Bug 1421860 – [3.4] Metrics Resolution of Heapster Image Should be 30s to Match cAdvisor

Bugzilla will be upgraded to version 5.0. The upgrade date is tentatively scheduled for 2 December 2018, pending final testing and feedback.

Bug 1421860 - [3.4] Metrics Resolution of Heapster Image Should be 30s to Match cAdvisor

Summary:	[3.4] Metrics Resolution of Heapster Image Should be 30s to Match cAdvisor

Status:	CLOSED ERRATA

Aliases:	None

Product:	OpenShift Container Platform
Classification:	Red Hat
Component:	Hawkular (Show other bugs)
Sub Component:
Version:	3.4.1
Hardware:	Unspecified Unspecified

Priority	medium Severity medium
Target Milestone:	---
Target Release:	3.4.z
Assigned To:	Matt Wringe
QA Contact:	Peng Li
Docs Contact:

URL:
Whiteboard:
Keywords:

Depends On:	1421834
Blocks:
	Show dependency tree / graph

Reported:	2017-02-13 16:36 EST by Matt Wringe
Modified:	2017-03-06 11:39 EST (History)
CC List:	5 users (show)

See Also:
Fixed In Version:
Doc Type:	Enhancement
Doc Text:	The metrics heapster data collection resolution has been changed to 30s.
Story Points:	---
Clone Of:	1421834
Environment:
Last Closed:	2017-03-06 11:39:07 EST
Type:	Bug
Regression:	---
Mount Type:	---
Documentation:	---
CRM:
Verified Versions:
Category:	---
oVirt Team:	---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---

Attachments	(Terms of Use)
Add an attachment (proposed patch, testcase, etc.)

External Trackers
Tracker	ID	Priority	Status	Summary	Last Updated
Red Hat Product Errata	RHSA-2017:0448	normal	SHIPPED_LIVE	Important: ansible and openshift-ansible security and bug fix update	2017-03-06 16:36:25 EST

Groups: None (edit)

Description Matt Wringe 2017-02-13 16:36:31 EST

+++ This bug was initially created as a clone of Bug #1421834 +++

In Kubernetes, we specify a cAdvisor housekeeping interval of 10s.  cAdvisor has backoff logic that could move this up to 15s, plus jitter logic that could double that, leaving us with an actual max housekeeping interval of 30s:

https://github.com/kubernetes/kubernetes/blob/master/vendor/github.com/google/cadvisor/manager/container.go#L382.

Thus, our Heapster image needs to have a metrics_resolution of at least 30s
in order to avoid reading the same metrics batch twice from the node, which causes Heapster to skip rate calculation, which causes CPU metrics to not appear.

--- Additional comment from Matt Wringe on 2017-02-13 14:37:41 EST ---

Does this only need to be applied to 3.5? or is this logic also in older versions like 3.4?

Comment 1 Matt Wringe 2017-02-13 17:24:39 EST

The only PR we are waiting on is for ansible: https://github.com/openshift/openshift-ansible/pull/3350

Comment 5 Peng Li 2017-02-15 20:59:52 EST

verified with openshift-ansible branch release-1.4

Comment 7 errata-xmlrpc 2017-03-06 11:39:07 EST

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2017:0448

Note You need to log in before you can comment on or make changes to this bug.