Bug 1017785

Summary:	Puppet Install Distributor should error or log when it can't write Puppet Modules to target directory
Product:	Red Hat Satellite	Reporter:	Mike McCune <mmccune>
Component:	Content Management	Assignee:	Mike McCune <mmccune>
Status:	CLOSED WONTFIX	QA Contact:	Og Maciel <omaciel>
Severity:	high	Docs Contact:
Priority:	unspecified
Version:	6.0.2	CC:	bkearney, dcleal, jmontleo, omaciel, stbenjam, sthirugn
Target Milestone:	Unspecified	Keywords:	Triaged
Target Release:	Unused
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2015-07-14 17:22:20 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:

Description Mike McCune 2013-10-10 13:51:21 UTC

If SELinux is enabled or for some reason the target directory for the Pulp puppet_install_distributor isn't available to be written to the publish task will still succeed and nothing will be written to the logfiles.

This makes it difficult as a user to know what is going on and why your puppet modules are not showing up in the target directory.

Initially we would at least see logging in /var/log/pulp/pulp.log that it wasn't able to write to the directory.

Ideally the pulp publish task would fail and we would get an error report.

Comment 1 RHEL Program Management 2013-10-10 14:06:31 UTC

Since this issue was entered in Red Hat Bugzilla, the release flag has been
set to ? to ensure that it is properly evaluated for this release.

Comment 3 Mike McCune 2013-10-10 16:21:47 UTC

associated pulp bug:

https://bugzilla.redhat.com/show_bug.cgi?id=1017815

Comment 4 Mike McCune 2013-10-10 16:22:07 UTC

requires a rebuild of Pulp 2.3

Comment 7 Dominic Cleal 2014-09-01 10:03:53 UTC

DEVELOPER TESTED:

Verified that a single log entry is made in /var/log/messages (syslog, not pulp.log, which doesn't exist):

Sep  1 05:56:29 cloud-qe-10 pulp: pulp.server.managers.repo.publish:INFO: publish failed for repo [Default_Organization-DEV-dcleal_puppet] with distributor ID [Default_Organization-DEV-dcleal_puppet]

No indication of *why* the failure occurred is listed as far as I can see.  The promotion task however succeeded, which suggests the error report requested in comment #0 wasn't implemented.  This usability isn't great.  Will leave up to Mike or QE to decide on this.

Verification was on Satellite-6.0.4-RHEL-6-20140829.0 by creating a CV containing a Puppet module, successfully publishing version 1 to /etc/puppet/environments/.  chown root /etc/puppet/environments, then promoted to DEV.

Comment 9 Dominic Cleal 2014-09-01 10:04:58 UTC

(In reply to Dominic Cleal from comment #7)
> Verified that a single log entry is made in /var/log/messages (syslog, not
> pulp.log, which doesn't exist):
> 
> Sep  1 05:56:29 cloud-qe-10 pulp: pulp.server.managers.repo.publish:INFO:
> publish failed for repo [Default_Organization-DEV-dcleal_puppet] with
> distributor ID [Default_Organization-DEV-dcleal_puppet]
> 
> No indication of *why* the failure occurred is listed as far as I can see.

Ah, bug #1017815 suggests this would be DEBUG level, which isn't enabled.

Comment 12 Og Maciel 2014-09-03 13:36:55 UTC

Though there is an error in the logs, the UI shows that the task was completed successfully, which is not the expected behavior. This fails QE with latest build (Satellite-6.0.4-RHEL-7-20140829.0 and Satellite-6.0.4-RHEL-6-20140829.0).

Comment 13 Bryan Kearney 2015-07-14 17:22:20 UTC

We do not believe this is an issue an longer. if you continue to see it, please re-open.