Skip to content

make sure the pending task is still in pendingJobs queue #233

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
smuzaffar wants to merge 5 commits into
base: V00-34-XX
Choose a base branch
from
Open

make sure the pending task is still in pendingJobs queue #233

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
6989757
make sure the pending task is still in pendingJobs queue
smuzaffar Jan 20, 2026
8efd5ea
resursively mark job broken if a deps is broken
smuzaffar Jan 26, 2026
b7bcd05
do not fail on forceDone tasks
smuzaffar Jan 26, 2026
5b2ea80
release resources if fails to schedule a pending job
smuzaffar Jan 27, 2026
7ab46ca
Update scheduler.py
smuzaffar Jan 27, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 23 additions & 13 deletions scheduler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,10 +16,11 @@ def transition(what, fromList, toList, key="unknow"):
try:
print("DEBUG:%s: transition %s" % (key, what))
fromList.remove(what)
toList.append(what)
except ValueError as e:
print (what + " not in source list")
raise e
toList.append(what)
return False
return True

class Scheduler(object):
# A simple job scheduler.
Expand Down Expand Up @@ -147,12 +148,18 @@ def __doRescheduleParallel(self):
parallelJobs = [j for j in self.pendingJobs if self.jobs[j]["scheduler"] == "parallel"]
# First of all clean up the pending parallel jobs from all those
# which have broken dependencies.
for taskId in parallelJobs:
brokenDeps = [dep for dep in self.jobs[taskId]["deps"] if dep in self.brokenJobs]
if not brokenDeps:
continue
transition(taskId, self.pendingJobs, self.brokenJobs, "parallel:pending->broken")
self.errors[taskId] = "The following dependencies could not complete:\n%s" % "\n".join(brokenDeps)
while True:
has_broken = False
for taskId in parallelJobs[:]:
brokenDeps = [dep for dep in self.jobs[taskId]["deps"] if dep in self.brokenJobs]
if not brokenDeps:
continue
has_broken = True
parallelJobs.remove(taskId)
transition(taskId, self.pendingJobs, self.brokenJobs, "parallel:pending->broken")
self.errors[taskId] = "The following dependencies could not complete:\n%s" % "\n".join(brokenDeps)
if not has_broken:
break

# If no tasks left, quit. Notice we need to check also for serial jobs
# since they might queue more parallel payloads.
Expand All @@ -175,7 +182,8 @@ def __doRescheduleParallel(self):
if dumpMsg:
self.log("Pending tasks: %s: %s" % (taskId, pendingDeps),30)
continue
allJobs.append({"id": taskId, "priorty": self.jobs[taskId]["priorty"]})
if taskId in self.pendingJobs:
allJobs.append({"id": taskId, "priorty": self.jobs[taskId]["priorty"]})
buildJobs =[]
downloadJobs = []
forceJobs = []
Expand All @@ -199,10 +207,12 @@ def __doRescheduleParallel(self):
else:
buildJobs = buildJobs[:bldCount]
for taskId in forceJobs + downloadJobs + buildJobs:
taskType = taskId.split("-")[0]
self.runningJobsCount[taskType] += 1
transition(taskId, self.pendingJobs, self.runningJobs, "parallel:pending->running")
self.__scheduleParallel(taskId, self.jobs[taskId]["spec"], priorty=self.jobs[taskId]["priorty"])
if transition(taskId, self.pendingJobs, self.runningJobs, "parallel:pending->running"):
taskType = taskId.split("-")[0]
self.runningJobsCount[taskType] += 1
self.__scheduleParallel(taskId, self.jobs[taskId]["spec"], priorty=self.jobs[taskId]["priorty"])
elif self.resourceManager:
self.resourceManager.releaseResourcesForExternal(taskId)

# Update the job with the result of running.
def __updateJobStatus(self, taskId, error):
Expand Down