Jira (PDB-2495) Use real 'noop' status from puppet agent report, when it has the feature to do this

Kenneth Barber (JIRA)

unread,

Mar 7, 2016, 9:44:03 AM3/7/16

to puppe...@googlegroups.com

Kenneth Barber created an issue

PuppetDB /

PDB-2495

Use real 'noop' status from puppet agent report, when it has the feature to do this

Issue Type:	New Feature
Assignee:	Unassigned
Created:	2016/03/07 6:43 AM
Priority:	Normal
Reporter:	Kenneth Barber

Right now, we simulate what is a noop report in our terminus:

https://github.com/puppetlabs/puppetdb/blob/master/puppet/lib/puppet/reports/puppetdb.rb#L40-L41

However, this isn't a true report of noop, anyone who happens to make a single resource noop will show up as the whole report being noop.

What we really need is to consume a proper 'noop' status from the agent, and if there isn't one falling back to the prior simulation for backwards compatibility purposes.

Add Comment

This message was sent by Atlassian JIRA (v6.4.12#64027-sha1:e3691cc)

Scott Walker (JIRA)

unread,

Mar 7, 2016, 11:57:03 AM3/7/16

to puppe...@googlegroups.com

Scott Walker commented on

PDB-2495

Re: Use real 'noop' status from puppet agent report, when it has the feature to do this

Would it be possible to implement the fallback mechanism before the PUP ticket? That would solve the immediate user problem without blocking on that one.

Add Comment

Kenneth Barber (JIRA)

unread,

Mar 7, 2016, 12:23:04 PM3/7/16

to puppe...@googlegroups.com

Kenneth Barber commented on

PDB-2495

Re: Use real 'noop' status from puppet agent report, when it has the feature to do this

You mean, write a fallback that tests for this new imaginary field? Wouldn't we like to know this is going to be implemented at all, in the way we desire it before writing against a guessed API?

Add Comment

Scott Walker (JIRA)

unread,

Mar 8, 2016, 6:14:03 AM3/8/16

to puppe...@googlegroups.com

Scott Walker commented on

PDB-2495

Re: Use real 'noop' status from puppet agent report, when it has the feature to do this

Agreed, so if the PUP change isn't implemented at all we still need the simulation approach used today, albeit modified. I was hoping we could phase the fix. Phase 1 would be to change the noop simulation to handle individual noop resources. Phase 2 would wrap a conditional around that same logic that runs it only if the report doesn't indicate the value. Phase 2 is blocked on the PUP ticket but phase 1 isn't (and would bring immediate value).

Add Comment

Kenneth Barber (JIRA)

unread,

Mar 9, 2016, 5:29:02 AM3/9/16

to puppe...@googlegroups.com

Kenneth Barber commented on

PDB-2495

Re: Use real 'noop' status from puppet agent report, when it has the feature to do this

Okay, so it sounds like you have another proposal for how to change the simulation, can you describe it here? We'll probably raise another ticket for that fix, since its different to this proper fix.

Add Comment

Scott Walker (JIRA)

unread,

Mar 9, 2016, 10:31:02 AM3/9/16

to puppe...@googlegroups.com

Scott Walker commented on

PDB-2495

Re: Use real 'noop' status from puppet agent report, when it has the feature to do this

I think the simulation today is something like:

"If there is one or more events with noop status and no events with failure status then this report is noop"

I think it needs to be:

"If there are only events with noop status then this report is noop"

Add Comment

Chris Cowell (JIRA)

unread,

May 11, 2016, 1:17:09 PM5/11/16

to puppe...@googlegroups.com

Chris Cowell commented on

PDB-2495

Re: Use real 'noop' status from puppet agent report, when it has the feature to do this

Kenneth Barber or Scott Walker Do we have an ETA for this ticket? I ask because Config Mgmt ticket PE-13095 is blocked by it and I wanted to make sure this didn't fall through the cracks.