def test_create_scratch_uri(self):
# "walrus" bucket will be ignored; it doesn't start with "mrjob-"
self.add_mock_s3_data({'walrus': {}, 'zebra': {}})
runner = EMRJobRunner(conf_path=False, s3_sync_wait_time=0.01)
# bucket name should be mrjob- plus 16 random hex digits
s3_scratch_uri = runner._opts['s3_scratch_uri']
assert_equal(s3_scratch_uri[:11], 's3://mrjob-')
assert_equal(s3_scratch_uri[27:], '/tmp/')
# bucket shouldn't actually exist yet
scratch_bucket, _ = parse_s3_uri(s3_scratch_uri)
assert_not_in(scratch_bucket, self.mock_s3_fs.keys())
# need to do something to ensure that the bucket actually gets
# created. let's launch a (mock) job flow
jfid = runner.make_persistent_job_flow()
assert_in(scratch_bucket, self.mock_s3_fs.keys())
runner.make_emr_conn().terminate_jobflow(jfid)
# once our scratch bucket is created, we should re-use it
runner2 = EMRJobRunner(conf_path=False)
assert_equal(runner2._opts['s3_scratch_uri'], s3_scratch_uri)
s3_scratch_uri = runner._opts['s3_scratch_uri']
def test_create_scratch_uri(self):
# "walrus" bucket will be ignored; it doesn't start with "mrjob-"
self.add_mock_s3_data({'walrus': {}, 'zebra': {}})
runner = EMRJobRunner(conf_path=False, s3_sync_wait_time=0.01)
# bucket name should be mrjob- plus 16 random hex digits
s3_scratch_uri = runner._opts['s3_scratch_uri']
assert_equal(s3_scratch_uri[:11], 's3://mrjob-')
assert_equal(s3_scratch_uri[27:], '/tmp/')
# bucket shouldn't actually exist yet
scratch_bucket, _ = parse_s3_uri(s3_scratch_uri)
assert_not_in(scratch_bucket, self.mock_s3_fs.keys())
# need to do something to ensure that the bucket actually gets
# created. let's launch a (mock) job flow
jfid = runner.make_persistent_job_flow()
assert_in(scratch_bucket, self.mock_s3_fs.keys())
runner.make_emr_conn().terminate_jobflow(jfid)
# once our scratch bucket is created, we should re-use it
runner2 = EMRJobRunner(conf_path=False)
assert_equal(runner2._opts['s3_scratch_uri'], s3_scratch_uri)
s3_scratch_uri = runner._opts['s3_scratch_uri']
def s3_cleanup(glob_path, time_old, dry_run=False, conf_path=None):
"""Delete all files older than *time_old* in *path*.
If *dry_run* is ``True``, then just log the files that need to be
deleted without actually deleting them
"""
runner = EMRJobRunner(conf_path=conf_path)
s3_conn = runner.make_s3_conn()
log.info('Deleting all files in %s that are older than %s' % (glob_path, time_old))
for path in runner.ls(glob_path):
bucket_name, key_name = parse_s3_uri(path)
bucket = s3_conn.get_bucket(bucket_name)
for key in bucket.list(key_name):
last_modified = iso8601_to_datetime(key.last_modified)
age = datetime.utcnow() - last_modified
if age > time_old:
# Delete it
log.info('Deleting %s; is %s old' % (key.name, age))
if not dry_run:
key.delete()
def simulate_progress(self, jobflow_id, now=None):
"""Simulate progress on the given job flow. This is automatically
run when we call describe_jobflow().
:type jobflow_id: str
:param jobflow_id: fake job flow ID
:type now: py:class:`datetime.datetime`
:param now: alternate time to use as the current time (should be UTC)
"""
if now is None:
now = datetime.datetime.utcnow()
if self.simulation_steps_left <= 0:
raise AssertionError(
'Simulated progress too many times; bailing out')
self.simulation_steps_left -= 1
job_flow = self.mock_emr_job_flows[jobflow_id]
# if job is STARTING, move it along to WAITING
if job_flow.state == 'STARTING':
job_flow.state = 'WAITING'
job_flow.startdatetime = to_iso8601(now)
# if job is done, don't advance it
if job_flow.state in ('COMPLETED', 'TERMINATED', 'FAILED'):
return
# if SHUTTING_DOWN, finish shutting down
if job_flow.state == 'SHUTTING_DOWN':
if job_flow.reason == 'Shut down as step failed':
job_flow.state = 'FAILED'
else:
job_flow.state = 'TERMINATED'
job_flow.enddatetime = to_iso8601(now)
return
# if a step is currently running, advance it
for step_num, step in enumerate(job_flow.steps):
# skip steps that are already done
if step.state in ('COMPLETED', 'FAILED', 'CANCELLED'):
continue
if step.name in ('Setup Hadoop Debugging', ):
step.state = 'COMPLETED'
continue
# found currently running step! going to handle it, then exit
if step.state == 'PENDING':
step.state = 'RUNNING'
step.startdatetime = to_iso8601(now)
return
assert step.state == 'RUNNING'
step.enddatetime = to_iso8601(now)
# check if we're supposed to have an error
if (jobflow_id, step_num) in self.mock_emr_failures:
step.state = 'FAILED'
reason = self.mock_emr_failures[(jobflow_id, step_num)]
if reason:
job_flow.reason = reason
if step.actiononfailure == 'TERMINATE_JOB_FLOW':
job_flow.state = 'SHUTTING_DOWN'
if not reason:
job_flow.reason = 'Shut down as step failed'
return
step.state = 'COMPLETED'
# create fake output if we're supposed to write to S3
output_uri = self._get_step_output_uri(step)
if output_uri and S3_URI_RE.match(output_uri):
mock_output = self.mock_emr_output.get(
(jobflow_id, step_num)) or ['']
bucket_name, key_name = parse_s3_uri(output_uri)
# write output to S3
for i, bytes in enumerate(mock_output):
add_mock_s3_data(self.mock_s3_fs, {
bucket_name: {key_name + 'part-%05d' % i: bytes}})
elif (jobflow_id, step_num) in self.mock_emr_output:
raise AssertionError(
"can't use output for job flow ID %s, step %d "
"(it doesn't output to S3)" %
(jobflow_id, step_num))
# done!
return
# no pending steps. shut down job if appropriate
if job_flow.keepjobflowalivewhennosteps:
job_flow.state = 'WAITING'
job_flow.reason = 'Waiting for steps to run'
else:
job_flow.state = 'COMPLETED'
job_flow.reason = 'Steps Completed'
def simulate_progress(self, jobflow_id, now=None):
"""Simulate progress on the given job flow. This is automatically
run when we call describe_jobflow().
:type jobflow_id: str
:param jobflow_id: fake job flow ID
:type now: py:class:`datetime.datetime`
:param now: alternate time to use as the current time (should be UTC)
"""
if now is None:
now = datetime.datetime.utcnow()
if self.simulation_steps_left <= 0:
raise AssertionError(
'Simulated progress too many times; bailing out')
self.simulation_steps_left -= 1
job_flow = self.mock_emr_job_flows[jobflow_id]
# if job is STARTING, move it along to WAITING
if job_flow.state == 'STARTING':
job_flow.state = 'WAITING'
job_flow.startdatetime = to_iso8601(now)
# if job is done, don't advance it
if job_flow.state in ('COMPLETED', 'TERMINATED', 'FAILED'):
return
# if SHUTTING_DOWN, finish shutting down
if job_flow.state == 'SHUTTING_DOWN':
if job_flow.reason == 'Shut down as step failed':
job_flow.state = 'FAILED'
else:
job_flow.state = 'TERMINATED'
job_flow.enddatetime = to_iso8601(now)
return
# if a step is currently running, advance it
for step_num, step in enumerate(job_flow.steps):
# skip steps that are already done
if step.state in ('COMPLETED', 'FAILED', 'CANCELLED'):
continue
# found currently running step! going to handle it, then exit
if step.state == 'PENDING':
step.state = 'RUNNING'
step.startdatetime = to_iso8601(now)
return
assert step.state == 'RUNNING'
step.enddatetime = to_iso8601(now)
# check if we're supposed to have an error
if (jobflow_id, step_num) in self.mock_emr_failures:
step.state = 'FAILED'
reason = self.mock_emr_failures[(jobflow_id, step_num)]
if reason:
job_flow.reason = reason
if step.actiononfailure == 'TERMINATE_JOB_FLOW':
job_flow.state = 'SHUTTING_DOWN'
if not reason:
job_flow.reason = 'Shut down as step failed'
return
step.state = 'COMPLETED'
# create fake output if we're supposed to write to S3
output_uri = self._get_step_output_uri(step)
if output_uri and S3_URI_RE.match(output_uri):
mock_output = self.mock_emr_output.get(
(jobflow_id, step_num)) or ['']
bucket_name, key_name = parse_s3_uri(output_uri)
# write output to S3
for i, bytes in enumerate(mock_output):
add_mock_s3_data(
self.mock_s3_fs,
{bucket_name: {
key_name + 'part-%05d' % i: bytes
}})
elif (jobflow_id, step_num) in self.mock_emr_output:
raise AssertionError(
"can't use output for job flow ID %s, step %d "
"(it doesn't output to S3)" % (jobflow_id, step_num))
# done!
return
# no pending steps. shut down job if appropriate
if job_flow.keepjobalivewhennosteps:
job_flow.state = 'WAITING'
job_flow.reason = 'Waiting for steps to run'
else:
job_flow.state = 'COMPLETED'
job_flow.reason = 'Steps Completed'
