Bug 349211

Summary: Anaconda hangs when trying to install F8T3/x86_64 on VMWare server 1.0.4
Product: [Fedora] Fedora Reporter: Gilboa Davara <gilboad>
Component: anacondaAssignee: Anaconda Maintenance Team <anaconda-maint-list>
Status: CLOSED CURRENTRELEASE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: low Docs Contact:
Priority: low    
Version: 8CC: katzj
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: f9 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2008-06-04 12:46:24 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:
Attachments:
Description Flags
Anaconda console log
none
Kernel log none

Description Gilboa Davara 2007-10-23 18:24:29 UTC 
Description of problem:
I'm trying to install F8T3/x86_64 on a 64bit VM under VMWare Server 1.0.4.
AFAICS the installation was pretty-much-strait forward.
However, when anaconda was trying to install the last packages, it hanged for hours.

Switching back to the text consoles doesn't really shows anything
critical (beyond a single syntax warning in Anaconda - see attached)
I'm keeping the VM image suspend, so please, if you require additional
information please let me ASAP.

Host configuration:
Fedoa 7, x86_64, 2.6.22.9-91.fc7.
2 x Opteron 275.
4GB RPM.
3 x 250GB SATA.
nVidia GF6800GT, nVidia binary drivers.

Guest configuration:
2 virtual CPUs.
10GB disk space.
512MB RAM.
2 bridged network devices.
2 host-only network devices.
USB support.
Sound support.
Comment 1 Gilboa Davara 2007-10-23 18:24:29 UTC 
Created attachment 235331 [details]
Anaconda console log
Comment 2 Gilboa Davara 2007-10-23 18:25:09 UTC 
Created attachment 235341 [details]
Kernel log
Comment 3 Chris Lumens 2007-10-23 20:05:56 UTC 
Looks like a python problem judging from the console log file.
Comment 4 James Antill 2007-10-23 20:57:28 UTC 
 Chris, if there was a double free problem in the python GC I'm pretty sure I'd
be flooded with bugs ... much more likely is that anaconda is calling some C
code that is doing the double free. Can you get glibc to print a
backtrace/core-dump like it normally does (post install) on double free?
Comment 5 Gilboa Davara 2007-10-24 00:40:32 UTC 
Even after I hard-wired /usr/bin/ls to /bin/ls and /usr/bin/sed to /bin/sed
(without it pstack aborted on an error) - pstack anaconda_pid returns nothing.

Beyond crashing anaconda by sending a kill -SEGV, any idea how I can yank the
call stack?

- Gilboa
Comment 6 James Antill 2007-10-24 03:50:50 UTC 
 Doing a kill -ABRT is probably the easiest, assuming nothing turns coredumps
off with resource limits ... chris what is the usual way to debug this easily?
Comment 7 Gilboa Davara 2007-10-24 06:20:00 UTC 
kill -ABRT doesn't return anything - anaconda just dies with no crash dump.
Reverting to last bad-known snapshot.

- Gilboa
Comment 8 Andy Lindeberg 2008-06-03 20:16:06 UTC 
Have you encountered this problem in Fedora 9?
Comment 9 Gilboa Davara 2008-06-04 09:12:31 UTC 
F9 installed cleanly.
Bug can be closed.

- Gilboa