1 # Copyright (C) The Arvados Authors. All rights reserved.
3 # SPDX-License-Identifier: Apache-2.0
5 from future import standard_library
6 standard_library.install_aliases()
7 from builtins import str
12 import urllib.request, urllib.parse, urllib.error
19 import arvados_cwl.util
20 import ruamel.yaml as yaml
22 from cwltool.errors import WorkflowException
23 from cwltool.process import UnsupportedRequirement, shortname
24 from cwltool.utils import aslist, adjustFileObjs, adjustDirObjs, visit_class
25 from cwltool.job import JobBase
27 import arvados.collection
29 from .arvdocker import arv_docker_get_image
31 from .runner import Runner, arvados_jobs_image, packed_workflow, trim_anonymous_location, remove_redundant_fields
32 from .fsaccess import CollectionFetcher
33 from .pathmapper import NoFollowPathMapper, trim_listing
34 from .perf import Perf
35 from ._version import __version__
37 logger = logging.getLogger('arvados.cwl-runner')
38 metrics = logging.getLogger('arvados.cwl-runner.metrics')
40 class ArvadosContainer(JobBase):
41 """Submit and manage a Crunch container request for executing a CWL CommandLineTool."""
43 def __init__(self, runner, job_runtime,
44 builder, # type: Builder
45 joborder, # type: Dict[Text, Union[Dict[Text, Any], List, Text]]
46 make_path_mapper, # type: Callable[..., PathMapper]
47 requirements, # type: List[Dict[Text, Text]]
48 hints, # type: List[Dict[Text, Text]]
51 super(ArvadosContainer, self).__init__(builder, joborder, make_path_mapper, requirements, hints, name)
52 self.arvrunner = runner
53 self.job_runtime = job_runtime
57 def update_pipeline_component(self, r):
60 def run(self, runtimeContext):
61 # ArvadosCommandTool subclasses from cwltool.CommandLineTool,
62 # which calls makeJobRunner() to get a new ArvadosContainer
63 # object. The fields that define execution such as
64 # command_line, environment, etc are set on the
65 # ArvadosContainer object by CommandLineTool.job() before
68 runtimeContext = self.job_runtime
71 "command": self.command_line,
73 "output_path": self.outdir,
75 "priority": runtimeContext.priority,
79 runtime_constraints = {}
81 if runtimeContext.project_uuid:
82 container_request["owner_uuid"] = runtimeContext.project_uuid
84 if self.arvrunner.secret_store.has_secret(self.command_line):
85 raise WorkflowException("Secret material leaked on command line, only file literals may contain secrets")
87 if self.arvrunner.secret_store.has_secret(self.environment):
88 raise WorkflowException("Secret material leaked in environment, only file literals may contain secrets")
90 resources = self.builder.resources
91 if resources is not None:
92 runtime_constraints["vcpus"] = math.ceil(resources.get("cores", 1))
93 runtime_constraints["ram"] = math.ceil(resources.get("ram") * 2**20)
98 "capacity": math.ceil(resources.get("outdirSize", 0) * 2**20)
102 "capacity": math.ceil(resources.get("tmpdirSize", 0) * 2**20)
106 scheduling_parameters = {}
108 rf = [self.pathmapper.mapper(f) for f in self.pathmapper.referenced_files]
109 rf.sort(key=lambda k: k.resolved)
111 for resolved, target, tp, stg in rf:
114 if prevdir and target.startswith(prevdir):
116 if tp == "Directory":
119 targetdir = os.path.dirname(target)
120 sp = resolved.split("/", 1)
121 pdh = sp[0][5:] # remove "keep:"
122 mounts[targetdir] = {
123 "kind": "collection",
124 "portable_data_hash": pdh
126 if pdh in self.pathmapper.pdh_to_uuid:
127 mounts[targetdir]["uuid"] = self.pathmapper.pdh_to_uuid[pdh]
129 if tp == "Directory":
132 path = os.path.dirname(sp[1])
133 if path and path != "/":
134 mounts[targetdir]["path"] = path
135 prevdir = targetdir + "/"
137 with Perf(metrics, "generatefiles %s" % self.name):
138 if self.generatefiles["listing"]:
139 vwd = arvados.collection.Collection(api_client=self.arvrunner.api,
140 keep_client=self.arvrunner.keep_client,
141 num_retries=self.arvrunner.num_retries)
142 generatemapper = NoFollowPathMapper(self.generatefiles["listing"], "", "",
145 sorteditems = sorted(generatemapper.items(), key=lambda n: n[1].target)
147 logger.debug("generatemapper is %s", sorteditems)
149 with Perf(metrics, "createfiles %s" % self.name):
150 for f, p in sorteditems:
154 if p.target.startswith("/"):
155 dst = p.target[len(self.outdir)+1:] if p.target.startswith(self.outdir+"/") else p.target[1:]
159 if p.type in ("File", "Directory", "WritableFile", "WritableDirectory"):
160 if p.resolved.startswith("_:"):
163 source, path = self.arvrunner.fs_access.get_collection(p.resolved)
164 vwd.copy(path or ".", dst, source_collection=source)
165 elif p.type == "CreateFile":
166 if self.arvrunner.secret_store.has_secret(p.resolved):
167 mountpoint = p.target if p.target.startswith("/") else os.path.join(self.outdir, p.target)
168 secret_mounts[mountpoint] = {
170 "content": self.arvrunner.secret_store.retrieve(p.resolved)
173 with vwd.open(dst, "w") as n:
176 def keepemptydirs(p):
177 if isinstance(p, arvados.collection.RichCollectionBase):
179 p.open(".keep", "w").close()
186 if not runtimeContext.current_container:
187 runtimeContext.current_container = arvados_cwl.util.get_current_container(self.arvrunner.api, self.arvrunner.num_retries, logger)
188 info = arvados_cwl.util.get_intermediate_collection_info(self.name, runtimeContext.current_container, runtimeContext.intermediate_output_ttl)
189 vwd.save_new(name=info["name"],
190 owner_uuid=runtimeContext.project_uuid,
191 ensure_unique_name=True,
192 trash_at=info["trash_at"],
193 properties=info["properties"])
196 for f, p in sorteditems:
197 if (not p.target or self.arvrunner.secret_store.has_secret(p.resolved) or
198 (prev is not None and p.target.startswith(prev))):
200 if p.target.startswith("/"):
201 dst = p.target[len(self.outdir)+1:] if p.target.startswith(self.outdir+"/") else p.target[1:]
204 mountpoint = p.target if p.target.startswith("/") else os.path.join(self.outdir, p.target)
205 mounts[mountpoint] = {"kind": "collection",
206 "portable_data_hash": vwd.portable_data_hash(),
208 if p.type.startswith("Writable"):
209 mounts[mountpoint]["writable"] = True
210 prev = p.target + "/"
212 container_request["environment"] = {"TMPDIR": self.tmpdir, "HOME": self.outdir}
214 container_request["environment"].update(self.environment)
217 sp = self.stdin[6:].split("/", 1)
218 mounts["stdin"] = {"kind": "collection",
219 "portable_data_hash": sp[0],
223 mounts["stderr"] = {"kind": "file",
224 "path": "%s/%s" % (self.outdir, self.stderr)}
227 mounts["stdout"] = {"kind": "file",
228 "path": "%s/%s" % (self.outdir, self.stdout)}
230 (docker_req, docker_is_req) = self.get_requirement("DockerRequirement")
232 docker_req = {"dockerImageId": "arvados/jobs:"+__version__}
234 container_request["container_image"] = arv_docker_get_image(self.arvrunner.api,
236 runtimeContext.pull_image,
237 runtimeContext.project_uuid,
238 runtimeContext.force_docker_pull,
239 runtimeContext.tmp_outdir_prefix)
241 network_req, _ = self.get_requirement("NetworkAccess")
243 runtime_constraints["API"] = network_req["networkAccess"]
245 api_req, _ = self.get_requirement("http://arvados.org/cwl#APIRequirement")
247 runtime_constraints["API"] = True
249 runtime_req, _ = self.get_requirement("http://arvados.org/cwl#RuntimeConstraints")
251 if "keep_cache" in runtime_req:
252 runtime_constraints["keep_cache_ram"] = math.ceil(runtime_req["keep_cache"] * 2**20)
253 if "outputDirType" in runtime_req:
254 if runtime_req["outputDirType"] == "local_output_dir":
255 # Currently the default behavior.
257 elif runtime_req["outputDirType"] == "keep_output_dir":
258 mounts[self.outdir]= {
259 "kind": "collection",
263 partition_req, _ = self.get_requirement("http://arvados.org/cwl#PartitionRequirement")
265 scheduling_parameters["partitions"] = aslist(partition_req["partition"])
267 intermediate_output_req, _ = self.get_requirement("http://arvados.org/cwl#IntermediateOutput")
268 if intermediate_output_req:
269 self.output_ttl = intermediate_output_req["outputTTL"]
271 self.output_ttl = self.arvrunner.intermediate_output_ttl
273 if self.output_ttl < 0:
274 raise WorkflowException("Invalid value %d for output_ttl, cannot be less than zero" % container_request["output_ttl"])
276 storage_class_req, _ = self.get_requirement("http://arvados.org/cwl#OutputStorageClass")
277 if storage_class_req and storage_class_req.get("intermediateStorageClass"):
278 container_request["output_storage_classes"] = aslist(storage_class_req["intermediateStorageClass"])
280 container_request["output_storage_classes"] = runtimeContext.intermediate_storage_classes.strip().split(",")
282 if self.timelimit is not None and self.timelimit > 0:
283 scheduling_parameters["max_run_time"] = self.timelimit
285 extra_submit_params = {}
286 if runtimeContext.submit_runner_cluster:
287 extra_submit_params["cluster_id"] = runtimeContext.submit_runner_cluster
289 container_request["output_name"] = "Output for step %s" % (self.name)
290 container_request["output_ttl"] = self.output_ttl
291 container_request["mounts"] = mounts
292 container_request["secret_mounts"] = secret_mounts
293 container_request["runtime_constraints"] = runtime_constraints
294 container_request["scheduling_parameters"] = scheduling_parameters
296 enable_reuse = runtimeContext.enable_reuse
298 reuse_req, _ = self.get_requirement("WorkReuse")
300 enable_reuse = reuse_req["enableReuse"]
301 reuse_req, _ = self.get_requirement("http://arvados.org/cwl#ReuseRequirement")
303 enable_reuse = reuse_req["enableReuse"]
304 container_request["use_existing"] = enable_reuse
306 if runtimeContext.runnerjob.startswith("arvwf:"):
307 wfuuid = runtimeContext.runnerjob[6:runtimeContext.runnerjob.index("#")]
308 wfrecord = self.arvrunner.api.workflows().get(uuid=wfuuid).execute(num_retries=self.arvrunner.num_retries)
309 if container_request["name"] == "main":
310 container_request["name"] = wfrecord["name"]
311 container_request["properties"]["template_uuid"] = wfuuid
313 self.output_callback = self.arvrunner.get_wrapped_callback(self.output_callback)
316 if runtimeContext.submit_request_uuid:
317 response = self.arvrunner.api.container_requests().update(
318 uuid=runtimeContext.submit_request_uuid,
319 body=container_request,
320 **extra_submit_params
321 ).execute(num_retries=self.arvrunner.num_retries)
323 response = self.arvrunner.api.container_requests().create(
324 body=container_request,
325 **extra_submit_params
326 ).execute(num_retries=self.arvrunner.num_retries)
328 self.uuid = response["uuid"]
329 self.arvrunner.process_submitted(self)
331 if response["state"] == "Final":
332 logger.info("%s reused container %s", self.arvrunner.label(self), response["container_uuid"])
334 logger.info("%s %s state is %s", self.arvrunner.label(self), response["uuid"], response["state"])
335 except Exception as e:
336 logger.exception("%s error submitting container\n%s", self.arvrunner.label(self), e)
337 logger.debug("Container request was %s", container_request)
338 self.output_callback({}, "permanentFail")
340 def done(self, record):
343 container = self.arvrunner.api.containers().get(
344 uuid=record["container_uuid"]
345 ).execute(num_retries=self.arvrunner.num_retries)
346 if container["state"] == "Complete":
347 rcode = container["exit_code"]
348 if self.successCodes and rcode in self.successCodes:
349 processStatus = "success"
350 elif self.temporaryFailCodes and rcode in self.temporaryFailCodes:
351 processStatus = "temporaryFail"
352 elif self.permanentFailCodes and rcode in self.permanentFailCodes:
353 processStatus = "permanentFail"
355 processStatus = "success"
357 processStatus = "permanentFail"
359 processStatus = "permanentFail"
361 if processStatus == "permanentFail" and record["log_uuid"]:
362 logc = arvados.collection.CollectionReader(record["log_uuid"],
363 api_client=self.arvrunner.api,
364 keep_client=self.arvrunner.keep_client,
365 num_retries=self.arvrunner.num_retries)
366 label = self.arvrunner.label(self)
369 "%s (%s) error log:" % (label, record["uuid"]), maxlen=40)
371 if record["output_uuid"]:
372 if self.arvrunner.trash_intermediate or self.arvrunner.intermediate_output_ttl:
373 # Compute the trash time to avoid requesting the collection record.
374 trash_at = ciso8601.parse_datetime_as_naive(record["modified_at"]) + datetime.timedelta(0, self.arvrunner.intermediate_output_ttl)
375 aftertime = " at %s" % trash_at.strftime("%Y-%m-%d %H:%M:%S UTC") if self.arvrunner.intermediate_output_ttl else ""
376 orpart = ", or" if self.arvrunner.trash_intermediate and self.arvrunner.intermediate_output_ttl else ""
377 oncomplete = " upon successful completion of the workflow" if self.arvrunner.trash_intermediate else ""
378 logger.info("%s Intermediate output %s (%s) will be trashed%s%s%s." % (
379 self.arvrunner.label(self), record["output_uuid"], container["output"], aftertime, orpart, oncomplete))
380 self.arvrunner.add_intermediate_output(record["output_uuid"])
382 if container["output"]:
383 outputs = done.done_outputs(self, container, "/tmp", self.outdir, "/keep")
384 except WorkflowException as e:
385 # Only include a stack trace if in debug mode.
386 # A stack trace may obfuscate more useful output about the workflow.
387 logger.error("%s unable to collect output from %s:\n%s",
388 self.arvrunner.label(self), container["output"], e, exc_info=(e if self.arvrunner.debug else False))
389 processStatus = "permanentFail"
391 logger.exception("%s while getting output object:", self.arvrunner.label(self))
392 processStatus = "permanentFail"
394 self.output_callback(outputs, processStatus)
397 class RunnerContainer(Runner):
398 """Submit and manage a container that runs arvados-cwl-runner."""
400 def arvados_job_spec(self, runtimeContext):
401 """Create an Arvados container request for this workflow.
403 The returned dict can be used to create a container passed as
404 the +body+ argument to container_requests().create().
407 adjustDirObjs(self.job_order, trim_listing)
408 visit_class(self.job_order, ("File", "Directory"), trim_anonymous_location)
409 visit_class(self.job_order, ("File", "Directory"), remove_redundant_fields)
412 for param in sorted(self.job_order.keys()):
413 if self.secret_store.has_secret(self.job_order[param]):
414 mnt = "/secrets/s%d" % len(secret_mounts)
415 secret_mounts[mnt] = {
417 "content": self.secret_store.retrieve(self.job_order[param])
419 self.job_order[param] = {"$include": mnt}
423 "output_path": "/var/spool/cwl",
424 "cwd": "/var/spool/cwl",
425 "priority": self.priority,
426 "state": "Committed",
427 "container_image": arvados_jobs_image(self.arvrunner, self.jobs_image),
429 "/var/lib/cwl/cwl.input.json": {
431 "content": self.job_order
435 "path": "/var/spool/cwl/cwl.output.json"
438 "kind": "collection",
442 "secret_mounts": secret_mounts,
443 "runtime_constraints": {
444 "vcpus": math.ceil(self.submit_runner_cores),
445 "ram": 1024*1024 * (math.ceil(self.submit_runner_ram) + math.ceil(self.collection_cache_size)),
448 "use_existing": False, # Never reuse the runner container - see #15497.
452 if self.embedded_tool.tool.get("id", "").startswith("keep:"):
453 sp = self.embedded_tool.tool["id"].split('/')
454 workflowcollection = sp[0][5:]
455 workflowname = "/".join(sp[1:])
456 workflowpath = "/var/lib/cwl/workflow/%s" % workflowname
457 container_req["mounts"]["/var/lib/cwl/workflow"] = {
458 "kind": "collection",
459 "portable_data_hash": "%s" % workflowcollection
462 packed = packed_workflow(self.arvrunner, self.embedded_tool, self.merged_map)
463 workflowpath = "/var/lib/cwl/workflow.json#main"
464 container_req["mounts"]["/var/lib/cwl/workflow.json"] = {
468 if self.embedded_tool.tool.get("id", "").startswith("arvwf:"):
469 container_req["properties"]["template_uuid"] = self.embedded_tool.tool["id"][6:33]
472 # --local means execute the workflow instead of submitting a container request
473 # --api=containers means use the containers API
474 # --no-log-timestamps means don't add timestamps (the logging infrastructure does this)
475 # --disable-validate because we already validated so don't need to do it again
476 # --eval-timeout is the timeout for javascript invocation
477 # --parallel-task-count is the number of threads to use for job submission
478 # --enable/disable-reuse sets desired job reuse
479 # --collection-cache-size sets aside memory to store collections
480 command = ["arvados-cwl-runner",
483 "--no-log-timestamps",
484 "--disable-validate",
486 "--eval-timeout=%s" % self.arvrunner.eval_timeout,
487 "--thread-count=%s" % self.arvrunner.thread_count,
488 "--enable-reuse" if self.enable_reuse else "--disable-reuse",
489 "--collection-cache-size=%s" % self.collection_cache_size]
492 command.append("--output-name=" + self.output_name)
493 container_req["output_name"] = self.output_name
496 command.append("--output-tags=" + self.output_tags)
498 if runtimeContext.debug:
499 command.append("--debug")
501 if runtimeContext.storage_classes != "default":
502 command.append("--storage-classes=" + runtimeContext.storage_classes)
504 if runtimeContext.intermediate_storage_classes != "default":
505 command.append("--intermediate-storage-classes=" + runtimeContext.intermediate_storage_classes)
508 command.append("--on-error=" + self.on_error)
510 if self.intermediate_output_ttl:
511 command.append("--intermediate-output-ttl=%d" % self.intermediate_output_ttl)
513 if self.arvrunner.trash_intermediate:
514 command.append("--trash-intermediate")
516 if self.arvrunner.project_uuid:
517 command.append("--project-uuid="+self.arvrunner.project_uuid)
520 command.append("--enable-dev")
522 command.extend([workflowpath, "/var/lib/cwl/cwl.input.json"])
524 container_req["command"] = command
529 def run(self, runtimeContext):
530 runtimeContext.keepprefix = "keep:"
531 job_spec = self.arvados_job_spec(runtimeContext)
532 if self.arvrunner.project_uuid:
533 job_spec["owner_uuid"] = self.arvrunner.project_uuid
535 extra_submit_params = {}
536 if runtimeContext.submit_runner_cluster:
537 extra_submit_params["cluster_id"] = runtimeContext.submit_runner_cluster
539 if runtimeContext.submit_request_uuid:
540 if "cluster_id" in extra_submit_params:
541 # Doesn't make sense for "update" and actually fails
542 del extra_submit_params["cluster_id"]
543 response = self.arvrunner.api.container_requests().update(
544 uuid=runtimeContext.submit_request_uuid,
546 **extra_submit_params
547 ).execute(num_retries=self.arvrunner.num_retries)
549 response = self.arvrunner.api.container_requests().create(
551 **extra_submit_params
552 ).execute(num_retries=self.arvrunner.num_retries)
554 self.uuid = response["uuid"]
555 self.arvrunner.process_submitted(self)
557 logger.info("%s submitted container_request %s", self.arvrunner.label(self), response["uuid"])
559 workbench1 = self.arvrunner.api.config()["Services"]["Workbench1"]["ExternalURL"]
560 workbench2 = self.arvrunner.api.config()["Services"]["Workbench2"]["ExternalURL"]
563 url = "{}processes/{}".format(workbench2, response["uuid"])
565 url = "{}container_requests/{}".format(workbench1, response["uuid"])
567 logger.info("Monitor workflow progress at %s", url)
570 def done(self, record):
572 container = self.arvrunner.api.containers().get(
573 uuid=record["container_uuid"]
574 ).execute(num_retries=self.arvrunner.num_retries)
575 container["log"] = record["log_uuid"]
577 logger.exception("%s while getting runner container", self.arvrunner.label(self))
578 self.arvrunner.output_callback({}, "permanentFail")
580 super(RunnerContainer, self).done(container)