Bug #36130
job invocation shows wrong info after remote execution job (regression from 6.11)
Description
Cloned from https://bugzilla.redhat.com/show_bug.cgi?id=2167396
Description of problem:
A customer reported, that after they migrated to Satellite 6.12 (from 6.11) the job invocation overview is showing the wrong info. After running an ansible remote execution job on several hosts, if one host failed, all hosts are shown as failed.
Version-Release number of selected component (if applicable):
Satellite 6.12.1
How reproducible:
always
Steps to Reproduce:
1. Prepare to run any ansible job in Satellite 6.12.1 and run it on a group of hosts
2. Force the job to fail on one host (for example with a wrong DNF repository, which causes this one host to fail "really")
3. Start the job
4. Ansible-runner returns error code 2 (partial failure) instead of 0 and this causes Satellite to mark every host in this job run as failed -> not just the one, which really failed
Actual results:
All hosts are shown as failed when only one has failed
Expected results:
Show correct info about remote execution on hosts (failed / successful)
Additional info:
The customer is relying on this info for patching but does not get the right info currently. This seems to be a regression from Satellite 6.11.
Related issues
Associated revisions
Refs #36130 - Update code for compatibility with Ansible 2.12.2
Set the exit status to success only for rescued tasks.
History
#1
Updated by Nofar Alfassi 4 months ago
Updated by - Assignee set to Nofar Alfassi
- Subject changed from job invocation shows wrong info after remote execution job 9regression from 6.11) to job invocation shows wrong info after remote execution job 9regression from 6.11)
#2
Updated by Oliver Ilian 3 months ago
Updated by - Subject changed from job invocation shows wrong info after remote execution job 9regression from 6.11) to job invocation shows wrong info after remote execution job (regression from 6.11)
Do we know when this issue can be fixed? It is urgent for my customer.
#3
Updated by Nofar Alfassi 3 months ago
Oliver Ilian wrote:
Do we know when this issue can be fixed? It is urgent for my customer.
I'm currently working on it, so as soon as it's done.
#4
Updated by The Foreman Bot 3 months ago
Updated by - Status changed from New to Ready For Testing
- Pull request https://github.com/theforeman/smart_proxy_ansible/pull/75 added
#5
Updated by The Foreman Bot 3 months ago
- Fixed in Releases smart_proxy_ansible-3.5.1 added
#6
Updated by Nofar Alfassi 3 months ago
- Status changed from Ready For Testing to Closed
Applied in changeset smart_proxy_ansible|a9c9ab2d4d0d1c38aefff07c8eb7f9c1dbcb6ab3.
#7
Updated by Nofar Alfassi 3 months ago
- Target version changed from foreman_ansible-10.0.0 to 1689
- Fixed in Releases smart_proxy_ansible-3.5.3 added
- Fixed in Releases deleted (
smart_proxy_ansible-3.5.1)
#8
Updated by Nofar Alfassi 3 months ago
- Target version changed from 1689 to smart_proxy_ansible-3.5.3
#9
Updated by The Foreman Bot 3 months ago
- Pull request https://github.com/theforeman/smart_proxy_ansible/pull/77 added
#10
Updated by Nofar Alfassi 3 months ago
- Related to Bug #36210: Set correct exit status for Ansible job run on multiple hosts with a non-existing module added
#11
Updated by Nofar Alfassi 3 months ago
- Related to deleted (Bug #36210: Set correct exit status for Ansible job run on multiple hosts with a non-existing module)
#12
Updated by Nofar Alfassi 3 months ago
- Precedes Bug #36210: Set correct exit status for Ansible job run on multiple hosts with a non-existing module added
#13
Updated by Nofar Alfassi 3 months ago
- Fixed in Releases smart_proxy_ansible-3.5.4 added
- Fixed in Releases deleted (
smart_proxy_ansible-3.5.3)
#14
Updated by Nofar Alfassi 3 months ago
- Has duplicate Bug #36206: One or more unreachable hosts result in "failed" job for every host added
Fixes #36130 - show correct exit code for multi-host job execution
Override per-host exit status only if the run ends early due to an error