time that parameter estimation begins, in case the preferred event is updated with high latency.

gwcelery.conf.check_vector_prepost = {'CWB': [0.5, 0.5], 'Fermi': [2, 2], 'HardwareInjection': [2, 2], 'LIB': [0.5, 0.5], 'MBTAOnline': [2, 2], 'SNEWS': [10, 10], 'Swift': [2, 2], 'gstlal': [2, 2], 'oLIB': [0.5, 0.5], 'pycbc': [2, 2], 'spiir': [2, 2]}

Seconds before and after the superevent start and end times which the DQ vector check will include in its check. Pipeline dependent.

gwcelery.conf.uses_gatedhoft = {'CWB': True, 'Fermi': False, 'HardwareInjection': False, 'LIB': False, 'MBTAOnline': True, 'SNEWS': False, 'Swift': False, 'gstlal': True, 'oLIB': False, 'pycbc': True, 'spiir': True}

Whether or not a pipeline uses gated h(t). Determines whether or not the DMT-DQ_VECTOR will be analyzed for data quality.

gwcelery.conf.llhoft_glob = '/dev/shm/kafka/{detector}_O2/*.gwf'

File glob for playground low-latency h(t) frames. Currently points to O2 replay data.

gwcelery.conf.llhoft_channels = {'H1:DMT-DQ_VECTOR': 'dmt_dq_vector_bits', 'H1:GDS-CALIB_STATE_VECTOR': 'ligo_state_vector_bits', 'L1:DMT-DQ_VECTOR': 'dmt_dq_vector_bits', 'L1:GDS-CALIB_STATE_VECTOR': 'ligo_state_vector_bits', 'V1:DQ_ANALYSIS_STATE_VECTOR': 'virgo_state_vector_bits'}

Low-latency h(t) state vector configuration. This is a dictionary consisting of a channel and its bitmask, as defined in gwcelery.tasks.detchar.

gwcelery.conf.idq_channels = ['H1:IDQ-PGLITCH_OVL_32_2048', 'L1:IDQ-PGLITCH_OVL_32_2048']

Low-latency iDQ p(glitch) channel names from O2 replay.

gwcelery.conf.idq_pglitch_thresh = 0.95

If P(Glitch) is above this threshold, and idq_veto for the pipeline is true, DQV will be labeled for the event.

gwcelery.conf.idq_veto = {'CWB': False, 'Fermi': False, 'HardwareInjection': False, 'LIB': False, 'MBTAOnline': False, 'SNEWS': False, 'Swift': False, 'gstlal': False, 'oLIB': False, 'pycbc': False, 'spiir': False}

If true for a pipeline, iDQ values above the threshold defined in thres will cause DQV to be labeled. Currently all False, pending iDQ review (should be done before O3).

gwcelery.conf.low_latency_frame_types = {'H1': 'H1_O2_llhoft', 'L1': 'L1_O2_llhoft', 'V1': 'V1_O2_llhoft'}

Types of low latency frames used in Parameter Estimation with LALInference (see gwcelery.tasks.lalinference) and in cache creation for detchar checks (see gwcelery.tasks.detchar).

gwcelery.conf.high_latency_frame_types = {'H1': None, 'L1': None, 'V1': None}

Types of high latency frames used in Parameter Estimation with LALInference and in cache creation for detchar checks. They do not exist for O2Replay data. (see gwcelery.tasks.lalinference and gwcelery.tasks.detchar)

gwcelery.conf.strain_channel_names = {'H1': 'H1:GDS-CALIB_STRAIN_O2Replay', 'L1': 'L1:GDS-CALIB_STRAIN_O2Replay', 'V1': 'V1:Hrec_hoft_16384Hz_O2Replay'}

Names of h(t) channels used in Parameter Estimation with LALInference (see gwcelery.tasks.lalinference)

gwcelery.conf.state_vector_channel_names = {'H1': 'H1:GDS-CALIB_STATE_VECTOR', 'L1': 'L1:GDS-CALIB_STATE_VECTOR', 'V1': 'V1:DQ_ANALYSIS_STATE_VECTOR'}

Names of state vector channels used in Parameter Estimation with LALInference (see gwcelery.tasks.lalinference)

gwcelery.conf.pe_threshold = 4.133597883597884e-07

FAR threshold in Hz for Parameter Estimation. PE group now applies 1/(4 weeks) as a threshold. 86400 seconds = 1 day and 28 days = 4 weeks.

gwcelery.conf.pe_results_path = '/home/docs/public_html/online_pe'

Path to the results of Parameter Estimation (see gwcelery.tasks.lalinference)

gwcelery.conf.pe_results_url = 'https://ldas-jobs.ligo.caltech.edu/~docs/online_pe/'

URL of page where all the results of Parameter Estimation are outputted (see gwcelery.tasks.lalinference)

gwcelery.conf.raven_coincidence_windows = {'GRB_Burst': [-600, 60], 'GRB_CBC': [-5, 1], 'SNEWS': [-10, 10]}

Time coincidence windows passed to ligo-raven. External events and superevents of the appropriate type are considered to be coincident if within time window of each other.

gwcelery.conf.mock_events_simulate_multiple_uploads = False

If True, then upload each mock event several times in rapid succession with random jitter in order to simulate multiple pipeline uploads.

gwcelery.conf.development module

Application configuration for gracedb-dev1.ligo.org. Inherits all settings from gwcelery.conf.test, with the exceptions below.

gwcelery.conf.development.gracedb_host = 'gracedb-dev1.ligo.org'

GraceDB host.

gwcelery.conf.development.sentry_environment = 'development'

Record this environment tag <https://docs.sentry.io/enriching-error-data/environments/>)` in Sentry log messages.

gwcelery.conf.playground module

Application configuration for gracedb-playground.ligo.org.

gwcelery.conf.playground.sentry_environment = 'playground'

Record this environment tag <https://docs.sentry.io/enriching-error-data/environments/>)` in Sentry log messages.

gwcelery.conf.playground.mock_events_simulate_multiple_uploads = True

If True, then upload each mock event several times in rapid succession with random jitter in order to simulate multiple pipeline uploads.

gwcelery.conf.playground.orchestrator_timeout = 120.0

The orchestrator will wait this many seconds from the time of the creation of a new superevent to the time that annotations begin, in order to let the superevent manager’s decision on the preferred event stabilize.

gwcelery.conf.production module

Application configuration for gracedb.ligo.org. Inherits all settings from gwcelery.conf.playground, with the exceptions below.

gwcelery.conf.production.expose_to_public = True

Set to True if events meeting the public alert threshold really should be exposed to the public.

gwcelery.conf.production.lvalert_host = 'lvalert.cgca.uwm.edu'

LVAlert host.

gwcelery.conf.production.gracedb_host = 'gracedb.ligo.org'

GraceDB host.

gwcelery.conf.production.voevent_broadcaster_address = ':5341'

The VOEvent broker will bind to this address to send GCNs. This should be a string of the form host:port. If host is empty, then listen on all available interfaces.

gwcelery.conf.production.voevent_broadcaster_whitelist = ['capella2.gsfc.nasa.gov']

List of hosts from which the broker will accept connections. If empty, then completely disable the broker’s broadcast capability.

gwcelery.conf.production.llhoft_glob = '/dev/shm/kafka/{detector}/*.gwf'

File glob for low-latency h(t) frames.

gwcelery.conf.production.low_latency_frame_types = {'H1': 'H1_llhoft', 'L1': 'L1_llhoft', 'V1': 'V1_llhoft'}

Types of frames used in Parameter Estimation with LALInference (see gwcelery.tasks.lalinference) and in cache creation for detchar checks (see gwcelery.tasks.detchar).

gwcelery.conf.production.high_latency_frame_types = {'H1': 'H1_HOFT_C00', 'L1': 'L1_HOFT_C00', 'V1': 'V1Online'}

Types of high latency frames used in Parameter Estimation with LALInference (see gwcelery.tasks.lalinference) and in cache creation for detchar checks (see gwcelery.tasks.detchar).

gwcelery.conf.production.idq_channels = ['H1:IDQ-PGLITCH_OVL_16_4096', 'L1:IDQ-PGLITCH_OVL_16_4096']

Low-latency iDQ p(glitch) channel names from live O3 frames

gwcelery.conf.production.strain_channel_names = {'H1': 'H1:GDS-CALIB_STRAIN_CLEAN', 'L1': 'L1:GDS-CALIB_STRAIN_CLEAN', 'V1': 'V1:Hrec_hoft_16384Hz'}

Names of h(t) channels used in Parameter Estimation with LALInference (see gwcelery.tasks.lalinference)

gwcelery.conf.production.sentry_environment = 'production'

Record this environment tag <https://docs.sentry.io/enriching-error-data/environments/>)` in Sentry log messages.

gwcelery.conf.test module

Application configuration for gracedb-test.ligo.org. Inherits all settings from gwcelery.conf.playground, with the exceptions below.

gwcelery.conf.test.lvalert_host = 'lvalert-test.cgca.uwm.edu'

LVAlert host.

gwcelery.conf.test.gracedb_host = 'gracedb-test.ligo.org'

GraceDB host.

gwcelery.conf.test.sentry_environment = 'test'

Record this environment tag <https://docs.sentry.io/enriching-error-data/environments/>)` in Sentry log messages.