Project

General

Profile

Actions

Bug #20114

closed

It takes too long for parent task to stop after its sub-tasks have stopped

Added by Adam Ruzicka almost 7 years ago. Updated over 5 years ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Category:
Dynflow
Target version:
-
Difficulty:
Triaged:
Fixed in Releases:
Found in Releases:

Description

Cloned from https://bugzilla.redhat.com/show_bug.cgi?id=1465296

Description of problem:
Bulk tasks are receiving events when their sub-tasks finish. This gets rather slow with large number of sub-tasks. We should take another approach and make the parent task poll the state of its sub-tasks.

Version-Release number of selected component (if applicable):

How reproducible:
Easily triggered by using REX at scale

Steps to Reproduce:
1. Run a remote execution job on high number (thousands) of hosts
2. Wait until all of its sub-tasks are stopped
3. It may reach into state where all the sub-tasks are stopped but the parent task is still running for another hour or so (depends on numbers of hosts)

Actual results:
Parent task keeps running for hours after its sub-tasks have stopped

Expected results:
The parent task is stopped shortly after its sub-tasks have stopped

Additional info:


Related issues 1 (0 open1 closed)

Related to Foreman Remote Execution - Refactor #20466: Use the polling sub plans functionality in RunHostsJobClosedAdam Ruzicka08/01/2017Actions
Actions #1

Updated by Adam Ruzicka almost 7 years ago

  • Category set to Dynflow
  • Target version set to 113
Actions #2

Updated by Ivan Necas over 6 years ago

  • Related to Refactor #20466: Use the polling sub plans functionality in RunHostsJob added
Actions #3

Updated by Ivan Necas over 6 years ago

  • Status changed from New to Closed

Marking as resolved, as this should be fixed by #20466

Actions

Also available in: Atom PDF