Bug 491207

Summary:	Anaconda crashes trying to retrieve kickstart file via nfs
Product:	[Fedora] Fedora	Reporter:	Orion Poplawski <orion>
Component:	anaconda	Assignee:	David Cantrell <dcantrell>
Status:	CLOSED RAWHIDE	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	medium	Docs Contact:
Priority:	low
Version:	rawhide	CC:	pjones, rmaximo, vanmeeuwen+fedora
Target Milestone:	---
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2009-04-30 01:06:19 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 Orion Poplawski 2009-03-19 19:53:44 UTC

Description of problem:

Today's (20090319) rawhide.

This may be limited to this hardware as a similar install on a xen guest started okay.

loader received SIGSEGV! Backtrace:
loader(loadSegvHandler)
libc
loader(doPwMount)
loader(getFileFromNfs)
loader(getKickstartFile)
loader(main)

VT3:
INFO : file location: nfs:saga:/export/data1/ks/rawhide-laptop64.cfg

Pinging the machine while it came up, did get two responses before it crashed.
No record of mount attempt in nfs server log.

2 for 2 on this machine now.

Comment 1 Chris Lumens 2009-03-19 20:25:12 UTC

David's already got a line on this one.

Comment 2 Orion Poplawski 2009-03-19 21:17:56 UTC

I don't know if this is related, but in the xen install that booted okay, I go the following in my %post script:

+ mkdir -p /data/backup1
+ mount -v saga:/export/backup1 /data/backup1
mount: no type was given - I'll assume nfs because of the colon
mount.nfs: timeout set for Thu Mar 19 14:36:02 2009
mount.nfs: text-based options: 'addr=192.168.0.12'
saga:/export/backup1 on /data/backup1 type nfs (rw)
+ mkdir -p /nfs/local
+ mount -v earth:/export/local /nfs/local
mount: no type was given - I'll assume nfs because of the colon
mount.nfs: mount(2): Input/output error
mount.nfs: mount system call failed
mount.nfs: timeout set for Thu Mar 19 14:36:02 2009
mount.nfs: text-based options: 'addr=192.168.0.8'

Running system is capable of mounting earth:/export/local fine.

earth is CentOS 5.2, saga is Fedora 10.

Comment 3 Orion Poplawski 2009-03-20 14:47:31 UTC

Okay, the crash has been fixed.  I now see:

Error downloading kickstart file

Running tshark on the nfs server I see no traffic from the install machine.  I can ping it though.  Changing to FQDN does not help. 

VT3:

doing kickstart... setting it up
get_connection (2051): NetworkManager connected
url is saga:/export/data1/ks/rawhide-laptop64.cfg
file location: nfs:saga:/export/data1/ks/rawhide-laptop64.cfg
ERROR: failed to mount nfs source

Comment 4 David Cantrell 2009-04-30 01:06:19 UTC

With the latest nightly tree, I cannot reproduce this issue.  I am able to pass:

ks=nfs:HOSTNAME:/path/to/ks.cfg

The interface is brought up with dhcp, the kickstart file is downloaded, and installation begins.