PXEboot interface is left with IPv6 autoconf turned on after initial install, before ansible playbook execution
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
StarlingX |
Fix Released
|
Low
|
Andre Kantek |
Bug Description
Brief Description
During tests of B&R it was observed that the system was keeping IPv6 autoconf turned on in controller-0 after the PXEboot installation
This happened because the kickstart script is setting the loopback instead of the interface
root@yow-
auto eno1
iface eno1 inet dhcp
post-up echo 0 > /proc/sys/
Severity
<Minor: System/Feature is usable with minor issue>
Steps to Reproduce
Install the system without this correction and observe if an IPv6 address is not added to the system. E.g.
root@yow-
lo UNKNOWN 127.0.0.1/8 ::1/128
eno1 UP 128.224.151.215/23 2620:10a:
Observe that the autoconfigured IPv6 (2620:10a:
Expected Behavior
There should not be autoconfigured addresses in the PXE boot after installation
Actual Behavior
There are autoconfigured addresses in the PXE boot after installation
Reproducibility
Reproducible, if the switch connected to the pxe interface is replying the autoconf data
System Configuration
AIO-DX or AIO-SX
Load info (eg: 2022-03-
Latest StarlingX
Test Activity
Regression Testing
Workaround
1) Remove the offending IPv6 address and
2) set the desired interface with autoconf OFF
echo 0 > /proc/sys/
3) proceed with the ansible playbook restore configuration
description: | updated |
Changed in starlingx: | |
importance: | Undecided → Low |
tags: | added: stx.9.0 stx.networking |
Changed in starlingx: | |
assignee: | nobody → Andre Kantek (akantek) |
Fix proposed to branch: master /review. opendev. org/c/starlingx /metal/ +/900937
Review: https:/