Bug 985843

The exception in the screenshot is:
AttributeError: 'NoneType' object has no attribute 'unpack'

I finally installed it, but the list of problems is pretty long - first of all - NetworkManager is not started properly - and it causes anaconda failure. 

The only way to get F20 installed was to create f19 image, add rawhide repos, update, and then cycle service NetworkManager start/stop a couple of times until it started working.

(In reply to Krzysztof Daniel from comment #0)
...
> I downloaded a boot.iso from
> http://dl.fedoraproject.org/pub/fedora/linux/development/rawhide/x86_64/os/
> images/
...

That link returns "404 Not Found" now ...

If you can switch to the installer shell console (ctrl-alt-f2), the anaconda log files are in /tmp. If there is an anaconda-tb-* file in /tmp, could you attach it?

No such file: just
anaconda, packaging, storage, ifcfg, program, all with .log extension,  but nothing interesting there.

How are you starting the installer? text mode?

In graphical mode, you should be getting an exception reporting dialog that lets you report a bug. Can you reproduce the exception in graphical mode?

I start empty VM image with boot.iso as a cdrom. I select 'Install Fedora rawhide'. It looks like GRUB. Then I get green [OK] messages, and then

Starting installer, one moment.
Then the screenshot follows.

OK, thanks. I'm not sure what is going on, then. Maybe the kernel command line has "text" on it by default in that image, or the exception occurs before X is started. Anyway, if you can reproduce this with an image that is online, please post a comment.

I also used http://koji.fedoraproject.org/koji/taskinfo?taskID=5639955.

Steps to reproduce are:
* start live image
* start command line, type liveinst -d

The same effect, the same stack trace as on the screenshot.

Starting installer, one moment...
/usr/lib64/python2.7/site-packages/gi/overrides/GLib.py:53: PyGIDeprecationWarning: threads_init longer needs to be called. See: https://bugzilla.gnome.org/show_bug.cgi?id=686914
  PyGIDeprecationWarning)
anaconda 20.2-1 for Fedora 20 (pre-release) started.
Traceback (most recent call last):
  File "/sbin/anaconda", line 968, in <module>
    setupDisplay(anaconda, opts, addon_paths)
  File "/sbin/anaconda", line 483, in setupDisplay
    if not nm_is_connecting() and not nm_is_connected():
  File "/usr/lib64/python2.7/site-packages/pyanaconda/nm.py", line 96, in nm_is_connecting
    return nm_state() == NetworkManager.State.CONNECTING
  File "/usr/lib64/python2.7/site-packages/pyanaconda/nm.py", line 82, in nm_state
    return proxy.get_cached_property("State").unpack()


Only after executing this series of commands:
[liveuser@localhost ~]$ su
[root@localhost liveuser]# service NetworkManager stop
Redirecting to /bin/systemctl stop  NetworkManager.service
^C
[root@localhost liveuser]# service NetworkManager start
Redirecting to /bin/systemctl start  NetworkManager.service
Job for NetworkManager.service canceled.
Job for NetworkManager.service canceled.
[root@localhost liveuser]# 

Anaconda was able to start (and after a few tries I was able to complete wizards, but never to actually install anything).

In general, the installer is *very* slow (RHEL and F19 VMs work fine) and it takes tens of seconds between a click and a reaction.

Thanks, Krzysztof. I didn't think of trying a nightly:
http://alt.fedoraproject.org/pub/alt/nightly-composes/

Confirming your report with:
$ qemu-kvm -m 4096 -hda f19-test-3.img -cdrom ~/xfr/fedora/nightly-composes/Fedora-20-Nightly-x86_64-Live-xfce-20130722.09-1.iso -vga std -boot menu=on

After the Live image boots, there is no ip address assigned, yet NetworkManager is running. Restarting NetworkManager hangs. After killing the restart, a second restart succeeds. This all looks like a systemd/NetworkManager/LiveCD problem and not an anaconda problem.

[liveuser@localhost ~]$ sudo su

[root@localhost liveuser]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: ens3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
    link/ether 52:54:00:12:34:56 brd ff:ff:ff:ff:ff:ff

[root@localhost liveuser]# systemctl status NetworkManager
...

[root@localhost liveuser]# systemctl restart NetworkManager

^C
[root@localhost liveuser]# systemctl status NetworkManager
...

[root@localhost liveuser]# systemctl restart NetworkManager

[root@localhost liveuser]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: ens3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 52:54:00:12:34:56 brd ff:ff:ff:ff:ff:ff
    inet 10.0.2.15/24 brd 10.0.2.255 scope global ens3
       valid_lft forever preferred_lft forever
    inet6 fe80::5054:ff:fe12:3456/64 scope link 
       valid_lft forever preferred_lft forever

[root@localhost liveuser]# systemctl status NetworkManager
...

(In reply to Steve Tyler from comment #10)
...
> ... This all looks like a
> systemd/NetworkManager/LiveCD problem and not an anaconda problem.
...

This sounds identical:
Bug 985627 - NM fails to bring up any interfaces at first boot on Rawhide since kernel-3.11.0-0.rc0.git7.1.fc20.x86_64

Well, not really identical. I clearly identified separate problems and filed separate bugs. This report is a mess.

See also:

https://bugzilla.redhat.com/show_bug.cgi?id=986068
https://bugzilla.redhat.com/show_bug.cgi?id=986069

*** This bug has been marked as a duplicate of bug 985627 ***

(In reply to Adam Williamson from comment #12)
> Well, not really identical. I clearly identified separate problems and filed
> separate bugs. This report is a mess.
Ain't you special, Adam.
The analysis clearly evolved in this bug ...

Or, as I put it, "this report is a mess". :)