Bug 1498213 – Increase ARP cache size on loadbalancers

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

Bug 1498213 - Increase ARP cache size on loadbalancers

Summary:	Increase ARP cache size on loadbalancers

Status:	CLOSED ERRATA

Aliases:	None

Product:	OpenShift Container Platform
Classification:	Red Hat
Component:	Installer (Show other bugs)
Sub Component:
Version:	3.7.0
Hardware:	Unspecified Unspecified

Priority	unspecified Severity medium
Target Milestone:	---
Target Release:	3.7.0
Assigned To:	jmencak
QA Contact:	Johnny Liu
Docs Contact:

URL:
Whiteboard:	aos-scalability-37
Keywords:	NeedsTestCase

Depends On:
Blocks:
	Show dependency tree / graph

Reported:	2017-10-03 14:21 EDT by jmencak
Modified:	2017-12-12 01:03 EST (History)
CC List:	4 users (show)

See Also:
Fixed In Version:
Doc Type:	Known Issue
Doc Text:	This is a known issue as openshift tuned profiles were never set on RHEL Atomic Host, only RHEL.
Story Points:	---
Clone Of:
Environment:
Last Closed:	2017-11-28 17:14:33 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-2017:3188	normal	SHIPPED_LIVE	Moderate: Red Hat OpenShift Container Platform 3.7 security, bug, and enhancement update	2017-11-28 21:34:54 EST

Groups: None (edit)

Description jmencak 2017-10-03 14:21:16 EDT

Description of problem:
On RHEL Atomic Host the ARP garbage collection thresholds are too low causing problems with OCP HA deployments with 1k+ nodes.

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

How reproducible:
Always

Steps to Reproduce:
1. Install RHEL Atomic Host OCP HA cluster with a loadbalancer and query sysctl values for net.ipv[46].neigh.default.gc_thresh[1-3]

Actual results:
net.ipv4.neigh.default.gc_thresh1 = 128
net.ipv4.neigh.default.gc_thresh2 = 512
net.ipv4.neigh.default.gc_thresh3 = 1024
net.ipv6.neigh.default.gc_thresh1 = 128
net.ipv6.neigh.default.gc_thresh2 = 512
net.ipv6.neigh.default.gc_thresh3 = 1024

Expected results:
net.ipv4.neigh.default.gc_thresh1 = 8192
net.ipv4.neigh.default.gc_thresh2 = 32768
net.ipv4.neigh.default.gc_thresh3 = 65536
net.ipv6.neigh.default.gc_thresh1 = 8192
net.ipv6.neigh.default.gc_thresh2 = 32768
net.ipv6.neigh.default.gc_thresh3 = 65536

Additional info:
https://github.com/openshift/openshift-ansible/pull/5645

Comment 1 Johnny Liu 2017-10-13 01:41:43 EDT

Verified this bug with openshift-ansible-3.7.0-0.148.0.git.0.b35eb14.el7.noarch, and PASS.

After installation, go to check:
on LB host:
# sysctl -a |grep "neigh.default.gc_thresh"
net.ipv4.neigh.default.gc_thresh1 = 8192
net.ipv4.neigh.default.gc_thresh2 = 32768
net.ipv4.neigh.default.gc_thresh3 = 65536
net.ipv6.neigh.default.gc_thresh1 = 8192
net.ipv6.neigh.default.gc_thresh2 = 32768
net.ipv6.neigh.default.gc_thresh3 = 65536


On node host:
#  sysctl -a |grep "neigh.default.gc_thresh"
net.ipv4.neigh.default.gc_thresh1 = 8192
net.ipv4.neigh.default.gc_thresh2 = 32768
net.ipv4.neigh.default.gc_thresh3 = 65536
net.ipv6.neigh.default.gc_thresh1 = 8192
net.ipv6.neigh.default.gc_thresh2 = 32768
net.ipv6.neigh.default.gc_thresh3 = 65536

Comment 4 errata-xmlrpc 2017-11-28 17:14:33 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-2017:3188

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