Log job failure even when there are retries configured

changes.d/6169.fix

@@ -0,0 +1 @@
+Ensure that job submit/failure is logged, even when retries are planned.

cylc/flow/task_events_mgr.py

@@ -744,7 +744,7 @@ def process_message(
                 # Already failed.
                 return True
             if self._process_message_failed(
-                itask, event_time, self.JOB_FAILED, forced
+                itask, event_time, self.JOB_FAILED, forced, message
             ):
                 self.spawn_children(itask, TASK_OUTPUT_FAILED)
 
@@ -795,7 +795,7 @@ def process_message(
             self.workflow_db_mgr.put_update_task_jobs(
                 itask, {"run_signal": signal})
             if self._process_message_failed(
-                itask, event_time, self.JOB_FAILED, forced
+                itask, event_time, self.JOB_FAILED, forced, message
             ):
                 self.spawn_children(itask, TASK_OUTPUT_FAILED)
 
@@ -812,7 +812,7 @@ def process_message(
             self.workflow_db_mgr.put_update_task_jobs(
                 itask, {"run_signal": aborted_with})
             if self._process_message_failed(
-                itask, event_time, aborted_with, forced
+                itask, event_time, aborted_with, forced, message
             ):
                 self.spawn_children(itask, TASK_OUTPUT_FAILED)
 
@@ -928,11 +928,15 @@ def _process_message_check(
             return False
 
         severity_lvl: int = LOG_LEVELS.get(severity, INFO)
+        # Don't log submit/failure messages here:
+        if flag != self.FLAG_POLLED and message in {
+            self.EVENT_SUBMIT_FAILED, f'{FAIL_MESSAGE_PREFIX}ERR'
+        }:
+            return True
         # Demote log level to DEBUG if this is a message that duplicates what
         # gets logged by itask state change anyway (and not manual poll)
         if severity_lvl > DEBUG and flag != self.FLAG_POLLED and message in {
             self.EVENT_SUBMITTED, self.EVENT_STARTED, self.EVENT_SUCCEEDED,
-            self.EVENT_SUBMIT_FAILED, f'{FAIL_MESSAGE_PREFIX}ERR'
         }:
             severity_lvl = DEBUG
         LOG.log(severity_lvl, f"[{itask}] {flag}{message}{timestamp}")
@@ -1297,10 +1301,17 @@ def _retry_task(self, itask, wallclock_time, submit_retry=False):
         if itask.state_reset(TASK_STATUS_WAITING):
             self.data_store_mgr.delta_task_state(itask)
 
-    def _process_message_failed(self, itask, event_time, message, forced):
+    def _process_message_failed(
+            self, itask, event_time, message, forced, full_message
+    ):
         """Helper for process_message, handle a failed message.
 
         Return True if no retries (hence go to the failed state).
+
+        Args:
+            full_message:
+                If we have retries lined up we still tell users what
+                happened to cause the this attempt to fail.
         """
         no_retries = False
         if event_time is None:
@@ -1327,12 +1338,18 @@ def _process_message_failed(self, itask, event_time, message, forced):
                 self.data_store_mgr.delta_task_output(
                     itask, TASK_OUTPUT_FAILED)
                 self.data_store_mgr.delta_task_state(itask)
+            LOG.error(
+                f'[{itask}] {full_message or self.EVENT_FAILED} - '
+            )
         else:
             # There is an execution retry lined up.
             timer = itask.try_timers[TimerFlags.EXECUTION_RETRY]
             self._retry_task(itask, timer.timeout)
             delay_msg = f"retrying in {timer.delay_timeout_as_str()}"
-            LOG.warning(f"[{itask}] {delay_msg}")
+            LOG.warning(
+                f'[{itask}] {full_message or self.EVENT_FAILED} - '
+                f'{delay_msg}'
+            )
             msg = f"{self.JOB_FAILED}, {delay_msg}"
             self.setup_event_handlers(itask, self.EVENT_RETRY, msg)
         self._reset_job_timers(itask)
@@ -1404,7 +1421,6 @@ def _process_message_submit_failed(
         Return True if no retries (hence go to the submit-failed state).
         """
         no_retries = False
-        LOG.critical(f"[{itask}] {self.EVENT_SUBMIT_FAILED}")
         if event_time is None:
             event_time = get_current_time_string()
         self.workflow_db_mgr.put_update_task_jobs(itask, {
@@ -1430,6 +1446,7 @@ def _process_message_submit_failed(
                 self.data_store_mgr.delta_task_output(
                     itask, TASK_OUTPUT_SUBMIT_FAILED)
                 self.data_store_mgr.delta_task_state(itask)
+            LOG.error(f"[{itask}] {self.EVENT_SUBMIT_FAILED}")
         else:
             # There is a submission retry lined up.
             timer = itask.try_timers[TimerFlags.SUBMISSION_RETRY]

tests/functional/cylc-remove/00-simple/flow.cylc

@@ -15,7 +15,7 @@
         script = false
     [[cleaner]]
         script = """
-cylc__job__poll_grep_workflow_log -E '1/b/01:running.* \(received\)failed'
+cylc__job__poll_grep_workflow_log -E '1/b/01.* failed'
 # Remove the unhandled failed task
 cylc remove "$CYLC_WORKFLOW_ID//1/b"
 # Remove waiting 1/c

tests/functional/cylc-remove/02-cycling/flow.cylc

@@ -17,8 +17,8 @@
 [runtime]
     [[remover]]
         script = """
-            cylc__job__poll_grep_workflow_log -E '2020/bar/01:running.* \(received\)failed'
-            cylc__job__poll_grep_workflow_log -E '2021/baz/01:running.* \(received\)failed'
+            cylc__job__poll_grep_workflow_log -E '2020/bar/01.* failed'
+            cylc__job__poll_grep_workflow_log -E '2021/baz/01.* failed'
             # Remove the two unhandled failed tasks.
             cylc remove "$CYLC_WORKFLOW_ID//*/ba*:failed"
             # Remove the two unsatisfied waiting tasks.

tests/functional/cylc-trigger/02-filter-failed/flow.cylc

@@ -18,9 +18,9 @@
     [[fixer]]
         script = """
             cylc__job__wait_cylc_message_started
-            cylc__job__poll_grep_workflow_log -E '1/fixable1/01:running.* \(received\)failed'
-            cylc__job__poll_grep_workflow_log -E '1/fixable2/01:running.* \(received\)failed'
-            cylc__job__poll_grep_workflow_log -E '1/fixable3/01:running.* \(received\)failed'
+            cylc__job__poll_grep_workflow_log -E '\[1/fixable1/01:failed\] failed/ERR'
+            cylc__job__poll_grep_workflow_log -E '\[1/fixable2/01:failed\] failed/ERR'
+            cylc__job__poll_grep_workflow_log -E '\[1/fixable3/01:failed\] failed/ERR'
             cylc trigger "${CYLC_WORKFLOW_ID}//1/fixable*"
         """
     [[Z]]

tests/functional/cylc-trigger/04-filter-names/flow.cylc

@@ -22,11 +22,11 @@
     [[fixer]]
         script = """
             cylc__job__wait_cylc_message_started
-            cylc__job__poll_grep_workflow_log -E '1/fixable-1a/01.* \(received\)failed'
-            cylc__job__poll_grep_workflow_log -E '1/fixable-1b/01.* \(received\)failed'
-            cylc__job__poll_grep_workflow_log -E '1/fixable-2a/01.* \(received\)failed'
-            cylc__job__poll_grep_workflow_log -E '1/fixable-2b/01.* \(received\)failed'
-            cylc__job__poll_grep_workflow_log -E '1/fixable-3/01.* \(received\)failed'
+            cylc__job__poll_grep_workflow_log -E '1/fixable-1a/01.* failed'
+            cylc__job__poll_grep_workflow_log -E '1/fixable-1b/01.* failed'
+            cylc__job__poll_grep_workflow_log -E '1/fixable-2a/01.* failed'
+            cylc__job__poll_grep_workflow_log -E '1/fixable-2b/01.* failed'
+            cylc__job__poll_grep_workflow_log -E '1/fixable-3/01.* failed'
             cylc trigger "${CYLC_WORKFLOW_ID}//" \
                 '//1/FIXABLE-1' '//1/fixable-2*' '//1/fixable-3'
         """

tests/functional/hold-release/11-retrying/flow.cylc

@@ -18,7 +18,7 @@ t-retry-able => t-analyse
     [[t-hold-release]]
         script = """
             cylc__job__poll_grep_workflow_log -E \
-                '1/t-retry-able/01:running.* \(received\)failed'
+                '\[1/t-retry-able:waiting\] failed/ERR'
 
             cylc__job__poll_grep_workflow_log -E \
                 '1/t-retry-able/01:running.* => waiting'

tests/functional/reload/10-runahead.t

@@ -30,7 +30,7 @@ run_fail "${TEST_NAME}" cylc play --debug --no-detach "${WORKFLOW_NAME}"
 #-------------------------------------------------------------------------------
 TEST_NAME=${TEST_NAME_BASE}-check-fail
 DB_FILE="$RUN_DIR/${WORKFLOW_NAME}/log/db"
-QUERY='SELECT COUNT(*) FROM task_states WHERE status == "failed"'
+QUERY="SELECT COUNT(*) FROM task_states WHERE status == 'failed'"
 cmp_ok <(sqlite3 "$DB_FILE" "$QUERY") <<< "4"
 #-------------------------------------------------------------------------------
 purge

tests/functional/reload/25-xtriggers.t

@@ -42,7 +42,7 @@ init_workflow "${TEST_NAME_BASE}" <<'__FLOW_CONFIG__'
     [[reload]]
         script = """
             # wait for "broken" to fail
-            cylc__job__poll_grep_workflow_log -E '1/broken/01.* \(received\)failed/ERR'
+            cylc__job__poll_grep_workflow_log -E '1/broken.*failed/ERR'
             # fix "broken" to allow it to pass
             sed -i 's/false/true/' "${CYLC_WORKFLOW_RUN_DIR}/flow.cylc"
             # reload the workflow
@@ -63,7 +63,7 @@ workflow_run_ok "${TEST_NAME_BASE}-run" cylc play "${WORKFLOW_NAME}" --no-detach
 log_scan "${TEST_NAME_BASE}-scan" \
     "$(cylc cat-log -m p "${WORKFLOW_NAME}")" \
     1 1 \
-    '1/broken.* (received)failed/ERR'
+    '1/broken.*failed/ERR'
 
 log_scan "${TEST_NAME_BASE}-scan" \
     "$(cylc cat-log -m p "${WORKFLOW_NAME}")" 1 1 \

tests/functional/reload/runahead/flow.cylc

@@ -20,7 +20,7 @@
         script = true
     [[reloader]]
         script = """
-            cylc__job__poll_grep_workflow_log -E "${CYLC_TASK_CYCLE_POINT}/foo/01:running.*\(received\)failed"
+            cylc__job__poll_grep_workflow_log -E "${CYLC_TASK_CYCLE_POINT}/foo/01:running.*failed"
             perl -pi -e 's/(runahead limit = )P1( # marker)/\1 P3\2/' $CYLC_WORKFLOW_RUN_DIR/flow.cylc
             cylc reload $CYLC_WORKFLOW_ID
         """

tests/functional/spawn-on-demand/10-retrigger/flow.cylc

@@ -18,7 +18,7 @@
       """
    [[triggerer]]
       script = """
-         cylc__job__poll_grep_workflow_log -E '1/oops/01:running.* \(received\)failed'
+         cylc__job__poll_grep_workflow_log -E '1/oops/01.* failed'
          cylc trigger "${CYLC_WORKFLOW_ID}//1/oops"
        """
    [[foo, bar]]

tests/functional/triggering/19-and-suicide/flow.cylc

@@ -16,7 +16,7 @@
     [[t0]]
         # https://github.com/cylc/cylc-flow/issues/2655
         # "1/t2" should not suicide on "1/t1:failed"
-        script = cylc__job__poll_grep_workflow_log -E '1/t1.* \(received\)failed'
+        script = cylc__job__poll_grep_workflow_log -E '1/t1.* failed'
     [[t1]]
         script = false
     [[t2]]

tests/integration/test_task_events_mgr.py

@@ -14,17 +14,14 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
-from itertools import product
 import logging
 from typing import Any as Fixture
 
 from cylc.flow.task_events_mgr import TaskJobLogsRetrieveContext
 from cylc.flow.scheduler import Scheduler
 from cylc.flow.data_store_mgr import (
     JOBS,
-    TASK_STATUSES_ORDERED,
     TASK_STATUS_WAITING,
-    TASK_STATUS_SUBMIT_FAILED,
 )
 
 
@@ -79,17 +76,22 @@ async def test__insert_task_job(flow, one_conf, scheduler, start, validate):
     with correct submit number.
     """
     conf = {
-        'scheduling': {'graph': {'R1': 'rhenas'}},
-        'runtime': {'rhenas': {'simulation': {
-            'fail cycle points': '1',
-            'fail try 1 only': False,
-    }}}}
+        "scheduling": {"graph": {"R1": "rhenas"}},
+        "runtime": {
+            "rhenas": {
+                "simulation": {
+                    "fail cycle points": "1",
+                    "fail try 1 only": False,
+                }
+            }
+        },
+    }
     id_ = flow(conf)
     schd = scheduler(id_)
     async with start(schd):
         # Set task to running:
-        itask =  schd.pool.get_tasks()[0]
-        itask.state.status = 'running'
+        itask = schd.pool.get_tasks()[0]
+        itask.state.status = "running"
         itask.submit_num += 1
 
         # Not run _insert_task_job yet:
@@ -170,3 +172,46 @@ async def test__always_insert_task_job(
             '1/broken/01': 'submit-failed',
             '1/broken2/01': 'submit-failed'
         }
+
+
+async def test__process_message_failed_with_retry(one, start):
+    """Log job failure, even if a retry is scheduled.
+
+    See: https://github.com/cylc/cylc-flow/pull/6169
+
+    """
+
+    async with start(one) as LOG:
+        fail_once = one.pool.get_tasks()[0]
+        # Add retry timers:
+        one.task_job_mgr._set_retry_timers(
+            fail_once, {
+                'execution retry delays': [1],
+                'submission retry delays': [1]
+            })
+
+        # Process submit failed message with and without retries:
+        one.task_events_mgr._process_message_submit_failed(
+            fail_once, None, 1, False)
+        last_record = LOG.records[-1]
+        assert last_record.levelno == logging.WARNING
+        assert '1/one:waiting(queued)' in last_record.message
+
+        one.task_events_mgr._process_message_submit_failed(
+            fail_once, None, 2, False)
+        last_record = LOG.records[-1]
+        assert last_record.levelno == logging.ERROR
+        assert 'submission failed' in last_record.message
+
+        # Process failed message with and without retries:
+        one.task_events_mgr._process_message_failed(
+            fail_once, None, 'failed', False, 'failed/OOK')
+        last_record = LOG.records[-1]
+        assert last_record.levelno == logging.WARNING
+        assert 'failed/OOK' in last_record.message
+
+        one.task_events_mgr._process_message_failed(
+            fail_once, None, 'failed', False, 'failed/OOK')
+        last_record = LOG.records[-1]
+        assert last_record.levelno == logging.ERROR
+        assert 'failed/OOK' in last_record.message