Project

General

Profile

Actions
Copy link

Bug #15126

closed

Rename NIC1 to eth0 on CentOS testnodes

Added by David Galloway about 8 years ago. Updated about 8 years ago.

Status:
Won't Fix
Priority:
High
Assignee:
David Galloway
Category:
Test Node
Target version:
-
% Done:

0%

Source:
other
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Crash signature (v1):
Crash signature (v2):

Description

Freshly installed testnodes (mira and smithi) are left with their primary NIC in a down state at the end of an ansible run.

Presumably this is because there's ansible logic in place for Ubuntu machines to set eth0 via DHCP.

In RHEL/CentOS, network --bootproto=dhcp --device=eth0 --onboot=on in the kickstart has taken care of this.

New udev naming rules break this.

Related issues 1 (0 open — 1 closed)

Blocks sepia - Bug #15120: smithis with bad rpmdbsResolvedDavid Galloway03/14/2016

Actions
Actions
Copy link
 #1

Updated by David Galloway about 8 years ago

  • Blocks Bug #15120: smithis with bad rpmdbs added
Actions
Copy link
 #2

Updated by David Galloway about 8 years ago

  • Status changed from New to Won't Fix

I was mistaken. This is not the case.

What was happening was our DHCP server was handing out a 2 hour lease.

Our testnodes got kickstarted in EDT and then changed to UTC during the initial ansible run.

2 hours pass and the system never tries to renew its DHCP lease because the time it's expecting never happens.

Instead, we've opted to change our default DHCP lease time to 12 hours.

See https://bugzilla.redhat.com/show_bug.cgi?id=1093803

Actions
Copy link

Also available in: Atom PDF