Bug 468456

Summary:	Fedora 10 Beta 1 won't netboot due to NetworkManager
Product:	[Fedora] Fedora	Reporter:	Jon Wallace <jon.wallace>
Component:	anaconda	Assignee:	Anaconda Maintenance Team <anaconda-maint-list>
Status:	CLOSED DUPLICATE	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	urgent	Docs Contact:
Priority:	medium
Version:	rawhide	CC:	anaconda-maint-list, dcantrell, dcbw, hdegoede, wtogami
Target Milestone:	---
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2008-10-25 08:11:20 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Jon Wallace 2008-10-24 19:52:10 UTC

Description of problem:
Trying to netboot FC 10 Beta 1 but it won't get past NetworkManager

Version-Release number of selected component (if applicable):
FC10 Beta 1

How reproducible:
Try netbooting it.

Steps to Reproduce:
1. Setup PXEboot on server
2. Boot system
3. Halts at NetworkManager
  
Actual results:
FATAL: Module loop not found.
FATAL: Module ext3 not found.
INFO: getting kickstart file
INFO: sending dhcp request through device eth0
ERROR: failed to start NetworkManager (6)

Expected results:


Additional info:

Comment 1 Dan Williams 2008-10-24 21:27:25 UTC

David, what sort of debugging info can we get here?

Comment 2 David Cantrell 2008-10-24 22:08:53 UTC

We launch NetworkManager so it uses tty4 for output, so messages should be over there.

The (6) is also a clue in the failed to start NetworkManager message.  That's the return code from get_connection() in loader.  Return code 6 means we got nothing back from dbus_connection_send_with_reply_and_block() in our wait loop that checks the state for NM_STATE_CONNECTED.  Before get_connection() returns 6, it logs that error to the anaconda log facility, so you would see that message on tty3 as well as whatever NM printed to tty4.

I notice that those errors are getting printed out only in DEBUG level, so make sure you start the installer and pass:

    loglevel=debug

At the boot: prompt.

Comment 3 Hans de Goede 2008-10-25 08:11:20 UTC

I think this is a dup of 462083, I'm pretty sure actually. This was fixed some time ago, but after the beta. Please try using the pxeboot images from the development (rawhide) tree (and point the installer to that same tree over http for the rest of the install).

*** This bug has been marked as a duplicate of bug 462083 ***