1 # Copyright (C) The Arvados Authors. All rights reserved.
3 # SPDX-License-Identifier: Apache-2.0
5 from __future__ import division
6 from builtins import next
7 from builtins import object
8 from builtins import str
9 from future.utils import viewvalues, viewitems
19 from functools import partial
22 from cwltool.errors import WorkflowException
23 import cwltool.workflow
24 from schema_salad.sourceline import SourceLine
25 import schema_salad.validate as validate
29 from arvados.keep import KeepClient
30 from arvados.errors import ApiError
32 import arvados_cwl.util
33 from .arvcontainer import RunnerContainer
34 from .arvjob import RunnerJob, RunnerTemplate
35 from .runner import Runner, upload_docker, upload_job_order, upload_workflow_deps
36 from .arvtool import ArvadosCommandTool, validate_cluster_target, ArvadosExpressionTool
37 from .arvworkflow import ArvadosWorkflow, upload_workflow
38 from .fsaccess import CollectionFsAccess, CollectionFetcher, collectionResolver, CollectionCache, pdh_size
39 from .perf import Perf
40 from .pathmapper import NoFollowPathMapper
41 from .task_queue import TaskQueue
42 from .context import ArvLoadingContext, ArvRuntimeContext
43 from ._version import __version__
45 from cwltool.process import shortname, UnsupportedRequirement, use_custom_schema
46 from cwltool.pathmapper import adjustFileObjs, adjustDirObjs, get_listing, visit_class
47 from cwltool.command_line_tool import compute_checksums
48 from cwltool.load_tool import load_tool
50 logger = logging.getLogger('arvados.cwl-runner')
51 metrics = logging.getLogger('arvados.cwl-runner.metrics')
53 DEFAULT_PRIORITY = 500
55 class RuntimeStatusLoggingHandler(logging.Handler):
57 Intercepts logging calls and report them as runtime statuses on runner
60 def __init__(self, runtime_status_update_func):
61 super(RuntimeStatusLoggingHandler, self).__init__()
62 self.runtime_status_update = runtime_status_update_func
63 self.updatingRuntimeStatus = False
65 def emit(self, record):
67 if record.levelno >= logging.ERROR:
69 elif record.levelno >= logging.WARNING:
71 if kind is not None and self.updatingRuntimeStatus is not True:
72 self.updatingRuntimeStatus = True
74 log_msg = record.getMessage()
76 # If the logged message is multi-line, use its first line as status
77 # and the rest as detail.
78 status, detail = log_msg.split('\n', 1)
79 self.runtime_status_update(
81 "%s: %s" % (record.name, status),
85 self.runtime_status_update(
87 "%s: %s" % (record.name, record.getMessage())
90 self.updatingRuntimeStatus = False
93 class ArvCwlExecutor(object):
94 """Execute a CWL tool or workflow, submit work (using either jobs or
95 containers API), wait for them to complete, and report output.
99 def __init__(self, api_client,
106 arvargs = argparse.Namespace()
107 arvargs.work_api = None
108 arvargs.output_name = None
109 arvargs.output_tags = None
110 arvargs.thread_count = 1
111 arvargs.collection_cache_size = None
113 self.api = api_client
115 self.workflow_eval_lock = threading.Condition(threading.RLock())
116 self.final_output = None
117 self.final_status = None
118 self.num_retries = num_retries
120 self.stop_polling = threading.Event()
123 self.final_output_collection = None
124 self.output_name = arvargs.output_name
125 self.output_tags = arvargs.output_tags
126 self.project_uuid = None
127 self.intermediate_output_ttl = 0
128 self.intermediate_output_collections = []
129 self.trash_intermediate = False
130 self.thread_count = arvargs.thread_count
131 self.poll_interval = 12
132 self.loadingContext = None
133 self.should_estimate_cache_size = True
134 self.fs_access = None
135 self.secret_store = None
137 if keep_client is not None:
138 self.keep_client = keep_client
140 self.keep_client = arvados.keep.KeepClient(api_client=self.api, num_retries=self.num_retries)
142 if arvargs.collection_cache_size:
143 collection_cache_size = arvargs.collection_cache_size*1024*1024
144 self.should_estimate_cache_size = False
146 collection_cache_size = 256*1024*1024
148 self.collection_cache = CollectionCache(self.api, self.keep_client, self.num_retries,
149 cap=collection_cache_size)
151 self.fetcher_constructor = partial(CollectionFetcher,
153 fs_access=CollectionFsAccess("", collection_cache=self.collection_cache),
154 num_retries=self.num_retries)
157 expected_api = ["containers", "jobs"]
158 for api in expected_api:
160 methods = self.api._rootDesc.get('resources')[api]['methods']
161 if ('httpMethod' in methods['create'] and
162 (arvargs.work_api == api or arvargs.work_api is None)):
168 if not self.work_api:
169 if arvargs.work_api is None:
170 raise Exception("No supported APIs")
172 raise Exception("Unsupported API '%s', expected one of %s" % (arvargs.work_api, expected_api))
174 if self.work_api == "jobs":
176 *******************************
177 Using the deprecated 'jobs' API.
179 To get rid of this warning:
181 Users: read about migrating at
182 http://doc.arvados.org/user/cwl/cwl-style.html#migrate
183 and use the option --api=containers
185 Admins: configure the cluster to disable the 'jobs' API as described at:
186 http://doc.arvados.org/install/install-api-server.html#disable_api_methods
187 *******************************""")
189 self.loadingContext = ArvLoadingContext(vars(arvargs))
190 self.loadingContext.fetcher_constructor = self.fetcher_constructor
191 self.loadingContext.resolver = partial(collectionResolver, self.api, num_retries=self.num_retries)
192 self.loadingContext.construct_tool_object = self.arv_make_tool
194 # Add a custom logging handler to the root logger for runtime status reporting
195 # if running inside a container
196 if arvados_cwl.util.get_current_container(self.api, self.num_retries, logger):
197 root_logger = logging.getLogger('')
199 # Remove existing RuntimeStatusLoggingHandlers if they exist
200 handlers = [h for h in root_logger.handlers if not isinstance(h, RuntimeStatusLoggingHandler)]
201 root_logger.handlers = handlers
203 handler = RuntimeStatusLoggingHandler(self.runtime_status_update)
204 root_logger.addHandler(handler)
206 self.runtimeContext = ArvRuntimeContext(vars(arvargs))
207 self.runtimeContext.make_fs_access = partial(CollectionFsAccess,
208 collection_cache=self.collection_cache)
210 validate_cluster_target(self, self.runtimeContext)
213 def arv_make_tool(self, toolpath_object, loadingContext):
214 if "class" in toolpath_object and toolpath_object["class"] == "CommandLineTool":
215 return ArvadosCommandTool(self, toolpath_object, loadingContext)
216 elif "class" in toolpath_object and toolpath_object["class"] == "Workflow":
217 return ArvadosWorkflow(self, toolpath_object, loadingContext)
218 elif "class" in toolpath_object and toolpath_object["class"] == "ExpressionTool":
219 return ArvadosExpressionTool(self, toolpath_object, loadingContext)
221 raise Exception("Unknown tool %s" % toolpath_object.get("class"))
223 def output_callback(self, out, processStatus):
224 with self.workflow_eval_lock:
225 if processStatus == "success":
226 logger.info("Overall process status is %s", processStatus)
229 logger.error("Overall process status is %s", processStatus)
232 self.api.pipeline_instances().update(uuid=self.pipeline["uuid"],
233 body={"state": state}).execute(num_retries=self.num_retries)
234 self.final_status = processStatus
235 self.final_output = out
236 self.workflow_eval_lock.notifyAll()
239 def start_run(self, runnable, runtimeContext):
240 self.task_queue.add(partial(runnable.run, runtimeContext),
241 self.workflow_eval_lock, self.stop_polling)
243 def process_submitted(self, container):
244 with self.workflow_eval_lock:
245 self.processes[container.uuid] = container
247 def process_done(self, uuid, record):
248 with self.workflow_eval_lock:
249 j = self.processes[uuid]
250 logger.info("%s %s is %s", self.label(j), uuid, record["state"])
251 self.task_queue.add(partial(j.done, record),
252 self.workflow_eval_lock, self.stop_polling)
253 del self.processes[uuid]
255 def runtime_status_update(self, kind, message, detail=None):
257 Updates the runtime_status field on the runner container.
258 Called when there's a need to report errors, warnings or just
259 activity statuses, for example in the RuntimeStatusLoggingHandler.
261 with self.workflow_eval_lock:
262 current = arvados_cwl.util.get_current_container(self.api, self.num_retries, logger)
265 runtime_status = current.get('runtime_status', {})
266 # In case of status being an error, only report the first one.
268 if not runtime_status.get('error'):
269 runtime_status.update({
272 if detail is not None:
273 runtime_status.update({
274 'errorDetail': detail
276 # Further errors are only mentioned as a count.
278 # Get anything before an optional 'and N more' string.
280 error_msg = re.match(
281 r'^(.*?)(?=\s*\(and \d+ more\)|$)', runtime_status.get('error')).groups()[0]
282 more_failures = re.match(
283 r'.*\(and (\d+) more\)', runtime_status.get('error'))
285 # Ignore tests stubbing errors
288 failure_qty = int(more_failures.groups()[0])
289 runtime_status.update({
290 'error': "%s (and %d more)" % (error_msg, failure_qty+1)
293 runtime_status.update({
294 'error': "%s (and 1 more)" % error_msg
296 elif kind in ['warning', 'activity']:
297 # Record the last warning/activity status without regard of
298 # previous occurences.
299 runtime_status.update({
302 if detail is not None:
303 runtime_status.update({
304 kind+"Detail": detail
307 # Ignore any other status kind
310 self.api.containers().update(uuid=current['uuid'],
312 'runtime_status': runtime_status,
313 }).execute(num_retries=self.num_retries)
314 except Exception as e:
315 logger.info("Couldn't update runtime_status: %s", e)
317 def wrapped_callback(self, cb, obj, st):
318 with self.workflow_eval_lock:
320 self.workflow_eval_lock.notifyAll()
322 def get_wrapped_callback(self, cb):
323 return partial(self.wrapped_callback, cb)
325 def on_message(self, event):
326 if event.get("object_uuid") in self.processes and event["event_type"] == "update":
327 uuid = event["object_uuid"]
328 if event["properties"]["new_attributes"]["state"] == "Running":
329 with self.workflow_eval_lock:
330 j = self.processes[uuid]
331 if j.running is False:
333 j.update_pipeline_component(event["properties"]["new_attributes"])
334 logger.info("%s %s is Running", self.label(j), uuid)
335 elif event["properties"]["new_attributes"]["state"] in ("Complete", "Failed", "Cancelled", "Final"):
336 self.process_done(uuid, event["properties"]["new_attributes"])
338 def label(self, obj):
339 return "[%s %s]" % (self.work_api[0:-1], obj.name)
341 def poll_states(self):
342 """Poll status of jobs or containers listed in the processes dict.
344 Runs in a separate thread.
348 remain_wait = self.poll_interval
351 self.stop_polling.wait(remain_wait)
352 if self.stop_polling.is_set():
354 with self.workflow_eval_lock:
355 keys = list(self.processes)
357 remain_wait = self.poll_interval
360 begin_poll = time.time()
361 if self.work_api == "containers":
362 table = self.poll_api.container_requests()
363 elif self.work_api == "jobs":
364 table = self.poll_api.jobs()
366 pageSize = self.poll_api._rootDesc.get('maxItemsPerResponse', 1000)
369 page = keys[:pageSize]
371 proc_states = table.list(filters=[["uuid", "in", page]]).execute(num_retries=self.num_retries)
373 logger.exception("Error checking states on API server: %s")
374 remain_wait = self.poll_interval
377 for p in proc_states["items"]:
379 "object_uuid": p["uuid"],
380 "event_type": "update",
385 keys = keys[pageSize:]
387 finish_poll = time.time()
388 remain_wait = self.poll_interval - (finish_poll - begin_poll)
390 logger.exception("Fatal error in state polling thread.")
391 with self.workflow_eval_lock:
392 self.processes.clear()
393 self.workflow_eval_lock.notifyAll()
395 self.stop_polling.set()
397 def add_intermediate_output(self, uuid):
399 self.intermediate_output_collections.append(uuid)
401 def trash_intermediate_output(self):
402 logger.info("Cleaning up intermediate output collections")
403 for i in self.intermediate_output_collections:
405 self.api.collections().delete(uuid=i).execute(num_retries=self.num_retries)
407 logger.warning("Failed to delete intermediate output: %s", sys.exc_info()[1], exc_info=(sys.exc_info()[1] if self.debug else False))
408 except (KeyboardInterrupt, SystemExit):
411 def check_features(self, obj, parentfield=""):
412 if isinstance(obj, dict):
413 if obj.get("writable") and self.work_api != "containers":
414 raise SourceLine(obj, "writable", UnsupportedRequirement).makeError("InitialWorkDir feature 'writable: true' not supported with --api=jobs")
415 if obj.get("class") == "DockerRequirement":
416 if obj.get("dockerOutputDirectory"):
417 if self.work_api != "containers":
418 raise SourceLine(obj, "dockerOutputDirectory", UnsupportedRequirement).makeError(
419 "Option 'dockerOutputDirectory' of DockerRequirement not supported with --api=jobs.")
420 if not obj.get("dockerOutputDirectory").startswith('/'):
421 raise SourceLine(obj, "dockerOutputDirectory", validate.ValidationException).makeError(
422 "Option 'dockerOutputDirectory' must be an absolute path.")
423 if obj.get("class") == "http://commonwl.org/cwltool#Secrets" and self.work_api != "containers":
424 raise SourceLine(obj, "class", UnsupportedRequirement).makeError("Secrets not supported with --api=jobs")
425 if obj.get("class") == "InplaceUpdateRequirement":
426 if obj["inplaceUpdate"] and parentfield == "requirements":
427 raise SourceLine(obj, "class", UnsupportedRequirement).makeError("InplaceUpdateRequirement not supported for keep collections.")
428 for k,v in viewitems(obj):
429 self.check_features(v, parentfield=k)
430 elif isinstance(obj, list):
431 for i,v in enumerate(obj):
432 with SourceLine(obj, i, UnsupportedRequirement, logger.isEnabledFor(logging.DEBUG)):
433 self.check_features(v, parentfield=parentfield)
435 def make_output_collection(self, name, storage_classes, tagsString, outputObj):
436 outputObj = copy.deepcopy(outputObj)
439 def capture(fileobj):
440 files.append(fileobj)
442 adjustDirObjs(outputObj, capture)
443 adjustFileObjs(outputObj, capture)
445 generatemapper = NoFollowPathMapper(files, "", "", separateDirs=False)
447 final = arvados.collection.Collection(api_client=self.api,
448 keep_client=self.keep_client,
449 num_retries=self.num_retries)
451 for k,v in generatemapper.items():
452 if v.type == "Directory" and v.resolved.startswith("_:"):
454 if v.type == "CreateFile" and (k.startswith("_:") or v.resolved.startswith("_:")):
455 with final.open(v.target, "wb") as f:
456 f.write(v.resolved.encode("utf-8"))
459 if not v.resolved.startswith("keep:"):
460 raise Exception("Output source is not in keep or a literal")
461 sp = v.resolved.split("/")
462 srccollection = sp[0][5:]
464 reader = self.collection_cache.get(srccollection)
465 srcpath = "/".join(sp[1:]) if len(sp) > 1 else "."
466 final.copy(srcpath, v.target, source_collection=reader, overwrite=False)
467 except arvados.errors.ArgumentError as e:
468 logger.error("Creating CollectionReader for '%s' '%s': %s", k, v, e)
471 logger.error("While preparing output collection: %s", e)
474 def rewrite(fileobj):
475 fileobj["location"] = generatemapper.mapper(fileobj["location"]).target
476 for k in ("listing", "contents", "nameext", "nameroot", "dirname"):
480 adjustDirObjs(outputObj, rewrite)
481 adjustFileObjs(outputObj, rewrite)
483 with final.open("cwl.output.json", "w") as f:
484 res = str(json.dumps(outputObj, sort_keys=True, indent=4, separators=(',',': '), ensure_ascii=False))
487 final.save_new(name=name, owner_uuid=self.project_uuid, storage_classes=storage_classes, ensure_unique_name=True)
489 logger.info("Final output collection %s \"%s\" (%s)", final.portable_data_hash(),
490 final.api_response()["name"],
491 final.manifest_locator())
493 final_uuid = final.manifest_locator()
494 tags = tagsString.split(',')
496 self.api.links().create(body={
497 "head_uuid": final_uuid, "link_class": "tag", "name": tag
498 }).execute(num_retries=self.num_retries)
500 def finalcollection(fileobj):
501 fileobj["location"] = "keep:%s/%s" % (final.portable_data_hash(), fileobj["location"])
503 adjustDirObjs(outputObj, finalcollection)
504 adjustFileObjs(outputObj, finalcollection)
506 return (outputObj, final)
508 def set_crunch_output(self):
509 if self.work_api == "containers":
510 current = arvados_cwl.util.get_current_container(self.api, self.num_retries, logger)
514 self.api.containers().update(uuid=current['uuid'],
516 'output': self.final_output_collection.portable_data_hash(),
517 }).execute(num_retries=self.num_retries)
518 self.api.collections().update(uuid=self.final_output_collection.manifest_locator(),
521 }).execute(num_retries=self.num_retries)
523 logger.exception("Setting container output")
525 elif self.work_api == "jobs" and "TASK_UUID" in os.environ:
526 self.api.job_tasks().update(uuid=os.environ["TASK_UUID"],
528 'output': self.final_output_collection.portable_data_hash(),
529 'success': self.final_status == "success",
531 }).execute(num_retries=self.num_retries)
533 def apply_reqs(self, job_order_object, tool):
534 if "https://w3id.org/cwl/cwl#requirements" in job_order_object:
535 if tool.metadata.get("http://commonwl.org/cwltool#original_cwlVersion") == 'v1.0':
536 raise WorkflowException(
537 "`cwl:requirements` in the input object is not part of CWL "
538 "v1.0. You can adjust to use `cwltool:overrides` instead; or you "
539 "can set the cwlVersion to v1.1 or greater and re-run with "
541 job_reqs = job_order_object["https://w3id.org/cwl/cwl#requirements"]
543 tool.requirements.append(req)
545 def arv_executor(self, tool, job_order, runtimeContext, logger=None):
546 self.debug = runtimeContext.debug
548 tool.visit(self.check_features)
550 self.project_uuid = runtimeContext.project_uuid
552 self.fs_access = runtimeContext.make_fs_access(runtimeContext.basedir)
553 self.secret_store = runtimeContext.secret_store
555 self.trash_intermediate = runtimeContext.trash_intermediate
556 if self.trash_intermediate and self.work_api != "containers":
557 raise Exception("--trash-intermediate is only supported with --api=containers.")
559 self.intermediate_output_ttl = runtimeContext.intermediate_output_ttl
560 if self.intermediate_output_ttl and self.work_api != "containers":
561 raise Exception("--intermediate-output-ttl is only supported with --api=containers.")
562 if self.intermediate_output_ttl < 0:
563 raise Exception("Invalid value %d for --intermediate-output-ttl, cannot be less than zero" % self.intermediate_output_ttl)
565 if runtimeContext.submit_request_uuid and self.work_api != "containers":
566 raise Exception("--submit-request-uuid requires containers API, but using '{}' api".format(self.work_api))
568 if not runtimeContext.name:
569 runtimeContext.name = self.name = tool.tool.get("label") or tool.metadata.get("label") or os.path.basename(tool.tool["id"])
571 # Upload local file references in the job order.
572 job_order = upload_job_order(self, "%s input" % runtimeContext.name,
575 submitting = (runtimeContext.update_workflow or
576 runtimeContext.create_workflow or
577 (runtimeContext.submit and not
578 (tool.tool["class"] == "CommandLineTool" and
579 runtimeContext.wait and
580 not runtimeContext.always_submit_runner)))
582 loadingContext = self.loadingContext.copy()
583 loadingContext.do_validate = False
584 loadingContext.do_update = False
586 # Document may have been auto-updated. Reload the original
587 # document with updating disabled because we want to
588 # submit the original document, not the auto-updated one.
589 tool = load_tool(tool.tool["id"], loadingContext)
591 # Upload direct dependencies of workflow steps, get back mapping of files to keep references.
592 # Also uploads docker images.
593 merged_map = upload_workflow_deps(self, tool)
595 # Recreate process object (ArvadosWorkflow or
596 # ArvadosCommandTool) because tool document may have been
597 # updated by upload_workflow_deps in ways that modify
598 # inheritance of hints or requirements.
599 loadingContext.loader = tool.doc_loader
600 loadingContext.avsc_names = tool.doc_schema
601 loadingContext.metadata = tool.metadata
602 tool = load_tool(tool.tool, loadingContext)
604 existing_uuid = runtimeContext.update_workflow
605 if existing_uuid or runtimeContext.create_workflow:
606 # Create a pipeline template or workflow record and exit.
607 if self.work_api == "jobs":
608 tmpl = RunnerTemplate(self, tool, job_order,
609 runtimeContext.enable_reuse,
611 submit_runner_ram=runtimeContext.submit_runner_ram,
612 name=runtimeContext.name,
613 merged_map=merged_map,
614 loadingContext=loadingContext)
616 # cwltool.main will write our return value to stdout.
617 return (tmpl.uuid, "success")
618 elif self.work_api == "containers":
619 return (upload_workflow(self, tool, job_order,
622 submit_runner_ram=runtimeContext.submit_runner_ram,
623 name=runtimeContext.name,
624 merged_map=merged_map),
627 self.apply_reqs(job_order, tool)
629 self.ignore_docker_for_reuse = runtimeContext.ignore_docker_for_reuse
630 self.eval_timeout = runtimeContext.eval_timeout
632 runtimeContext = runtimeContext.copy()
633 runtimeContext.use_container = True
634 runtimeContext.tmpdir_prefix = "tmp"
635 runtimeContext.work_api = self.work_api
637 if self.work_api == "containers":
638 if self.ignore_docker_for_reuse:
639 raise Exception("--ignore-docker-for-reuse not supported with containers API.")
640 runtimeContext.outdir = "/var/spool/cwl"
641 runtimeContext.docker_outdir = "/var/spool/cwl"
642 runtimeContext.tmpdir = "/tmp"
643 runtimeContext.docker_tmpdir = "/tmp"
644 elif self.work_api == "jobs":
645 if runtimeContext.priority != DEFAULT_PRIORITY:
646 raise Exception("--priority not implemented for jobs API.")
647 runtimeContext.outdir = "$(task.outdir)"
648 runtimeContext.docker_outdir = "$(task.outdir)"
649 runtimeContext.tmpdir = "$(task.tmpdir)"
651 if runtimeContext.priority < 1 or runtimeContext.priority > 1000:
652 raise Exception("--priority must be in the range 1..1000.")
654 if self.should_estimate_cache_size:
657 def estimate_collection_cache(obj):
658 if obj.get("location", "").startswith("keep:"):
659 m = pdh_size.match(obj["location"][5:])
660 if m and m.group(1) not in visited:
661 visited.add(m.group(1))
662 estimated_size[0] += int(m.group(2))
663 visit_class(job_order, ("File", "Directory"), estimate_collection_cache)
664 runtimeContext.collection_cache_size = max(((estimated_size[0]*192) // (1024*1024))+1, 256)
665 self.collection_cache.set_cap(runtimeContext.collection_cache_size*1024*1024)
667 logger.info("Using collection cache size %s MiB", runtimeContext.collection_cache_size)
670 if runtimeContext.submit:
671 # Submit a runner job to run the workflow for us.
672 if self.work_api == "containers":
673 if tool.tool["class"] == "CommandLineTool" and runtimeContext.wait and (not runtimeContext.always_submit_runner):
674 runtimeContext.runnerjob = tool.tool["id"]
676 tool = RunnerContainer(self, tool, loadingContext, runtimeContext.enable_reuse,
679 submit_runner_ram=runtimeContext.submit_runner_ram,
680 name=runtimeContext.name,
681 on_error=runtimeContext.on_error,
682 submit_runner_image=runtimeContext.submit_runner_image,
683 intermediate_output_ttl=runtimeContext.intermediate_output_ttl,
684 merged_map=merged_map,
685 priority=runtimeContext.priority,
686 secret_store=self.secret_store,
687 collection_cache_size=runtimeContext.collection_cache_size,
688 collection_cache_is_default=self.should_estimate_cache_size)
689 elif self.work_api == "jobs":
690 tool = RunnerJob(self, tool, loadingContext, runtimeContext.enable_reuse,
693 submit_runner_ram=runtimeContext.submit_runner_ram,
694 name=runtimeContext.name,
695 on_error=runtimeContext.on_error,
696 submit_runner_image=runtimeContext.submit_runner_image,
697 merged_map=merged_map)
698 elif runtimeContext.cwl_runner_job is None and self.work_api == "jobs":
699 # Create pipeline for local run
700 self.pipeline = self.api.pipeline_instances().create(
702 "owner_uuid": self.project_uuid,
703 "name": runtimeContext.name if runtimeContext.name else shortname(tool.tool["id"]),
705 "state": "RunningOnClient"}).execute(num_retries=self.num_retries)
706 logger.info("Pipeline instance %s", self.pipeline["uuid"])
708 if runtimeContext.cwl_runner_job is not None:
709 self.uuid = runtimeContext.cwl_runner_job.get('uuid')
711 jobiter = tool.job(job_order,
712 self.output_callback,
715 if runtimeContext.submit and not runtimeContext.wait:
716 runnerjob = next(jobiter)
717 runnerjob.run(runtimeContext)
718 return (runnerjob.uuid, "success")
720 current_container = arvados_cwl.util.get_current_container(self.api, self.num_retries, logger)
721 if current_container:
722 logger.info("Running inside container %s", current_container.get("uuid"))
724 self.poll_api = arvados.api('v1', timeout=runtimeContext.http_timeout)
725 self.polling_thread = threading.Thread(target=self.poll_states)
726 self.polling_thread.start()
728 self.task_queue = TaskQueue(self.workflow_eval_lock, self.thread_count)
731 self.workflow_eval_lock.acquire()
733 # Holds the lock while this code runs and releases it when
734 # it is safe to do so in self.workflow_eval_lock.wait(),
735 # at which point on_message can update job state and
736 # process output callbacks.
738 loopperf = Perf(metrics, "jobiter")
740 for runnable in jobiter:
743 if self.stop_polling.is_set():
746 if self.task_queue.error is not None:
747 raise self.task_queue.error
750 with Perf(metrics, "run"):
751 self.start_run(runnable, runtimeContext)
753 if (self.task_queue.in_flight + len(self.processes)) > 0:
754 self.workflow_eval_lock.wait(3)
756 logger.error("Workflow is deadlocked, no runnable processes and not waiting on any pending processes.")
759 if self.stop_polling.is_set():
765 while (self.task_queue.in_flight + len(self.processes)) > 0:
766 if self.task_queue.error is not None:
767 raise self.task_queue.error
768 self.workflow_eval_lock.wait(3)
770 except UnsupportedRequirement:
773 if sys.exc_info()[0] is KeyboardInterrupt or sys.exc_info()[0] is SystemExit:
774 logger.error("Interrupted, workflow will be cancelled")
775 elif isinstance(sys.exc_info()[1], WorkflowException):
776 logger.error("Workflow execution failed:\n%s", sys.exc_info()[1], exc_info=(sys.exc_info()[1] if self.debug else False))
778 logger.exception("Workflow execution failed")
781 self.api.pipeline_instances().update(uuid=self.pipeline["uuid"],
782 body={"state": "Failed"}).execute(num_retries=self.num_retries)
784 if self.work_api == "containers" and not current_container:
785 # Not running in a crunch container, so cancel any outstanding processes.
786 for p in self.processes:
788 self.api.container_requests().update(uuid=p,
789 body={"priority": "0"}
790 ).execute(num_retries=self.num_retries)
794 self.workflow_eval_lock.release()
795 self.task_queue.drain()
796 self.stop_polling.set()
797 self.polling_thread.join()
798 self.task_queue.join()
800 if self.final_status == "UnsupportedRequirement":
801 raise UnsupportedRequirement("Check log for details.")
803 if self.final_output is None:
804 raise WorkflowException("Workflow did not return a result.")
806 if runtimeContext.submit and isinstance(tool, Runner):
807 logger.info("Final output collection %s", tool.final_output)
809 if self.output_name is None:
810 self.output_name = "Output of %s" % (shortname(tool.tool["id"]))
811 if self.output_tags is None:
812 self.output_tags = ""
814 storage_classes = runtimeContext.storage_classes.strip().split(",")
815 self.final_output, self.final_output_collection = self.make_output_collection(self.output_name, storage_classes, self.output_tags, self.final_output)
816 self.set_crunch_output()
818 if runtimeContext.compute_checksum:
819 adjustDirObjs(self.final_output, partial(get_listing, self.fs_access))
820 adjustFileObjs(self.final_output, partial(compute_checksums, self.fs_access))
822 if self.trash_intermediate and self.final_status == "success":
823 self.trash_intermediate_output()
825 return (self.final_output, self.final_status)