Bug 1577144

Summary:	ASB logs error about problem reading job state when deprovision
Product:	OpenShift Container Platform	Reporter:	Zihan Tang <zitang>
Component:	Service Broker	Assignee:	Shawn Hurley <shurley>
Status:	CLOSED CURRENTRELEASE	QA Contact:	Zihan Tang <zitang>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	3.10.0	CC:	aos-bugs, chezhang, jiazha, jmatthew, zhsun
Target Milestone:	---
Target Release:	3.10.0
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	No Doc Update
Doc Text:	undefined	Story Points:	---
Clone Of:		Environment:
Last Closed:	2018-12-20 21:12:41 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:
Embargoed:

Description Zihan Tang 2018-05-11 10:12:42 UTC

Description of problem:
When deprovision ,ASB logs error : 
[2018-05-11T10:04:10.526Z] [ERROR] - Could not find instance c3b999df-54f7-11e8-971f-0a580a800003 associated with job state 8fd3cd84-9409-4506-9dae-e81c72313eb3 - bundleinstances.automationbroker.io "c3b999df-54f7-11e8-971f-0a580a800003" not found
[2018-05-11T10:04:10.526Z] [ERROR] - problem reading job state: [c3b999df-54f7-11e8-971f-0a580a800003]. error: [bundleinstances.automationbroker.io "c3b999df-54f7-11e8-971f-0a580a800003" not found]


Version-Release number of selected component (if applicable):
1.2.10

How reproducible:
always
Steps to Reproduce:
1. provision a apb like postgresql 
2. deprovision the serviceinstance ,check the asb  log 

Actual results:
deprovision succeed. but asb logs error:
[2018-05-11T10:04:10.526Z] [ERROR] - Could not find instance c3b999df-54f7-11e8-971f-0a580a800003 associated with job state 8fd3cd84-9409-4506-9dae-e81c72313eb3 - bundleinstances.automationbroker.io "c3b999df-54f7-11e8-971f-0a580a800003" not found
[2018-05-11T10:04:10.526Z] [ERROR] - problem reading job state: [c3b999df-54f7-11e8-971f-0a580a800003]. error: [bundleinstances.automationbroker.io "c3b999df-54f7-11e8-971f-0a580a800003" not found]

Expected results:
no errors when deprovision 

Additional info:

Comment 1 Shawn Hurley 2018-05-14 16:45:29 UTC

PR: https://github.com/openshift/ansible-service-broker/pull/949

This will change the log statement about job state to be a Warning as well as update the message to be more clear.

Comment 2 David Zager 2018-05-16 19:15:40 UTC

https://brewweb.engineering.redhat.com/brew/buildinfo?buildID=695427

Comment 3 Zihan Tang 2018-05-22 06:05:30 UTC

Image is ready, change it to ON_QA.

Comment 4 Zihan Tang 2018-05-22 06:08:03 UTC

Verify failed.

Only 1 ERROR log changed to WARNING.  

the first line is still in ERROR.

[2018-05-22T05:55:56.41Z] [ERROR] - Could not find instance 16a03b07-5d71-11e8-8814-0a580a800004 associated with job state 23a48153-246d-4772-884a-d903f8a3fd88 - bundleinstances.automationbroker.io "16a03b07-5d71-11e8-8814-0a580a800004" not found
[2018-05-22T05:55:56.41Z] [WARNING] - unable to find job state: [16a03b07-5d71-11e8-8814-0a580a800004]. error: [bundleinstances.automationbroker.io "16a03b07-5d71-11e8-8814-0a580a800004" not found]

Comment 5 Shawn Hurley 2018-05-23 12:18:03 UTC

I don't think that we will be changing that log statement.

We will work on a better way to log these statements in the next release but we are not going to change this one. The reason being the DAO just has a single get method. This method will actually be in error state if you can not find the instance 9/10 times. We have no way to selectively turn off the log statement just of a single call of the method. 

@John is the above correct?

Comment 6 John Matthews 2018-05-24 19:18:32 UTC

For 3.10 we only intend to change the 1 message Shawn mentioned.

Open to considering an alternative approach to reduce the first ERROR message, if that is interesting then please file a new BZ for that message and we can consider for 3.11.

Comment 7 Zihan Tang 2018-05-29 09:56:18 UTC

Based on comment6 & comment 4 , mark this as Verified.
I'll open another bug for 3.11 to trace the other error message.