Skip to content

refactor: the pod failure analysis checks each container in two flat steps #460

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

Merged
HardMax71 merged 1 commit into from
Oct 23, 2026
Merged

refactor: the pod failure analysis checks each container in two flat steps #460

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
67 changes: 26 additions & 41 deletions backend/app/services/pod_monitor/event_mapper.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,12 @@

_UNREADABLE_OUTPUT = "The output couldn't be read from the pod's log."

_WAITING_ERRORS = {
"ImagePullBackOff": ExecutionErrorType.SYSTEM_ERROR,
"ErrImagePull": ExecutionErrorType.SYSTEM_ERROR,
"CrashLoopBackOff": ExecutionErrorType.SCRIPT_ERROR,
}


class WatchEventType(StringEnum):
"""Kubernetes watch event types."""
Expand Down Expand Up @@ -423,56 +429,35 @@ class FailureInfo:

def _analyze_failure(self, pod: k8s_client.V1Pod) -> FailureInfo:
"""Analyze pod failure and determine error type"""
# Default failure info
default = self.FailureInfo(
message=(pod.status.message if pod.status else None) or "Pod failed",
error_type=ExecutionErrorType.SYSTEM_ERROR,
)

if not pod.status:
return default
status = pod.status
if not status:
return self.FailureInfo(message="Pod failed", error_type=ExecutionErrorType.SYSTEM_ERROR)

# Check for resource limits
if pod.status.reason == "Evicted":
if status.reason == "Evicted":
return self.FailureInfo(
message="Pod evicted due to resource constraints", error_type=ExecutionErrorType.RESOURCE_LIMIT
)

# Check container statuses
for status in pod.status.container_statuses or []:
# Terminated container
if status.state and status.state.terminated:
terminated = status.state.terminated
if terminated.exit_code != 0:
return self.FailureInfo(
message=terminated.message
or pod.status.message
or f"Container exited with code {terminated.exit_code}",
error_type=ExecutionErrorType.SCRIPT_ERROR,
exit_code=terminated.exit_code,
)

# Waiting container
if status.state and status.state.waiting:
waiting = status.state.waiting
error_type_map = {
"ImagePullBackOff": ExecutionErrorType.SYSTEM_ERROR,
"ErrImagePull": ExecutionErrorType.SYSTEM_ERROR,
"CrashLoopBackOff": ExecutionErrorType.SCRIPT_ERROR,
}

if error_type := error_type_map.get(waiting.reason):
return self.FailureInfo(
message=waiting.message or f"Container waiting: {waiting.reason}", error_type=error_type
)

# Check for OOM
if "OOMKilled" in (pod.status.message or ""):
for container in status.container_statuses or []:
state = container.state or k8s_client.V1ContainerState()
if state.terminated and state.terminated.exit_code != 0:
exit_code = state.terminated.exit_code
return self.FailureInfo(
message=state.terminated.message or status.message or f"Container exited with code {exit_code}",
error_type=ExecutionErrorType.SCRIPT_ERROR,
exit_code=exit_code,
)
if state.waiting and (error_type := _WAITING_ERRORS.get(state.waiting.reason)):
return self.FailureInfo(
message=state.waiting.message or f"Container waiting: {state.waiting.reason}", error_type=error_type
)

if "OOMKilled" in (status.message or ""):
return self.FailureInfo(
message="Container killed due to out of memory", error_type=ExecutionErrorType.RESOURCE_LIMIT
)

return default
return self.FailureInfo(message=status.message or "Pod failed", error_type=ExecutionErrorType.SYSTEM_ERROR)

async def _extract_logs(self, pod: k8s_client.V1Pod) -> PodLogs | None:
"""Read a finished pod's result, or None while its main container hasn't terminated.
Expand Down
Loading