External-Mirrors/awx
2
0
Fork 0
mirror of https://github.com/ansible/awx.git synced 2026-04-04 17:55:06 -02:30
Code Issues Packages Projects Releases Wiki Activity

Add RECEPTOR_KEEP_WORK_ON_ERROR setting

Browse Source 
If RECEPTOR_KEEP_WORK_ON_ERROR is set to true receptor work unit will not be automatically released

Co-Authored-By: Chris Meyers <chrismeyersfsu@users.noreply.github.com>
This commit is contained in:
Hao Liu
2024-07-22 12:32:53 -04:00
parent 7691365aea
commit 139d8f0ae2
5 changed files with 24 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
awx/main/tasks/system.py
View File

@@ -54,7 +54,7 @@ from awx.main.models import (
Job,
convert_jsonfields,
)
from awx.main.constants import ACTIVE_STATES
from awx.main.constants import ACTIVE_STATES, ERROR_STATES
from awx.main.dispatch.publish import task
from awx.main.dispatch import get_task_queuename, reaper
from awx.main.utils.common import ignore_inventory_computed_fields, ignore_inventory_group_removal
@@ -685,6 +685,8 @@ def awx_receptor_workunit_reaper():
unit_ids = [id for id in receptor_work_list]
jobs_with_unreleased_receptor_units = UnifiedJob.objects.filter(work_unit_id__in=unit_ids).exclude(status__in=ACTIVE_STATES)
if settings.RECEPTOR_KEEP_WORK_ON_ERROR:
jobs_with_unreleased_receptor_units = jobs_with_unreleased_receptor_units.exclude(status__in=ERROR_STATES)
for job in jobs_with_unreleased_receptor_units:
logger.debug(f"{job.log_format} is not active, reaping receptor work unit {job.work_unit_id}")
receptor_ctl.simple_command(f"work cancel {job.work_unit_id}")
@@ -704,7 +706,10 @@ def awx_k8s_reaper():
logger.debug("Checking for orphaned k8s pods for {}.".format(group))
pods = PodManager.list_active_jobs(group)
time_cutoff = now() - timedelta(seconds=settings.K8S_POD_REAPER_GRACE_PERIOD)
for job in UnifiedJob.objects.filter(pk__in=pods.keys(), finished__lte=time_cutoff).exclude(status__in=ACTIVE_STATES):
reap_job_candidates = UnifiedJob.objects.filter(pk__in=pods.keys(), finished__lte=time_cutoff).exclude(status__in=ACTIVE_STATES)
if settings.RECEPTOR_KEEP_WORK_ON_ERROR:
reap_job_candidates = reap_job_candidates.exclude(status__in=ERROR_STATES)
for job in reap_job_candidates:
logger.debug('{} is no longer active, reaping orphaned k8s pod'.format(job.log_format))
try:
pm = PodManager(job)