X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/39fc2f223fae40dc4fb160758e76ca39304b44af..9f86e39c7720f1291194fc8eea867a175e94e07f:/sdk/cwl/arvados_cwl/arvcontainer.py

diff --git a/sdk/cwl/arvados_cwl/arvcontainer.py b/sdk/cwl/arvados_cwl/arvcontainer.py
index 4d0fde7440..be8e557bd8 100644
--- a/sdk/cwl/arvados_cwl/arvcontainer.py
+++ b/sdk/cwl/arvados_cwl/arvcontainer.py
@@ -367,6 +367,12 @@ class ArvadosContainer(JobBase):
                 logger.warning("%s API revision is %s, revision %s is required to support setting properties on output collections.",
                                self.arvrunner.label(self), self.arvrunner.api._rootDesc["revision"], "20220510")
 
+        ram_multiplier = [1]
+
+        oom_retry_req, _ = self.get_requirement("http://arvados.org/cwl#OutOfMemoryRetry")
+        if oom_retry_req and oom_retry_req.get('memoryRetryMultipler'):
+            ram_multiplier.append(oom_retry_req.get('memoryRetryMultipler'))
+
         if runtimeContext.runnerjob.startswith("arvwf:"):
             wfuuid = runtimeContext.runnerjob[6:runtimeContext.runnerjob.index("#")]
             wfrecord = self.arvrunner.api.workflows().get(uuid=wfuuid).execute(num_retries=self.arvrunner.num_retries)
@@ -374,17 +380,20 @@ class ArvadosContainer(JobBase):
                 container_request["name"] = wfrecord["name"]
             container_request["properties"]["template_uuid"] = wfuuid
 
-        self.output_callback = self.arvrunner.get_wrapped_callback(self.output_callback)
+        if self.attempt_count == 0:
+            self.output_callback = self.arvrunner.get_wrapped_callback(self.output_callback)
 
         try:
             ram = runtime_constraints["ram"]
 
-            for i in range(1, 4):
+            self.uuid = runtimeContext.submit_request_uuid
+
+            for i in ram_multiplier:
                 runtime_constraints["ram"] = ram * i
 
-                if runtimeContext.submit_request_uuid:
+                if self.uuid:
                     response = self.arvrunner.api.container_requests().update(
-                        uuid=runtimeContext.submit_request_uuid,
+                        uuid=self.uuid,
                         body=container_request,
                         **extra_submit_params
                     ).execute(num_retries=self.arvrunner.num_retries)
@@ -393,22 +402,21 @@ class ArvadosContainer(JobBase):
                         body=container_request,
                         **extra_submit_params
                     ).execute(num_retries=self.arvrunner.num_retries)
-                    runtimeContext.submit_request_uuid = response["uuid"]
+                    self.uuid = response["uuid"]
 
                 if response["container_uuid"] is not None:
                     break
 
             if response["container_uuid"] is None:
-                runtime_constraints["ram"] = ram * (self.attempt_count+1)
+                runtime_constraints["ram"] = ram * ram_multiplier[self.attempt_count]
 
             container_request["state"] = "Committed"
             response = self.arvrunner.api.container_requests().update(
-                uuid=runtimeContext.submit_request_uuid,
+                uuid=self.uuid,
                 body=container_request,
                 **extra_submit_params
             ).execute(num_retries=self.arvrunner.num_retries)
 
-            self.uuid = response["uuid"]
             self.arvrunner.process_submitted(self)
             self.attempt_count += 1
 
@@ -422,6 +430,11 @@ class ArvadosContainer(JobBase):
             self.output_callback({}, "permanentFail")
 
     def out_of_memory_retry(self, record, container):
+        oom_retry_req, _ = self.get_requirement("http://arvados.org/cwl#OutOfMemoryRetry")
+        if oom_retry_req is None:
+            return False
+
+        # Sometimes it gets killed with no warning
         if container["exit_code"] == 137:
             return True
 
@@ -434,14 +447,10 @@ class ArvadosContainer(JobBase):
         def callback(v1, v2, v3):
             loglines[0] = v3
 
-        done.logtail(logc, callback, "", maxlen=200)
-
-        oom_matches = r'(bad_alloc|out ?of ?memory|Container using over 95% of memory)'
-
-        print("Checking loglines", loglines[0])
-
-        print("Match", re.search(oom_matches, loglines[0], re.IGNORECASE | re.MULTILINE))
+        done.logtail(logc, callback, "", maxlen=1000)
 
+        # Check allocation failure
+        oom_matches = oom_retry_req.get('memoryErrorRegex') or r'(bad_alloc|out ?of ?memory|memory ?error|container using over 9.% of memory)'
         if re.search(oom_matches, loglines[0], re.IGNORECASE | re.MULTILINE):
             return True
 
@@ -467,16 +476,17 @@ class ArvadosContainer(JobBase):
                 else:
                     processStatus = "permanentFail"
 
-                if processStatus == "permanentFail" and self.out_of_memory_retry(record, container):
-                    logger.info("%s Container failed with out of memory error, retrying with more RAM.",
+                if processStatus == "permanentFail" and self.attempt_count == 1 and self.out_of_memory_retry(record, container):
+                    logger.warning("%s Container failed with out of memory error, retrying with more RAM.",
                                  self.arvrunner.label(self))
                     self.job_runtime.submit_request_uuid = None
+                    self.uuid = None
                     self.run(None)
                     retried = True
                     return
 
                 if rcode == 137:
-                    logger.warning("%s Container may have been killed for using too much RAM.  Try resubmitting with a higher 'ramMin'.",
+                    logger.warning("%s Container may have been killed for using too much RAM.  Try resubmitting with a higher 'ramMin' or use the arv:OutOfMemoryRetry feature.",
                                  self.arvrunner.label(self))
             else:
                 processStatus = "permanentFail"