Bug 441598 - Install crashes with traceback - The name org.freedesktop.Hal was not provided by any .service files
Summary: Install crashes with traceback - The name org.freedesktop.Hal was not provide...
Alias: None
Product: Fedora
Classification: Fedora
Component: anaconda
Version: rawhide
Hardware: All
OS: Linux
Target Milestone: ---
Assignee: Anaconda Maintenance Team
QA Contact: Fedora Extras Quality Assurance
Depends On:
Blocks: F9Blocker
TreeView+ depends on / blocked
Reported: 2008-04-08 21:43 UTC by Orion Poplawski
Modified: 2008-04-11 18:02 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Last Closed: 2008-04-11 18:02:03 UTC

Attachments (Terms of Use)
ps -fe (3.70 KB, text/plain)
2008-04-09 16:35 UTC, Orion Poplawski
no flags Details

Description Orion Poplawski 2008-04-08 21:43:02 UTC
Description of problem:

Trying pxe/kickstart text mode install of today's rawhide.  Worked on one
laptop, on another I get:

File "/usr/lib/python2.5/site-packages/dbus/connection.py", line 607, in
  message, timeout)
dbus.exceptions.DBusException: org.freedesktop.DBus.Error.ServiceUnknown: The
anme org.freedesktop.Hal was not provided by any .service files

Main difference is that I'm trying to re-use the /boot, swap, and /export
partitions from an existing install.

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

Comment 1 Orion Poplawski 2008-04-08 21:47:21 UTC
Hmm, removed a CD-ROM disc that was in the machine and it seems to get past that

Comment 2 Jeremy Katz 2008-04-09 00:35:35 UTC
If you can reproduce, does it look like hal is running when you get the traceback?

Comment 3 Orion Poplawski 2008-04-09 02:17:57 UTC
Very reproducible.  hald has been running the whole time.

Comment 4 Orion Poplawski 2008-04-09 02:19:13 UTC
Gah, sorry - got this mixed up with another bug.  I'll see if I can check - but
it crashes very quickly...

Comment 5 Orion Poplawski 2008-04-09 16:35:25 UTC
Created attachment 301844 [details]
ps -fe

Okay, with nokill I can still use VT2.	hald is definitely running.  Perhaps it
is taking too long to come up pinging the CDROM?

Comment 6 Jeremy Katz 2008-04-09 19:22:38 UTC
Is the cd being used for the kickstart as a source?  Also, where is the dbus
call being made (ie, what's the rest of the traceback)

Comment 7 Orion Poplawski 2008-04-09 19:42:15 UTC
The cd just happened to be in the drive (though it is a Fedora CD).  I'm doing a
PXE boot, http install.  Rest of the traceback scrolls off the screen.  Last
item in anaconda.log is:

09:58:11 INFO    : Display mode = t
09:58:11 INFO    : Method =

Comment 8 Jeremy Katz 2008-04-09 19:57:09 UTC
The only callers of it that I see in a quick grep are wrapped in try/except. 
And shift-pgup should work to scrollback and see more of the traceback.

Comment 9 Orion Poplawski 2008-04-09 20:18:01 UTC
Ah, never knew about shift-pgup.  cool.

/usr/bin/anaconda line 869 in probeHW
                  line 456 in xserver.probeHW
 rhpxl/xserver.py line 84
 rhpxl/videocard.py line 126

Comment 10 Jeremy Katz 2008-04-10 01:32:34 UTC
Hrmm -- definitely should be running at that point.  Sticking on the blocker
list so that we can not lose track of this

Comment 11 Jerry Vonau 2008-04-10 01:45:27 UTC
Cool, I thought it was just me.... I saw this earlier with also. 
I also saw after the above:

/dbus/bus.py line 244 get_object
/dbus/proxies.py line 241 __init__
/dbus/bus.py line 183 activate_name_owner
/dbus/bus.py line 281 start_service_by_name

then /dbus/connection.py", line 607 

hope it helps.

Comment 12 Jerry Vonau 2008-04-11 14:25:33 UTC
fixed with for me.

Comment 13 Orion Poplawski 2008-04-11 18:02:03 UTC
And for me.. Looks like the new rhpxl package fixed it.

Note You need to log in before you can comment on or make changes to this bug.