class ProcessorJob(models.Model):
"""Records information about running a processor."""
class Meta:
db_table = "processor_jobs"
indexes = [
models.Index(
fields=["created_at"],
name="processor_jobs_created_at",
# A partial index might be better here, given our queries we don't
# need to index the whole table. We need to update to Django 2.2
# for this to be supported.
# condition=Q(success=None, retried=False, no_retry=False)
# https://github.com/AlexsLemonade/refinebio/issues/1454
),
]
# Managers
objects = models.Manager()
failed_objects = FailedJobsManager()
hung_objects = HungJobsManager()
lost_objects = LostJobsManager()
# This field will contain an enumerated value specifying which
# processor pipeline was applied during the processor job.
pipeline_applied = models.CharField(max_length=256)
original_files = models.ManyToManyField(
"OriginalFile", through="ProcessorJobOriginalFileAssociation")
datasets = models.ManyToManyField("DataSet",
through="ProcessorJobDataSetAssociation")
no_retry = models.BooleanField(default=False)
abort = models.BooleanField(default=False)
# Resources
ram_amount = models.IntegerField(default=2048)
volume_index = models.CharField(max_length=3, null=True)
# Tracking
start_time = models.DateTimeField(null=True)
end_time = models.DateTimeField(null=True)
success = models.NullBooleanField(null=True)
nomad_job_id = models.CharField(max_length=256, null=True)
# This field represents how many times this job has been
# retried. It starts at 0 and each time the job has to be retried
# it will be incremented.
num_retries = models.IntegerField(default=0)
# This field indicates whether or not this job has been retried
# already or not.
retried = models.BooleanField(default=False)
# This point of this field is to identify which worker ran the
# job. A few fields may actually be required or something other
# than just an id.
worker_id = models.CharField(max_length=256, null=True)
# This field corresponds to the version number of the
# data_refinery_workers project that was used to run the job.
worker_version = models.CharField(max_length=128, null=True)
# This field allows jobs to specify why they failed.
failure_reason = models.TextField(null=True)
# If the job is retried, this is the id of the new job
retried_job = models.ForeignKey("self",
on_delete=models.SET_NULL,
null=True)
created_at = models.DateTimeField(editable=False, default=timezone.now)
last_modified = models.DateTimeField(default=timezone.now)
def get_samples(self) -> Set[Sample]:
samples = set()
for original_file in self.original_files.all():
for sample in original_file.samples.all():
samples.add(sample)
return samples
def kill_nomad_job(self) -> bool:
if not self.nomad_job_id:
return False
try:
nomad_host = get_env_variable("NOMAD_HOST")
nomad_port = get_env_variable("NOMAD_PORT", "4646")
nomad_client = Nomad(nomad_host, port=int(nomad_port), timeout=30)
nomad_client.job.deregister_job(self.nomad_job_id)
except nomad.api.exceptions.BaseNomadException:
return False
return True
def save(self, *args, **kwargs):
""" On save, update timestamps """
current_time = timezone.now()
if not self.id:
self.created_at = current_time
self.last_modified = current_time
return super(ProcessorJob, self).save(*args, **kwargs)
def __str__(self):
return "ProcessorJob " + str(self.pk) + ": " + str(
self.pipeline_applied)
class SurveyJob(models.Model):
"""Records information about a Surveyor Job."""
class Meta:
db_table = "survey_jobs"
# Managers
objects = models.Manager()
failed_objects = FailedJobsManager()
hung_objects = HungJobsManager()
lost_objects = LostJobsManager()
source_type = models.CharField(max_length=256)
success = models.NullBooleanField(null=True)
no_retry = models.BooleanField(default=False)
nomad_job_id = models.CharField(max_length=256, null=True)
ram_amount = models.IntegerField(default=256)
# The start time of the job
start_time = models.DateTimeField(null=True)
# The end time of the job
end_time = models.DateTimeField(null=True)
# This field represents how many times this job has been
# retried. It starts at 0 and each time the job has to be retried
# it will be incremented.
num_retries = models.IntegerField(default=0)
# This field indicates whether or not this job has been retried
# already or not.
retried = models.BooleanField(default=False)
# This field allows jobs to specify why they failed.
failure_reason = models.TextField(null=True)
created_at = models.DateTimeField(editable=False, default=timezone.now)
last_modified = models.DateTimeField(default=timezone.now)
def save(self, *args, **kwargs):
""" On save, update timestamps """
current_time = timezone.now()
if not self.id:
self.created_at = current_time
self.last_modified = current_time
return super(SurveyJob, self).save(*args, **kwargs)
def get_properties(self) -> Dict:
""" Return all associated SurveyJobKeyValues as a dict"""
return {pair.key: pair.value for pair in self.surveyjobkeyvalue_set.all()}
def get_accession_code(self):
""" Return `experiment_accession_code`, the most important code."""
try:
kvp = self.surveyjobkeyvalue_set.get(key="experiment_accession_code")
return kvp.value
except Exception:
return None
def kill_nomad_job(self) -> bool:
if not self.nomad_job_id:
return False
try:
nomad_host = get_env_variable("NOMAD_HOST")
nomad_port = get_env_variable("NOMAD_PORT", "4646")
nomad_client = Nomad(nomad_host, port=int(nomad_port), timeout=30)
nomad_client.job.deregister_job(self.nomad_job_id)
except nomad.api.exceptions.BaseNomadException:
return False
return True
def __str__(self):
return "SurveyJob " + str(self.pk) + ": " + str(self.source_type)
class ProcessorJob(models.Model):
"""Records information about running a processor."""
class Meta:
db_table = "processor_jobs"
indexes = [
models.Index(
fields=["created_at"],
name="processor_jobs_created_at",
# A partial index might be better here, given our queries we don't
# need to index the whole table. We need to update to Django 2.2
# for this to be supported.
# condition=Q(success=None, retried=False, no_retry=False)
# https://github.com/AlexsLemonade/refinebio/issues/1454
),
]
# Managers
objects = models.Manager()
failed_objects = FailedJobsManager()
hung_objects = HungJobsManager()
lost_objects = LostJobsManager()
unqueued_objects = UnqueuedJobsManager()
# This field will contain an enumerated value specifying which
# processor pipeline was applied during the processor job.
pipeline_applied = models.CharField(max_length=256)
original_files = models.ManyToManyField(
"OriginalFile", through="ProcessorJobOriginalFileAssociation"
)
datasets = models.ManyToManyField("DataSet", through="ProcessorJobDataSetAssociation")
no_retry = models.BooleanField(default=False)
abort = models.BooleanField(default=False)
# Resources
ram_amount = models.IntegerField(default=2048)
# The volume index is the instance id of an AWS EC2 machine. It looks like
# these are 19 characters, but just to be safe we'll make the max length a
# bit higher
volume_index = models.CharField(max_length=25, null=True)
# Which AWS Batch Job Queue the job was run in.
batch_job_queue = models.CharField(max_length=100, null=True)
# Tracking
start_time = models.DateTimeField(null=True)
end_time = models.DateTimeField(null=True)
success = models.BooleanField(null=True)
batch_job_id = models.CharField(max_length=256, null=True)
# This field represents how many times this job has been
# retried. It starts at 0 and each time the job has to be retried
# it will be incremented.
num_retries = models.IntegerField(default=0)
# This field indicates whether or not this job has been retried
# already or not.
retried = models.BooleanField(default=False)
# This point of this field is to identify which worker ran the
# job. A few fields may actually be required or something other
# than just an id.
worker_id = models.CharField(max_length=256, null=True)
# This field corresponds to the version number of the
# data_refinery_workers project that was used to run the job.
worker_version = models.CharField(max_length=128, null=True)
# This field allows jobs to specify why they failed.
failure_reason = models.TextField(null=True)
# If the job had data downloaded for it, this is the DownloaderJob that did so.
downloader_job = models.ForeignKey(
"data_refinery_common.DownloaderJob", on_delete=models.SET_NULL, null=True,
)
# If the job is retried, this is the id of the new job
retried_job = models.ForeignKey("self", on_delete=models.SET_NULL, null=True)
created_at = models.DateTimeField(editable=False, default=timezone.now)
last_modified = models.DateTimeField(default=timezone.now)
def get_samples(self) -> Set[Sample]:
samples = set()
for original_file in self.original_files.all():
for sample in original_file.samples.all():
samples.add(sample)
return samples
def save(self, *args, **kwargs):
""" On save, update timestamps """
current_time = timezone.now()
if not self.id:
self.created_at = current_time
self.last_modified = current_time
return super(ProcessorJob, self).save(*args, **kwargs)
def __str__(self):
return "ProcessorJob " + str(self.pk) + ": " + str(self.pipeline_applied)
class DownloaderJob(models.Model):
"""Records information about running a Downloader."""
class Meta:
db_table = "downloader_jobs"
indexes = [
models.Index(
fields=["created_at"],
name="downloader_jobs_created_at",
# condition=Q(success=None, retried=False, no_retry=False)
),
models.Index(fields=["worker_id"]),
]
# Managers
objects = models.Manager()
failed_objects = FailedJobsManager()
hung_objects = HungJobsManager()
lost_objects = LostJobsManager()
unqueued_objects = UnqueuedJobsManager()
# This field contains a string which corresponds to a valid
# Downloader Task. Valid values are enumerated in:
# data_refinery_common.enums.Downloaders
downloader_task = models.CharField(max_length=256)
accession_code = models.CharField(max_length=256, blank=True, null=True)
no_retry = models.BooleanField(default=False)
original_files = models.ManyToManyField(
"OriginalFile", through="DownloaderJobOriginalFileAssociation")
# Tracking
start_time = models.DateTimeField(null=True)
end_time = models.DateTimeField(null=True)
success = models.BooleanField(null=True)
batch_job_id = models.CharField(max_length=256, null=True)
# Resources
ram_amount = models.IntegerField(default=1024)
# The volume index is the instance id of an AWS EC2 machine. It looks like
# these are 19 characters, but just to be safe we'll make the max length a
# bit higher
# DEPRECATED
volume_index = models.CharField(max_length=25, null=True)
# Which AWS Batch Job Queue the job was run in.
batch_job_queue = models.CharField(max_length=100, null=True)
# This field represents how many times this job has been
# retried. It starts at 0 and each time the job has to be retried
# it will be incremented.
num_retries = models.IntegerField(default=0)
# This field indicates whether or not this job has been retried
# already or not.
retried = models.BooleanField(default=False)
# This point of this field is to identify which worker ran the
# job. A few fields may actually be required or something other
# than just an id.
worker_id = models.CharField(max_length=256, null=True)
# This field corresponds to the version number of the
# data_refinery_workers project that was used to run the job.
worker_version = models.CharField(max_length=128, null=True)
# This field allows jobs to specify why they failed.
failure_reason = models.TextField(null=True)
# If the job is retried, this is the id of the new job
retried_job = models.ForeignKey("self",
on_delete=models.PROTECT,
null=True)
# If the job was recreated because the data it downloaded got
# lost, deleted, or corrupted then this field will be true.
# This helps prevent an infinite loop of DownloaderJob recreation.
was_recreated = models.BooleanField(default=False)
created_at = models.DateTimeField(editable=False, default=timezone.now)
last_modified = models.DateTimeField(default=timezone.now)
def get_samples(self) -> Set[Sample]:
samples = set()
for original_file in self.original_files.all():
for sample in original_file.samples.all():
samples.add(sample)
return samples
def save(self, *args, **kwargs):
""" On save, update timestamps """
current_time = timezone.now()
if not self.id:
self.created_at = current_time
self.last_modified = current_time
return super(DownloaderJob, self).save(*args, **kwargs)
def __str__(self):
return "DownloaderJob " + str(self.pk) + ": " + str(
self.downloader_task)
class SurveyJob(models.Model):
"""Records information about a Surveyor Job."""
class Meta:
db_table = "survey_jobs"
# Managers
objects = models.Manager()
failed_objects = FailedJobsManager()
hung_objects = HungJobsManager()
lost_objects = LostJobsManager()
unqueued_objects = UnqueuedJobsManager()
source_type = models.CharField(max_length=256)
success = models.BooleanField(null=True)
no_retry = models.BooleanField(default=False)
batch_job_id = models.CharField(max_length=256, null=True)
# Which AWS Batch Job Queue the job was run in.
batch_job_queue = models.CharField(max_length=100, null=True)
ram_amount = models.IntegerField(default=1024)
# The start time of the job
start_time = models.DateTimeField(null=True)
# The end time of the job
end_time = models.DateTimeField(null=True)
# This field represents how many times this job has been
# retried. It starts at 0 and each time the job has to be retried
# it will be incremented.
num_retries = models.IntegerField(default=0)
# This field indicates whether or not this job has been retried
# already or not.
retried = models.BooleanField(default=False)
# If the job is retried, this is the id of the new job
retried_job = models.ForeignKey("self",
on_delete=models.SET_NULL,
null=True)
# This field allows jobs to specify why they failed.
failure_reason = models.TextField(null=True)
created_at = models.DateTimeField(editable=False, default=timezone.now)
last_modified = models.DateTimeField(default=timezone.now)
def save(self, *args, **kwargs):
""" On save, update timestamps """
current_time = timezone.now()
if not self.id:
self.created_at = current_time
self.last_modified = current_time
return super(SurveyJob, self).save(*args, **kwargs)
def get_properties(self) -> Dict:
""" Return all associated SurveyJobKeyValues as a dict"""
return {
pair.key: pair.value
for pair in self.surveyjobkeyvalue_set.all()
}
def get_accession_code(self):
""" Return `experiment_accession_code`, the most important code."""
try:
kvp = self.surveyjobkeyvalue_set.get(
key="experiment_accession_code")
return kvp.value
except Exception:
return None
def __str__(self):
return "SurveyJob " + str(self.pk) + ": " + str(self.source_type)