Updating report + questions

[nimnull22]

Dear All.

As long as now I can spend in front of Fedora 14 more then 15 min, because I removed DRI2 from xorg and it makes screen picture to be nice for eyes, I would like to share my thoughts about last update.
So kernel 2.6.34-11 for i686 - fine, works - no errors in dmesg so far.
Xfce modeles - works, I use xfce so I tested it.
Skype starts, and connects without any errors.
StarDict QT4 works.
Pidgin works, tray icon is blinking on new message.

Now some remarks.
"Folder" icon picture in Xfce - biggest size everywhere.

I use keyboard layouts plugin (xfce4-xkb-plugin) and Slim. So when I boot, keyboard plugin aware about second layout and toggle key combination, but if I Logout/Login I lose toggle option.

And, of course, DRI2 - it doesn't work on my
00:02.0 VGA compatible controller: Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller (rev 04). It spoils the screen picture to unusable condition. Without it loaded "glxgears" rotates smooth and shows more FPS then dri2 is loaded. (112/20)

Thanks.

[GoinEasy9]

I have similar results with the icons on Chrome, very big rectangles. Considering it's rawhide, a lot more could be going on ... and isn't. Gnome icons seem to be fine, SELinux icon disappeared like it did on F13.

All in all though, it is running pretty stable except for minor annoyances.

[nimnull22]

Skype tested. UVC video works, sound works microphone works. Tested under "sudo" on different user. No errors.





Got some errors today:

---> Package cjkuni-uming-fonts.noarch 0:0.2.20080216.1-37.fc14 set to be updated
--> Processing Dependency: cjkuni-fonts-common = 0.2.20080216.1-37.fc14 for package: cjkuni-uming-fonts-0.2.20080216.1-37.fc14.noarch
Error: Package: cjkuni-uming-fonts-0.2.20080216.1-37.fc14.noarch (rawhide)
Requires: cjkuni-fonts-common = 0.2.20080216.1-37.fc14
Installed: cjkuni-fonts-common-0.2.20080216.1-37.fc13.noarch (installed)
cjkuni-fonts-common = 0.2.20080216.1-37.fc13

It looks like there is no cjkuni-fonts-common = 0.2.20080216.1-37.fc14 on rawhide.

Thanks.

[dd_wizard]

There isn't one for F13, either. I'm sure it'll get built soon, but --exclude=cjkuni-* works for now.

dd_wizard

[nimnull22]

In addition today:
Error: Package: ekiga-3.2.6-3.fc14.i686 (@rawhide/13)
Requires: libedataserver-1.2.so.12
Removing: evolution-data-server-2.31.1-1.fc14.i686 (@rawhide/13)
libedataserver-1.2.so.12
Updated By: evolution-data-server-2.31.2-1.fc14.i686 (rawhide)
Not found
Error: Package: gnome-panel-2.30.0-2.fc14.i686 (@rawhide/13)
Requires: libedataserver-1.2.so.12
Removing: evolution-data-server-2.31.1-1.fc14.i686 (@rawhide/13)
libedataserver-1.2.so.12
Updated By: evolution-data-server-2.31.2-1.fc14.i686 (rawhide)
Not found
Error: Package: empathy-2.31.1-1.fc14.i686 (@rawhide/13)
Requires: libedataserver-1.2.so.12
Removing: evolution-data-server-2.31.1-1.fc14.i686 (@rawhide/13)
libedataserver-1.2.so.12
Updated By: evolution-data-server-2.31.2-1.fc14.i686 (rawhide)
Not found
Error: Package: cjkuni-uming-fonts-0.2.20080216.1-37.fc14.noarch (rawhide)
Requires: cjkuni-fonts-common = 0.2.20080216.1-37.fc14
Installed: cjkuni-fonts-common-0.2.20080216.1-37.fc13.noarch (installed)
cjkuni-fonts-common = 0.2.20080216.1-37.fc13
Error: Package: nautilus-sendto-2.28.4-2.fc14.i686 (@rawhide/13)
Requires: libedataserver-1.2.so.12
Removing: evolution-data-server-2.31.1-1.fc14.i686 (@rawhide/13)
libedataserver-1.2.so.12
Updated By: evolution-data-server-2.31.2-1.fc14.i686 (rawhide)
Not found

[nimnull22]

for 26.05.2010
still errors:
--> Running transaction check
--> Processing Dependency: libedataserver-1.2.so.12 for package: ekiga-3.2.6-3.fc14.i686
--> Processing Dependency: libedataserver-1.2.so.12 for package: gnome-panel-2.30.0-2.fc14.i686
--> Processing Dependency: libedataserver-1.2.so.12 for package: nautilus-sendto-2.28.4-2.fc14.i686
--> Processing Dependency: libedataserver-1.2.so.12 for package: empathy-2.31.1-1.fc14.i686
---> Package evolution-data-server.i686 0:2.31.2-2.fc14 set to be updated
--> Finished Dependency Resolution
Error: Package: gnome-panel-2.30.0-2.fc14.i686 (@rawhide/13)
Requires: libedataserver-1.2.so.12
Removing: evolution-data-server-2.31.1-1.fc14.i686 (@rawhide/13)
libedataserver-1.2.so.12
Updated By: evolution-data-server-2.31.2-2.fc14.i686 (rawhide)
Not found
Error: Package: ekiga-3.2.6-3.fc14.i686 (@rawhide/13)
Requires: libedataserver-1.2.so.12
Removing: evolution-data-server-2.31.1-1.fc14.i686 (@rawhide/13)
libedataserver-1.2.so.12
Updated By: evolution-data-server-2.31.2-2.fc14.i686 (rawhide)
Not found
Error: Package: nautilus-sendto-2.28.4-2.fc14.i686 (@rawhide/13)
Requires: libedataserver-1.2.so.12
Removing: evolution-data-server-2.31.1-1.fc14.i686 (@rawhide/13)
libedataserver-1.2.so.12
Updated By: evolution-data-server-2.31.2-2.fc14.i686 (rawhide)
Not found
Error: Package: empathy-2.31.1-1.fc14.i686 (@rawhide/13)
Requires: libedataserver-1.2.so.12
Removing: evolution-data-server-2.31.1-1.fc14.i686 (@rawhide/13)
libedataserver-1.2.so.12
Updated By: evolution-data-server-2.31.2-2.fc14.i686 (rawhide)
Not found


After update - icon size is normal. Solved.


I can't start "SELinux Troubleshooter":
Opps, sealert hit an error!

Traceback (most recent call last):
File "/usr/bin/sealert", line 978, in <module>
run_as_dbus_service(username)
File "/usr/bin/sealert", line 99, in run_as_dbus_service
app = SEAlert(user, dbus_service.presentation_manager, watch_setroubleshootd=True)
File "/usr/bin/sealert", line 599, in __init__
from setroubleshoot.browser import BrowserApplet
File "/usr/lib/python2.6/site-packages/setroubleshoot/browser.py", line 11, in <module>
import gtkhtml2
ImportError: No module named gi

Need to resolve this for dhcpcd:
type=1400 audit(1274896575.016:12876): avc: denied { write } for pid=2772 comm="dhcpcd" scontext=unconfined_u:system_r:dhcpc_t:s0-s0:c0.c1023 tcontext=system_u:object_r:sysctl_net_t:s0 tclass=file
type=1300 audit(1274896575.016:12876): arch=40000003 syscall=5 success=no exit=-13 a0=bfd06050 a1=241 a2=1b6 a3=8059986 items=0 ppid=1314 pid=2772 auid=500 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=pts0 ses=1 comm="dhcpcd" exe="/sbin/dhcpcd" subj=unconfined_u:system_r:dhcpc_t:s0-s0:c0.c1023 key=(null)


Thanks

[GoinEasy9]

I filed a BZ for the SELinux Troubleshooter, --skip-broken worked around evolution and evolution-data-server. There was a abrt/SELinux conflict that I got working installing a new selinux-policy version from koji. All in all, rawhide has been pretty stable.

[dd_wizard]

This has been around for a couple of days, too:[gene@Mobile-PC ~]$ sudo yum --disablerepo=rawhide update
Loaded plugins: fastestmirror, presto, refresh-packagekit, remove-with-leaves, show-leaves,
: verify
Loading mirror speeds from cached hostfile
* livna: rpm.livna.org
* rpmfusion-free-rawhide: mirror.web-ster.com
* rpmfusion-nonfree-rawhide: mirror.web-ster.com
Setting up Update Process
Resolving Dependencies
--> Running transaction check
---> Package ffmpeg.x86_64 0:0.6-0.3.20100429svn.fc14 set to be updated
---> Package ffmpeg-libs.x86_64 0:0.6-0.3.20100429svn.fc14 set to be updated
---> Package libquicktime.x86_64 0:1.1.5-1.fc14 set to be updated
---> Package live555.x86_64 0:0-0.27.2010.04.09.fc14 set to be updated
---> Package mplayer.x86_64 0:1.0-0.117.20100429svn.fc14 set to be updated
---> Package mplayer-common.x86_64 0:1.0-0.117.20100429svn.fc14 set to be updated
---> Package smplayer.x86_64 0:0.6.9-2.fc14 set to be updated
--> Processing Dependency: libx264.so.83()(64bit) for package: gstreamer-plugins-ugly-0.10.14-1.fc13.x86_64
---> Package x264-libs.x86_64 0:0.0.0-0.27.20100429gitd9db8b3.fc14 set to be updated
--> Finished Dependency Resolution
Error: Package: gstreamer-plugins-ugly-0.10.14-1.fc13.x86_64 (@rpmfusion-free-rawhide/13)
Requires: libx264.so.83()(64bit)
Removing: x264-libs-0.0.0-0.27.20100130git3659b81.fc13.x86_64 (@rpmfusion-free-updates/13)
libx264.so.83()(64bit)
Updated By: x264-libs-0.0.0-0.27.20100429gitd9db8b3.fc14.x86_64 (rpmfusion-free-rawhide)
Not found
You could try using --skip-broken to work around the problem
You could try running: rpm -Va --nofiles --nodigest

I filed a bug at RPMFusion about it yesterday, but it's not resolved yet. It looks like a couple of packages need to be rebuilt for F14.

dd_wizard

---------- Post added at 01:21 PM CDT ---------- Previous post was at 11:31 AM CDT ----------

@nimnull22: Did you post a bugzilla report about SELinux? I'm getting the same oops here.

dd_wizard

[nimnull22]

@nimnull22: Did you post a bugzilla report about SELinux? I'm getting the same oops here.

dd_wizard

Not yet, but I can.
Yesterday, GoinEasy9 said: "I filed a BZ for the SELinux Troubleshooter". If BZ=bugzilla, so he did it.

[dd_wizard]

Oops, he did! :)

dd_wizard

[nimnull22]

system-config-users has some problems.
First:
/usr/lib/python2.6/site-packages/gtk-2.0/gtk/__init__.py:40: RuntimeWarning: tp_compare didn't return -1 or -2 for exception
from gtk import _gtk
ImportError: No module named gi
Traceback (most recent call last):
File "/usr/share/system-config-users/mainWindow.py", line 856, in on_assignHighestUidCheckButton_toggled
if self.assignHighestUidCheckButton.get_active () == True:
ImportError: No module named gi

Second, there is no option for creation a "system user".

[GoinEasy9]

ImportError: No module named gi

Looks like the same problem that's going on with SELinux Troubleshooter.

[dd_wizard]

Interesting, that's how Exaile dies, too. I reported the bug against Exaile, but that's obviously not the culprit. Any ideas which package to report it against?

dd_wizard

---------- Post added at 04:39 PM CDT ---------- Previous post was at 04:31 PM CDT ----------

This the exaile bug report (https://bugzilla.redhat.com/show_bug.cgi?id=596571). Does it look a lot like your SE Linux bug, GoinEasy9?

dd_wizard

[GoinEasy9]

I filed under https://bugzilla.redhat.com/show_bug.cgi?id=596572 but it was marked as a duplicate of https://bugzilla.redhat.com/show_bug.cgi?id=596392 - I would look under the second BZ to see if that's closer to what Exaile is doing.

I'm going to try the pygobject2 from koji to see if that fixes the problem, was mentioned at the end of 596392.

[dd_wizard]

From the second link, it looks like a fix is in koji now. There's a new pygobject2 package in the pipeline. :)

dd_wizard

---------- Post added at 04:54 PM CDT ---------- Previous post was at 04:46 PM CDT ----------

Yup, pygobject2 updated to the latest koji and stuff works again! :D

dd_wizard

[GoinEasy9]

I did the daily update after I updated pygoobject2, and yes, your correct, everything works again ... LOL ... until tomorrow.

I updated BZ with the success story. It's like patting the devs on the head.

[dd_wizard]

I noticed that nautilus-sendto was rebuilt against the newest evolution-data-server, but gnome-panel hasn't. So that's still broken. :(

dd_wizard

[GoinEasy9]

evolution.i686 2.31.2-1.fc14 rawhide
evolution-data-server.i686 2.31.2-2.fc14 rawhide
nautilus-sendto.i686 2.28.4-3.fc14 rawhide

Those are the broken ones I come up with.

One of these days I'm going to have to start using evolution, so, I can complain when it's broken ... I feel so left out.

[nimnull22]

After today's update selinux works as usual. Thanks.

I have a question. How often intel drivers are updated? For instance i915, xorg's intel?
I have found a lot of patches on [intel-glx] mailing list, and especially the last one:
"[Intel-gfx] [PATCH] drm/i915: Honor sync polarity from VBT panel timing descriptors. I'm actually kind of shocked that it works at all otherwise."

So when we can see them ported to rawhide?

Thanks

[dd_wizard]

I'm not sure what the time-line will be, but you can check the xorg.x11-drv-intel (http://koji.fedoraproject.org/koji/packageinfo?packageID=7794) page for new versions of the driver. Follow the link to the latest fc14 build, and scroll down to the Changelog to see what's been happening.

dd_wizard

[GoinEasy9]

Tonight's updates dependency problems:
evolution.i686 2.31.2-1.fc14 rawhide
evolution-data-server.i686 2.31.2-2.fc14 rawhide
gnome-panel.i686 2.31.2-1.fc14 rawhide
gnome-panel-libs.i686 2.31.2-1.fc14 rawhide
nautilus-sendto.i686 2.28.4-3.fc14 rawhide

Some messages:
Updating : evince-2.31.2-1.fc14.i686 40/172
I/O warning : failed to load external entity "/etc/gconf/schemas/evince.schemas"
Failed to open `/etc/gconf/schemas/evince.schemas': No such file or directory
Updating : selinux-policy-targeted-3.8.1-2.fc14.noarch 59/172
libsemanage.semanage_install_active: setfiles returned error code 1.
semodule: Failed!

But all in all no problems, SELinux message had me worried, but, everything seems to be working.

[kyryder]

Some messages:
Updating : evince-2.31.2-1.fc14.i686 40/172
I/O warning : failed to load external entity "/etc/gconf/schemas/evince.schemas"
Failed to open `/etc/gconf/schemas/evince.schemas': No such file or directory
Updating : selinux-policy-targeted-3.8.1-2.fc14.noarch 59/172
libsemanage.semanage_install_active: setfiles returned error code 1.
semodule: Failed!

But all in all no problems, SELinux message had me worried, but, everything seems to be working.

I had that same evince message and dependency problems , but after grabbing the packages that are in your list of Tonight's updates dependency problems + a lib or two all is well from the koji mountain. Now I just have rpmfusion problems: :cool:
Packages skipped because of dependency problems:
ffmpeg-libs-0.6-0.3.20100429svn.fc14.i686 from rpmfusion-free-rawhide
libquicktime-1.1.5-1.fc14.i686 from rpmfusion-free-rawhide
mencoder-1.0-0.117.20100429svn.fc14.i686 from rpmfusion-free-rawhide
mplayer-1.0-0.117.20100429svn.fc14.i686 from rpmfusion-free-rawhide
mplayer-common-1.0-0.117.20100429svn.fc14.i686 from rpmfusion-free-rawhide
x264-libs-0.0.0-0.27.20100429gitd9db8b3.fc14.i686 from rpmfusion-free-rawhide

Ky

[nimnull22]

To be clear. Update 29.05.10 16:52 UTC

Updated:
ekiga.i686 0:3.2.6-4.fc14 empathy.i686 0:2.31.2-1.fc14
evolution-data-server.i686 0:2.31.2-2.fc14 gnome-icon-theme-extras.noarch 0:2.30.1-2.fc14
gnome-panel.i686 0:2.31.2-1.fc14 gnome-panel-libs.i686 0:2.31.2-1.fc14
ibus.i686 0:1.3.4-2.fc14 ibus-gtk.i686 0:1.3.4-2.fc14
ibus-libs.i686 0:1.3.4-2.fc14 libdrm.i686 0:2.4.21-0.1.fc14
linux-atm-libs.i686 0:2.5.1-1.fc14 mousetweaks.i686 0:2.31.2-1.fc14
nautilus-sendto.i686 0:2.28.4-3.fc14 nfs-utils-lib.i686 0:1.1.5-2.fc14
openldap.i686 0:2.4.22-2.fc14 orca.i686 0:2.31.2-1.fc14
pkgconfig.i686 1:0.25-1.fc14 selinux-policy.noarch 0:3.8.1-3.fc14
selinux-policy-targeted.noarch 0:3.8.1-3.fc14 sos.noarch 0:2.2-0.fc13
usbmuxd.i686 0:1.0.4-3.fc14 vim-common.i686 2:7.2.440-1.fc14
vim-enhanced.i686 2:7.2.440-1.fc14 vim-minimal.i686 2:7.2.440-1.fc14
webkitgtk.i686 0:1.3.1-1.fc14

Complete!
No problems here - everything was updated.



But, at the boot time I got:

...
SELinux: initialized (dev rootfs, type rootfs), uses genfs_contexts
SELinux: initialized (dev sysfs, type sysfs), uses genfs_contexts
type=1403 audit(1275152052.633:3): policy loaded auid=4294967295 ses=4294967295
dracut: Switching root
type=1403 audit(1275152052.633:3): policy loaded auid=4294967295 ses=4294967295
dracut: Switching root
type=1400 audit(1275152053.625:4): avc: denied { read write } for pid=293 comm="hostname" path="/dev/console" dev=devtmpfs ino=5566 scontext=system_u:system_r:hostname_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275152053.634:5): avc: denied { read write } for pid=293 comm="hostname" path="/dev/console" dev=devtmpfs ino=5566 scontext=system_u:system_r:hostname_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275152053.715:6): avc: denied { read write } for pid=296 comm="consoletype" path="/dev/null" dev=devtmpfs ino=4053 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275152053.721:7): avc: denied { read write } for pid=297 comm="consoletype" path="/dev/console" dev=devtmpfs ino=5566 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275152053.731:8): avc: denied { read write } for pid=297 comm="consoletype" path="/dev/console" dev=devtmpfs ino=5566 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275152053.744:9): avc: denied { read write } for pid=296 comm="consoletype" path="/dev/null" dev=devtmpfs ino=4053 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275152053.965:10): avc: denied { open } for pid=298 comm="mount" name="null" dev=devtmpfs ino=4053 scontext=system_u:system_r:mount_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275152053.976:11): avc: denied { read write } for pid=300 comm="consoletype" path="/dev/console" dev=devtmpfs ino=5566 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275152053.983:12): avc: denied { read write } for pid=300 comm="consoletype" path="/dev/console" dev=devtmpfs ino=5566 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file

Thanks.

[GoinEasy9]

I only did a restart last night, then a shutdown. Let's see if that's what's waiting for me when I get home...LOL. I opened up some of the SELinux gui's though, nothing out of the ordinary. We'll see.

[nimnull22]

Yesterday you did Updating : selinux-policy-targeted-3.8.1-2.fc14.noarch
I did not have any problem with selinux-policy-targeted-3.8.1-2.fc14

I had this message after update to selinux-policy-targeted.noarch 0:3.8.1-3.fc14.

[dd_wizard]

I got the same boot time selinux errors after updating to selinux-policy-targeted.noarch 0:3.8.1-3.fc14 also. I'm also getting AVC denial messages I haven't seen since I erased nspluginwrapper months ago:
SELinux is preventing /usr/lib64/firefox-3.6/firefox from making the program stack executable.
SELinux is preventing firefox from making its memory writable and executable.
As far as I can tell, there weren't any FF updates lately, and I haven't updated any plugins. Is anybody else seeing those AVC denials?

dd_wizard

[nimnull22]

Can you fill new BUG report for it?

[GoinEasy9]

Ahhh, didn't even see #3 selinux-policy-targeted.noarch 0:3.8.1-3.fc14

Hope it gets to 3.8.1-4 before I get home. :)

[dd_wizard]

Bug (https://bugzilla.redhat.com/show_bug.cgi?id=597616) submitted.

dd_wizard

[R3353]

in the meantime, safe-mode doesn't seg fault and works fine as long as you don't mind:
(firefox:2394): Gdk-WARNING **: XID collision, trouble ahead

(firefox:2394): Gdk-WARNING **: XID collision, trouble ahead

(firefox:2394): Gdk-WARNING **: XID collision, trouble ahead
.
.
.
:eek:

[dd_wizard]

in the meantime, safe-mode doesn't seg fault and works fine as long as you don't mind:
(firefox:2394): Gdk-WARNING **: XID collision, trouble ahead

(firefox:2394): Gdk-WARNING **: XID collision, trouble ahead

(firefox:2394): Gdk-WARNING **: XID collision, trouble ahead
.
.
.
:eek:
That's been around for a long time. The warning was put in upstream to troubleshoot a crashing problem, and has never been removed. Google a bit and you find some interesting reading. :rolleyes:

dd_wizard

[kyryder]

I got the same boot time selinux errors after updating to selinux-policy-targeted.noarch 0:3.8.1-3.fc14 also. I'm also getting AVC denial messages I haven't seen since I erased nspluginwrapper months ago:
SELinux is preventing /usr/lib64/firefox-3.6/firefox from making the program stack executable.
SELinux is preventing firefox from making its memory writable and executable.
As far as I can tell, there weren't any FF updates lately, and I haven't updated any plugins. Is anybody else seeing those AVC denials?

dd_wizard

a reboot and yup, me to :

Raw Audit Messages

node=localhost.localdomain type=AVC msg=audit(1275174204.510:60): avc: denied { execmem } for pid=15809 comm="firefox" scontext=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 tcontext=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 tclass=process

node=localhost.localdomain type=SYSCALL msg=audit(1275174204.510:60): arch=40000003 syscall=192 success=no exit=-13 a0=0 a1=10000 a2=7 a3=22 items=0 ppid=15792 pid=15809 auid=500 uid=500 gid=500 euid=500 suid=500 fsuid=500 egid=500 sgid=500 fsgid=500 tty=(none) ses=1 comm="firefox" exe="/usr/lib/firefox-3.6/firefox" subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 key=(null)


Firefox won't even launch. This could be a problem :(
Ky

[dd_wizard]

@kyryder:
Try changing SELINUX= from enforcing to permissive in /etc/selinux/config. That way you'll get the messages, but things will still work.

dd_wizard

[nimnull22]

Some interesting explanation:
All from SELinux Troubleshooter:

SummarySELinux is preventing firefox from making its memory writable and executable.
Detailed Description - The firefox application attempted to change the access protection of memory (e.g., allocated using malloc). This is a potential security problem. Firefox is probably not the problem here ,but one of its plugins. You could remove the plugin and the app would no longer require the access. If you figure out which plugin is causing the access request, please open a bug report on the plugin.

---------- Post added at 03:38 PM CDT ---------- Previous post was at 03:34 PM CDT ----------

There are two ways to fix this problem, you can install the nsspluginwrapper package, which will cause firefox to run its plugins under a separate process. This process will allow the execmem access. This is the safest choice. You could also turn off the allow_unconfined_nsplugin_transition boolean. setsebool -P allow_unconfined_nsplugin_transition=0

I am going to save it and check later to find out which way developers will go. Honestly, I do not like firefox plugins. Opera works without any problems.

[kyryder]

@kyryder:
Try changing SELINUX= from enforcing to permissive in /etc/selinux/config. That way you'll get the messages, but things will still work.

dd_wizard

Thanks, :) that should work and I will probably have to give it a shot, but for now I'm sshing to use firefox from another machine. ;)

Ky

[GoinEasy9]

Luckily, I got to see the problems before I updated tonight. I excluded selinux-policy* and everything is working. Did anyone try downgrading selinux-policy* instead of trying to work around the bug? When I tried to work around the abrt/selinux bug, I created more problems than I solved. Just sayin', It'll probably be fixed in the next version. Besides, I saw a new version of gnome-shell, the next thing I'm going to do is play with that.

Tom

[kyryder]

Luckily, I got to see the problems before I updated tonight. I excluded selinux-policy* and everything is working.

Tom

Now that's just cheating! :p and you are going to miss all the fun of putting things back together.;)

I haven't made any attempt to downgrade or fix the issue as of yet.

Ky

[GoinEasy9]

Now that's just cheating! and you are going to miss all the fun of putting things back together.
Heh, it was either play with the new version of gnome-shell or fix selinux. I chose the former. Besides, I ran Debian Sid for 2 years before coming to Fedora. Fixing things may be fun, but, putting things back together? I did way to much of that with Sid.

Tom

[GoinEasy9]

Well, wanted to play with gnome-shell, but, todays updates won't let me.

May 30 16:25:34 Updated: gnome-panel-libs-2.31.2-2.fc14.i686
May 30 16:25:37 Updated: mesa-dri-drivers-7.9-0.1.fc14.i686
May 30 16:25:42 Updated: ibus-pinyin-db-open-phrase-1.3.8-1.fc14.noarch
May 30 16:25:45 Updated: mesa-libGL-7.9-0.1.fc14.i686
May 30 16:25:47 Updated: mesa-libGLU-7.9-0.1.fc14.i686
May 30 16:25:48 Updated: glew-1.5.4-1.fc14.i686
May 30 16:25:49 Updated: glx-utils-7.9-0.1.fc14.i686
May 30 16:25:50 Updated: ibus-pinyin-1.3.8-1.fc14.i686
May 30 16:25:59 Updated: gnome-panel-2.31.2-2.fc14.i686
May 30 16:26:00 Updated: mesa-dri-drivers-experimental-7.9-0.1.fc14.i686

Read all about it, new mesa updates crash eye candy....news at 11. Heh.
https://bugzilla.redhat.com/show_bug.cgi?id=597875

[R3353]

@kyryder:
Try changing SELINUX= from enforcing to permissive in /etc/selinux/config. That way you'll get the messages, but things will still work.

running
firefox -safe-modealso works fine, other than the gdk-warning that has indeed been around for a long time (thanks for alerting me to that dd_wizard).

Did anyone try downgrading selinux-policy* instead of trying to work around the bug?

Neither selinux-policy package that received updates yesterday is eligible for downgrade.

[GoinEasy9]

I'm still holding on to selinux-policy* -3.8.1-2.fc14 until -4 version comes out. I didn't realize they weren't eligible for downgrades. Good thing I excluded the -3's from upgrade.

[dd_wizard]

You might try erasing mesa-dri-drivers-experimental-7.9-0.1.fc14.i686, GE9. Compiz works fine here with the non-experimental drivers after the update:

[gene@Mobile-PC ~]$ yum list installed gnome-panel* mesa*
Loaded plugins: fastestmirror, presto, refresh-packagekit, remove-with-leaves, show-leaves,
: verify
Installed Packages
gnome-panel.x86_64 2.31.2-2.fc14 @rawhide
gnome-panel-libs.x86_64 2.31.2-2.fc14 @rawhide
mesa-dri-drivers.x86_64 7.9-0.1.fc14 @rawhide
mesa-libGL.x86_64 7.9-0.1.fc14 @rawhide
mesa-libGL-devel.x86_64 7.9-0.1.fc14 @rawhide
mesa-libGLU.x86_64 7.9-0.1.fc14 @rawhide
mesa-libGLU-devel.x86_64 7.9-0.1.fc14 @rawhide
mesa-libOSMesa.x86_64 7.9-0.1.fc14 @rawhide
mesa-libOSMesa-devel.x86_64 7.9-0.1.fc14 @rawhide


dd_wizard

[R3353]

I'm still holding on to selinux-policy* -3.8.1-2.fc14 until -4 version comes out.

Wish I'd done the same. I was overly eager to update because of AVC denials preventing .xsession-errors append, read and/or write access to various files in my home directory that restorecon did not fix. This was a bug in F12 that meandered back into my system somehow. se-policy-3.8.1-3 fixed this, but now holds my web browsers hostage in a half-way house, if you will.

[nimnull22]

I do not know where do you all get your updates, but I do not have mesa experimental. I have only (i686, intel i915):
Installed Packages
mesa-dri-drivers.i686 7.9-0.1.fc14 @rawhide
mesa-libGL.i686 7.9-0.1.fc14 @rawhide
mesa-libGLU.i686 7.9-0.1.fc14 @rawhide

Second. Although I have as well "avc: denied { read write } messages at boot time, everything works.

Third. In "glxgears" test gears do not rotate smooth, unless I start to move my mouse around screen. Also only then I got ~ 60FPS, otherwise no more then ~20. Nothing changes, at least for i915GM.

[GoinEasy9]

@nimnull22
Geez, your running with Intel graphics too. I have a Nvidia 9800GT and have to use nouveau and the experimental drivers for 3D. I have a laptop with Intel i915 graphics, maybe I'll put a rawhide install on that. I'm really surprised to hear that you and dd_wizard don't need the experimental drivers using Intel.

I like this thread, I'm learning some very interesting things here.

[kyryder]

Just got some updates. No need for --skip-broken, at least for the moment.:D

Updated:
ffmpeg-libs.i686 0:0.6-0.3.20100429svn.fc14
gstreamer-plugins-ugly.i686 0:0.10.14-2.fc14
libquicktime.i686 0:1.1.5-1.fc14
mencoder.i686 0:1.0-0.117.20100429svn.fc14
mplayer.i686 0:1.0-0.117.20100429svn.fc14
mplayer-common.i686 0:1.0-0.117.20100429svn.fc14
x264-libs.i686 0:0.0.0-0.27.20100429gitd9db8b3.fc14

Ky

[nimnull22]

Update: rawhide report: 20100602 changes Rawhide Report

Still have at the boot time:
...
SELinux: initialized (dev sysfs, type sysfs), uses genfs_contexts
type=1403 audit(1275506213.608:3): policy loaded auid=4294967295 ses=4294967295
dracut: Switching root
type=1400 audit(1275506214.800:4): avc: denied { read write } for pid=305 comm="hostname" path="/dev/console" dev=devtmpfs ino=5578 scontext=system_u:system_r:hostname_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275506214.806:5): avc: denied { read write } for pid=305 comm="hostname" path="/dev/console" dev=devtmpfs ino=5578 scontext=system_u:system_r:hostname_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275506214.814:6): avc: denied { read write } for pid=308 comm="consoletype" path="/dev/console" dev=devtmpfs ino=5578 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275506214.820:7): avc: denied { read write } for pid=309 comm="consoletype" path="/dev/null" dev=devtmpfs ino=4053 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275506214.830:8): avc: denied { read write } for pid=309 comm="consoletype" path="/dev/null" dev=devtmpfs ino=4053 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275506214.838:9): avc: denied { read write } for pid=308 comm="consoletype" path="/dev/console" dev=devtmpfs ino=5578 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275506215.255:10): avc: denied { open } for pid=310 comm="mount" name="null" dev=devtmpfs ino=4053 scontext=system_u:system_r:mount_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file
type=1400 audit(1275506215.269:11): avc: denied { read write } for pid=312 comm="consoletype" path="/dev/console" dev=devtmpfs ino=5578 scontext=system_u:system_r:consoletype_t:s0 tcontext=system_u:object_r:device_t:s0 tclass=chr_file


Second, which I think more nasty:
================================================== =
[ INFO: suspicious rcu_dereference_check() usage. ]
---------------------------------------------------
include/net/inet_timewait_sock.h:227 invoked rcu_dereference_check() without protection!

other info that might help us debug this:


rcu_scheduler_active = 1, debug_locks = 0
3 locks held by X/1014:
#0: (&u->readlock){+.+.+.}, at: [<c07aaf41>] unix_stream_recvmsg+0xbc/0x425
#1: (&(&u->lock)->rlock){+.+...}, at: [<c07aaf8a>] unix_stream_recvmsg+0x105/0x425
#2: (net/ipv4/tcp_minisocks.c:41){+.-...}, at: [<c044d37c>] run_timer_softirq+0x123/0x233

stack backtrace:
Pid: 1014, comm: X Not tainted 2.6.34-18.fc14.i686.PAE #1
Call Trace:
[<c07cd90f>] ? printk+0x14/0x1d
[<c0468b06>] lockdep_rcu_dereference+0x7d/0x86
[<c076d39f>] twsk_net+0x3a/0x40
[<c076d842>] __inet_twsk_kill+0x6d/0xaf
[<c076db8b>] inet_twdr_do_twkill_work+0x50/0xa0
[<c076dc96>] inet_twdr_hangman+0x2f/0x89
[<c044d3f2>] run_timer_softirq+0x199/0x233
[<c076dc67>] ? inet_twdr_hangman+0x0/0x89
[<c0447bdc>] __do_softirq+0xc1/0x17d
[<c0447cd3>] do_softirq+0x3b/0x5f
[<c0447e16>] irq_exit+0x3a/0x6d
[<c041eaf5>] smp_apic_timer_interrupt+0x73/0x81
[<c07d0766>] apic_timer_interrupt+0x36/0x3c
[<c046007b>] ? read_profile+0xc2/0x17c
[<c07d0080>] ? _raw_spin_unlock_irqrestore+0x45/0x4d
[<c073638b>] skb_dequeue+0x48/0x4f
[<c07aaf92>] unix_stream_recvmsg+0x10d/0x425
[<c072fed6>] __sock_recvmsg+0x59/0x63
[<c0732125>] sock_aio_read+0xb3/0xbe
[<c04ea4bf>] do_sync_read+0x8f/0xca
[<c05903a5>] ? security_file_permission+0x14/0x16
[<c04ea597>] ? rw_verify_area+0x9d/0xc0
[<c04eaaf3>] vfs_read+0x90/0xe1
[<c04eabe2>] sys_read+0x40/0x62
[<c07d025c>] syscall_call+0x7/0xb

Most possible, that it happened after Opera starts on kernel 34-18.
Kernel 34-14 didn't have this message.

Thanks

[GoinEasy9]

@nimnull22
Is that with the new selinux-policy* 3.8.1-4.fc14?

Edit: Oh, I see, there's a new kernel too.

Edit: Oh, well, I'll give it a try...LOL

[nimnull22]

Yes, it was with selinux-policy* 3.8.1-4.fc14, problem still exists.

[GoinEasy9]

I never updated to 3.8.1-3, but, my update from 3.8.1-2 to -4 went successfully.
--skip-broken only with control-center and control-center-filesystem

So far, no notifications, and everything is working. BBL after I run it for a while.

Edit: I get the same notification in messages after opening chrome as you do after opening opera. I don't see it causing any problems though.

[nimnull22]

That is strange. I will write to bugzilla.
And another problems, now with ip6tables. I have:
stack backtrace:
Pid: 1014, comm:ip6tables Not tainted 2.6.34-18.fc14.i686.PAE #1
Call Trace:
....

But when I shut down laptop, how can I save this message, it doesn't go to dmesg?

---------- Post added at 12:27 PM CDT ---------- Previous post was at 12:23 PM CDT ----------

And when I press "Crtl-Alt-Del" because reboot process halts, I get the same audit messages - avc: denied bla bla. And as well I can't save it. 34-18 doesn't work properly.

[GoinEasy9]

I just searched messages and I don't have the ip6tables notification that you have. I also looked back and the same messages ([ INFO: suspicious rcu_dereference_check() usage. ]) came through with the 2.6.34-14 kernel.
When I had the problem with the 2.6.34 kernel not allowing shutdown, I couldn't find a way of logging shutdown messages. If someone has a hint about how to do that, I'm open for suggestions.
I also sent the kernel shutdown problems to kernelopps, or whatever it's called. You can also do a search on the kernel mailing list (or just Google it, easier) and see what you come up with.

[dd_wizard]

Bug #598475 (https://bugzilla.redhat.com/show_bug.cgi?id=598475) addresses the SE Linux errors at boot up. There's a fix shown in the bug comments, so I'm not sure if/when the fix will hit the repo. The fix is to edit /usr/share/dracut/modules.d/99base/selinux-loadpolicy.sh, then generate a new initrd. It worked for me.

Bug #593026 (https://bugzilla.redhat.com/show_bug.cgi?id=593026) addresses the [ INFO: suspicious rcu_dereference_check() usage. ] messages in /var/log/messages. Feel free to add a comment to that one, not much has happened on it.

dd_wizard

[nimnull22]

Update: rawhide report: 20100603 changes

selinux-policy 3.8.1-5.fc14 - still have errors at boot time.
Kernel 2.6.34-20: Wakeups-from-idle per second : 12.8, was 5-6 on 2.6.34-18:

C3 80.5ms (100.0%)
Wakeups-from-idle per second : 12.8 interval: 5.0s
no ACPI power usage estimate available

Top causes for wakeups:
28.7% ( 5.4) <kernel core> : hrtimer_start (tick_sched_timer)
21.3% ( 4.0) <interrupt> : ata_piix
21.3% ( 4.0) <kernel core> : __mod_timer (rh_timer_func)


But there is no "[ INFO: suspicious rcu_dereference_check() usage. ] "

[GoinEasy9]

I finally got the avc denials with Firefox starting yesterday with selinux-policy 3.8.1-4 ( I skipped 3.8.1-3). Today's update to 3.8.1-5 didn't help. Solution in https://bugzilla.redhat.com/show_bug.cgi?id=597858 to execute setsebool -P allow_execmem 0 didn't solve the problem either. Since I'm using Chrome now as my default browser it doesn't interfere much with using the desktop. I'm just using FF to test and report back to the bugzilla about progress.

@nimnull22 What are you using to check Wakeups-from-idle per second, and the causes for wakeups? I'd like to keep track also.

The [ INFO: suspicious rcu_dereference_check() usage. ] messages aren't in my /var/log/messages file any longer.

Also, filed new bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=600163
I'm getting a notification window asking for root password in order to start gnome-power-backlight helper.

[nimnull22]

Powertop - perfect program, to find out about CPU usage. Of course, I told about it just for information. I always check it, as long as I use laptop and that is quite important for me. I want it to be silent and cool. For that purpose I use xfce also.

I did not check Firefox, will do today after new update.

[GoinEasy9]

Rawhide's acting like rawhide...
Firefox still has avc errors, fixes didn't work for me. https://bugzilla.redhat.com/show_bug.cgi?id=597858
Nouveau still not working along with gnome-shell. https://bugzilla.redhat.com/show_bug.cgi?id=597875
Gnome power backlight helper notification popup at login screen. https://bugzilla.redhat.com/show_bug.cgi?id=600163
Gedit has now started segfaulting. https://bugzilla.redhat.com/show_bug.cgi?id=602039

Mama told me there'd be days like this, there'd be days like this my mama said.....

[nimnull22]

Actually not only Gedit:
operapluginwrap[1358]: segfault at 0 ip (null) sp bfaf945c error 4 in libm-2.12.90.so[110000+28000]
operapluginwrap[1370]: segfault at 0 ip (null) sp bfaa30bc error 4 in libuuid.so.1.3.0[110000+4000]
operapluginwrap[1394]: segfault at 0 ip (null) sp bff596ac error 4 in libX11.so.6.3.0[110000+135000]
operapluginwrap[1406]: segfault at 0 ip (null) sp bfd507ac error 4 in libXt.so.6.0.0[110000+58000]
gedit[2587]: segfault at 0 ip 04b7a438 sp bf87257c error 4 in libc-2.12.90.so[4b00000+18e000]

But as long I use xfce and leafpad, I can't join your party.

---------- Post added 2010-06-09 at 05:20 AM CDT ---------- Previous post was 2010-06-08 at 11:04 PM CDT ----------

Report about: 20100609 changes

Error: Package: gthumb-2.11.3-2.fc14.i686 (@rawhide)
Requires: libbrasero-burn.so.0
Removing: brasero-libs-2.31.2-1.fc14.i686 (@rawhide)
libbrasero-burn.so.0
Updated By: brasero-libs-2.31.3-1.fc14.i686 (rawhide)
Not found
Error: Package: sound-juicer-2.28.2-2.fc14.i686 (@rawhide/13)
Requires: libbrasero-media.so.0
Removing: brasero-libs-2.31.2-1.fc14.i686 (@rawhide)
libbrasero-media.so.0
Updated By: brasero-libs-2.31.3-1.fc14.i686 (rawhide)
Not found
Error: Package: rhythmbox-0.12.8-4.fc14.i686 (@rawhide)
Requires: libbrasero-media.so.0
Removing: brasero-libs-2.31.2-1.fc14.i686 (@rawhide)
libbrasero-media.so.0
Updated By: brasero-libs-2.31.3-1.fc14.i686 (rawhide)
Not found

2.
Updating : evince-2.31.3-1.fc14.i686
I/O warning : failed to load external entity "/etc/gconf/schemas/evince.schemas"
Failed to open `/etc/gconf/schemas/evince.schemas': No such file or directory


3.
Cleanup : setroubleshoot-2.2.83-1.fc13.i686
/usr/share/glib-2.0/schemas/org.gnome.eog.gschema.xml: Error on line 659 char 67: invalid name 'full_screen': invalid character '_'; only lowercase letters, numbers and dash ('-') are permitted.


4.
Non-fatal POSTTRANS scriptlet failure in rpm package 1:gedit-2.31.3-1.fc14.i686
Non-fatal POSTTRANS scriptlet failure in rpm package eog-2.31.3-1.fc14.i686


So, everything is good except for:
Skipped (dependency problems):
brasero.i686 0:2.31.3-1.fc14 brasero-libs.i686 0:2.31.3-1.fc14 brasero-nautilus.i686 0:2.31.3-1.fc14

---------- Post added at 07:24 AM CDT ---------- Previous post was at 05:20 AM CDT ----------

Interesting, but "gedit" starts, but when closes gives:
(gedit:2795): GLib-GIO-CRITICAL **: g_settings_set_value: assertion `G_IS_SETTINGS (settings)' failed
Segmentation fault


Firefox the same:
/usr/lib/firefox-3.6/run-mozilla.sh: line 131: 2652 Segmentation fault "$prog" ${1+"$@"}

And I still have AVC denied at the boot time.

[dd_wizard]

@nimnull22: You may have to use the fix in the bug report for the boot AVC denials. That did fix it for me. I got all the errors you saw on today's update, also. Oh well, it is rawhide. ;) I thought about erasing gthumb so the brasero update would succeed, but why bother? We should see a rebuilt gthumb soon.

dd_wizard

[nimnull22]

Of course I can apply fix from bugzilla, but I want to be a normal user, developers have to do it. I can wait, I report about it for others, in case if one has a question.

Thanks.

[GoinEasy9]

Seems like the gedit bug has been solved, it's no longer segfaulting on exit. Found a new avc denial on amarok, and was able to file bugs on clutter and mutter which are preventing nouveau from running gnome-shell and compiz, but, there are so many debuginfo's missing, I'm thinking that the BZ's I'm filing are useless.

[dd_wizard]

They may not be as helpful as they could be, GE9, but at the very least they'll let the devs know what's broken. ;)

dd_wizard

[nimnull22]

Update rawhide: 20100612 for i686

Error: Package: sound-juicer-2.28.2-2.fc14.i686 (@rawhide/13)
Requires: libbrasero-media.so.0
Removing: brasero-libs-2.31.2-1.fc14.i686 (@rawhide)
libbrasero-media.so.0
Updated By: brasero-libs-2.31.3-1.fc14.i686 (rawhide)
Not found

With --skip-broken:

Skipped (dependency problems):
brasero.i686 0:2.31.3-1.fc14 brasero-libs.i686 0:2.31.3-1.fc14 brasero-nautilus.i686 0:2.31.3-1.fc14 gthumb.i686 0:2.11.3-3.fc14
rhythmbox.i686 0:0.12.8-5.fc14

Replaced:
ibus-gtk.i686 0:1.3.4-2.fc14.

And after reboot:
$ gedit

(gedit:1328): GLib-GIO-CRITICAL **: g_settings_set_value: assertion `G_IS_SETTINGS (settings)' failed
Segmentation fault

Segmentation fault after close.
gedit[1328]: segfault at 0 ip 008dfb7b sp bfb92f50 error 4 in libgconf-2.so.4.1.5[8c4000+37000]



$ firefox
/usr/lib/firefox-3.6/run-mozilla.sh: line 131: 1362 Segmentation fault "$prog" ${1+"$@"}

[dd_wizard]

The sound-juicer problem sounds like Bug 602895 (https://bugzilla.redhat.com/show_bug.cgi?id=602895) I filed against gthumb. I see that gthumb, rhythmbox, and sound-juicer all have rpms rebuilt against brasero in koji now. You could download and localupdate them if you're in a hurry. :)

I filed a bug report for the new gedit seg fault, but I haven't seen the firefox crash yet. Perhaps because I'm running 64 bit.

dd_wizard

---------- Post added at 01:43 PM CDT ---------- Previous post was at 01:41 PM CDT ----------

Oh, if you haven't added this line:
ulimit -c 1024

to your ~/.bash_profile, you should nimnull22. That will put smallish core files in your home directory that you need to clean up once in a while. It will also make abrt start working again. :)

dd_wizard

[GoinEasy9]

gedit is still segfaulting upon close. I'm not going to file again until all the debuginfos's can be downloaded. abrt is working again, but the debug repos are still having problems.

Talk is that they're going to start pushing systemd into the mix, heh, we're going to need a working abrt for that.

[GoinEasy9]

Tonight, a new mesa-dri-drivers-experimental came in which restore my ability to play with compiz and gnome-shell using the nouveau drivers. Unfortunately, a bug in gnome-panel is bringing up blank windows in gnome-shell. Can't file a BZ, the debuginfo's aren't available/loadable.

[dd_wizard]

Does Bug 603226 (https://bugzilla.redhat.com/show_bug.cgi?id=603226) cover what you're seeing GoinEasy9? I get it every time I start gnome-shell, but after it restarts things work until mutter (the window manager for g-s) crashes. Then gnome-panel seg faults again and about 20-25% of the time I have to drop out of gnome-shell to get a panel back.

dd_wizard

[GoinEasy9]

A new version of blender came in compiled for F14, but it's segfaulting due to the nouveau driver. It's also generating an AVC denial. That's Firefox, amarok and blender with AVC denials (At least the ones I've found).

[dd_wizard]

Dump amarok for exaile if you're using gnome. You don't need all that KDE crap on your system. :p

dd_wizard

[GoinEasy9]

Hey dd, yes that's the same bug. It comes up as soon as I start gnome-shell. An then the window manager crashes and I'm left with empty windows. No panels and no windows...but the wallpaper looks nice and uncluttered. :p

[dd_wizard]

HAHA! :p

dd_wizard

[GoinEasy9]

I have exaile, it looks nice, just tested amarok to see if it would run, heh, oh well. Another app from my KDE days that I'll never use again.

---------- Post added at 08:03 PM CDT ---------- Previous post was at 07:56 PM CDT ----------

Damn, gnome-packagekit just segfaulted. It's the experimental nouveau driver. Just shut off 3D and no segfault. I've really got to install rawhide onto the Intel laptop, Geez.

[nimnull22]

Honestly, compiz works on my i915, but under XFCE on console I have artifacts around font, and effects are not really smooth. I also have gnome - and compiz works there as well, but sorry I do not use gnome and compiz. My needs cover only network area, for me xfce is more then enough.

[nimnull22]

Update 16.06.10

Everything is good, 34-38 works fine for my i686. There is no new errors yet.
But.

I have no any idea when does it start, but I notice today that skype is denied by selinux. And solution is strange:
Fix Command
Choose one of the following to allow access:

Allow unconfined executables to make their stack executable. This should never, ever be necessary. Probably indicates a badly coded executable, but could indicate an attack. This executable should be reported in bugzilla")
# setsebool -P allow_execstack 1

Allow unconfined executables to map a memory region as both executable and writable, this is dangerous and the executable should be reported in bugzilla")
# setsebool -P allow_execmem 1

Which one should I chose, and do I need to chose both?
Should I report to bugzilla?

[GoinEasy9]

Report everything to bugzilla, they can't fix things they don't know about.

[nimnull22]

There was not any updates since:
"rawhide report: 20100618 changes Rawhide Report"

Is it true, or I missed something?

[GoinEasy9]

I just got an email from the devel mailing list. Since it is public, I'll post it:

Just a quick warning:

I have built a glib update that affects some users of GDbus and other new apis.
We'll try to get everything straightened out by tomorrow, but you never know...

Matthias
I guess it's fair warning, maybe one should wait till tomorrow to do any updates.

---------- Post added at 05:01 PM CDT ---------- Previous post was at 03:23 PM CDT ----------

The next message explains what happens if you do an update:

Thanks for the warning (they really are appreciated).

I updated to koji from the last hour, just to be a bit insane.

With all the gnome and glib changes in place, gnome gets to the point of displaying the desktop icons. The panel never appears and the three desktop icons keep flashing rapidly, like it is trying to restart. ctl-alt-backspace doesn't kill it.

.xsession-errors says

nm-applet: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
polkit-gnome-authentication-agent-1: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
gnome-screensaver: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.
gnome-volume-control-applet: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.

(gnome-panel:3972): EggSMClient-WARNING **: Failed to connect to the session manager: IO error occured opening connection

The application 'gnome-panel' lost its connection to the display :0.0;
most likely the X server was shut down or you killed/destroyed
----------------------------------------------------------
darrell
In order words....WAIT...LOL

[dd_wizard]

How long? :)

dd_wizard

[GoinEasy9]

Don't know..I haven't seen any post today dealing with the problem, and, I don't really need to bork the whole system. If I see anything new I'll post, unless, you want to be brave? :D

Another note, openoffice.org-3.2.1-19.16.fc14 is built on gtk3, so installing gtk3-engines manually will improve how it looks. Without it, they said it would look like a throw-back to a different era...LOL.

[SlowJet]

OK, if you want it now here is how I did it.

0. optional - grab the 35 kernel to install and when done - reboot.

1. First download the fixed kijo rpms glib2, gimp, pypoppler and any other you like for today.
cups-1.4.4-3.fc14.i686.rpm ibus-libs-1.3.5-2.fc14.i686.rpm
cups-libs-1.4.4-3.fc14.i686.rpm lftp-4.0.9-2.fc14.i686.rpm
gimp-2.6.8-9.fc14.i686.rpm lftp-scripts-4.0.9-2.fc14.noarch.rpm
gimp-help-browser-2.6.8-9.fc14.i686.rpm logrotate-3.7.8-12.fc14.i686.rpm
gimp-libs-2.6.8-9.fc14.i686.rpm pypoppler-0.12.1-2.fc14.i686.rpm
glib2-2.25.9-2.fc14.i686.rpm setroubleshoot-2.2.89-1.fc14.i686.rpm
glib2-devel-2.25.9-2.fc14.i686.rpm setroubleshoot-server-2.2.89-1.fc14.i686.rpm
ibus-1.3.5-2.fc14.i686.rpm

2. su -
cd /home/useryou/downloads/ where ever you put the rpms's
yum localupdate *.rpm --nogpgcheck

3. cd /
yum update --skip-broken --nogpgcheck

reboot, there may be a few hot flashes but for the most part it works.
K35 works with vbox-3.2.4-beta1 as host

SJ

[dd_wizard]

After steps 0, 1, and 2, I just excluded perl* from the update and it went without hitches. The package list for step 1 got smaller by the time I did it, so some of them have hit the repos already. The 2.6.35 kernel is working good so far, too. Several packages need to be built against the new perl before it will update.

dd_wizard

---------- Post added at 03:27 PM CDT ---------- Previous post was at 02:45 PM CDT ----------

This looks really familiar... Evince and gnome-color-manager crash when started. The last time this happened, I think I tracked it down to dconf and filed a bugzilla against it. Also, gnome-panel used to just crash when starting gnome-shell. Now it crashes when logging in to gnome/openbox. I guess it's time to track down the gremlins.

dd_wizard

[R3353]

Is gnome-power-manager working for any of you? I can't seem to start the gui:

Gtk-Message: Failed to load module "pk-gtk-module": libpk-gtk-module.so: cannot open shared object file: No such file or directory
Segmentation fault

btw: 2.6.34-45.fc14.i686.PAE

[dd_wizard]

No, it's not. That may or may not be related to evince, gedit, and gcm all crashing at startup. Those three crashing was related to dconf or gconf2, the first time. I'm going to check my previous bugs and find the relevant one. For now, I'm filing an ABRT bug against gnome-power-manager but I think the problem is deeper than that.

dd_wizard

---------- Post added at 04:07 PM CDT ---------- Previous post was at 03:46 PM CDT ----------

Odd... GPM may be looking in the wrong place, because the library is where it's supposed to be.
[gene@Mobile-PC ~]$ gnome-power-manager
Gtk-Message: Failed to load module "pk-gtk-module": libpk-gtk-module.so: cannot open shared object file: No such file or directory
Segmentation fault (core dumped)
[gene@Mobile-PC ~]$ locate -e libpk-gtk-module.so
/usr/lib64/gtk-2.0/modules/libpk-gtk-module.so

Bug 607008 (https://bugzilla.redhat.com/show_bug.cgi?id=607008) filed.

dd_wizard

---------- Post added at 04:50 PM CDT ---------- Previous post was at 04:07 PM CDT ----------

Yup, gnome-power-manager looks just about everywhere except /usr/lib64/gtk-2.0/modules:
[gene@Mobile-PC abrt]$ strace gnome-power-manager 2>&1 | grep libpk-gtk-module.so
access("/home/gene/.gtk-3.0/3.0.0/x86_64-redhat-linux-gnu/modules/libpk-gtk-module.so", F_OK) = -1 ENOENT (No such file or directory)
access("/home/gene/.gtk-3.0/3.0.0/modules/libpk-gtk-module.so", F_OK) = -1 ENOENT (No such file or directory)
access("/home/gene/.gtk-3.0/x86_64-redhat-linux-gnu/modules/libpk-gtk-module.so", F_OK) = -1 ENOENT (No such file or directory)
access("/home/gene/.gtk-3.0/modules/libpk-gtk-module.so", F_OK) = -1 ENOENT (No such file or directory)
access("/usr/lib64/gtk-3.0/3.0.0/x86_64-redhat-linux-gnu/modules/libpk-gtk-module.so", F_OK) = -1 ENOENT (No such file or directory)
access("/usr/lib64/gtk-3.0/3.0.0/modules/libpk-gtk-module.so", F_OK) = -1 ENOENT (No such file or directory)
access("/usr/lib64/gtk-3.0/x86_64-redhat-linux-gnu/modules/libpk-gtk-module.so", F_OK) = -1 ENOENT (No such file or directory)
access("/usr/lib64/gtk-3.0/modules/libpk-gtk-module.so", F_OK) = -1 ENOENT (No such file or directory)
stat("libpk-gtk-module.so", 0x7ffff4be6f00) = -1 ENOENT (No such file or directory)
stat("libpk-gtk-module.so.so", 0x7ffff4be6f00) = -1 ENOENT (No such file or directory)
stat("libpk-gtk-module.so.la", 0x7ffff4be6f00) = -1 ENOENT (No such file or directory)
open("/lib64/tls/x86_64/libpk-gtk-module.so", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/lib64/tls/libpk-gtk-module.so", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/lib64/x86_64/libpk-gtk-module.so", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/lib64/libpk-gtk-module.so", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/lib64/tls/x86_64/libpk-gtk-module.so", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/lib64/tls/libpk-gtk-module.so", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/lib64/x86_64/libpk-gtk-module.so", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/lib64/libpk-gtk-module.so", O_RDONLY) = -1 ENOENT (No such file or directory)
write(2, "Gtk-Message: Failed to load modu"..., 131Gtk-Message: Failed to load module "pk-gtk-module": libpk-gtk-module.so: cannot open shared object file: No such file or directory
[
However, gpm still seg faults even with a sym link to the gtk-2.0 version in /usr/lib64. It finds the lib, but the 2.0 version is apparently incompatible. :(

dd_wizard

---------- Post added at 05:11 PM CDT ---------- Previous post was at 04:50 PM CDT ----------

@R3353: This works. :)

1. Download gnome-power-manager-2.31.1-3.fc14 (http://koji.fedoraproject.org/koji/buildinfo?buildID=178525) from koji to some directory.

2. In that directory, run:sudo yum --nogpgcheck downgrade ./gnome-power-manager*

3. nohup gnome-power-manager >/dev/null 2>&1 &

dd_wizard

[GoinEasy9]

It's not only glib, but they're talking about recompiling with new versions gcc and python. That, and all the existing problems, remind me of times when I use to leave Sid alone for a while so the devs could sync everything up.
I've got F13 running on the Intel laptop, but I'm going to wait till things settle down before I upgrade it to Rawhide.
BTW - F13 with SELinux enabled hasn't given me any problems so far. Not one burp from SELinux yet, I'm impressed. If it runs so smooth on F13, why all the AVC denials on F14?

[R3353]

Thank you kindly dd_wizzard. I'll give that a go in the morning.

[nimnull22]

I did it, with "--skip-broken".
So those was successfully skipped:
Packages skipped because of dependency problems:
evince-2.31.3-4.20100621git.fc14.i686 from rawhide
evince-djvu-2.31.3-4.20100621git.fc14.i686 from rawhide
evince-libs-2.31.3-4.20100621git.fc14.i686 from rawhide
2:gimp-2.6.8-8.fc14.i686 from rawhide
2:gimp-help-browser-2.6.8-8.fc14.i686 from rawhide
2:gimp-libs-2.6.8-8.fc14.i686 from rawhide
libebml-1.0.0-1.fc14.i686 from rawhide
libmatroska-1.0.0-1.fc14.i686 from rawhide
4:perl-5.12.1-123.fc14.i686 from rawhide
perl-Compress-Raw-Bzip2-2.027-1.fc14.i686 from rawhide
perl-Compress-Raw-Zlib-2.027-1.fc14.i686 from rawhide
perl-File-BaseDir-0.03-7.fc14.noarch from rawhide
perl-File-DesktopEntry-0.04-11.fc14.noarch from rawhide
perl-HTML-Parser-3.64-3.fc14.i686 from rawhide
perl-HTML-Tagset-3.20-5.fc14.noarch from rawhide
perl-IO-Compress-2.027-1.fc14.noarch from rawhide
1:perl-Module-Pluggable-3.90-123.fc14.noarch from rawhide
1:perl-Pod-Escapes-1.04-123.fc14.noarch from rawhide
1:perl-Pod-Simple-3.13-123.fc14.noarch from rawhide
perl-URI-1.54-2.fc14.noarch from rawhide
perl-XML-Parser-2.36-9.fc14.i686 from rawhide
4:perl-libs-5.12.1-123.fc14.i686 from rawhide
perl-libwww-perl-5.834-1.fc14.noarch from rawhide
3:perl-version-0.82-123.fc14.noarch from rawhide
poppler-0.14.0-1.fc14.i686 from rawhide
poppler-glib-0.14.0-1.fc14.i686 from rawhide
poppler-utils-0.14.0-1.fc14.i686 from rawhide
1:xchat-2.8.8-3.fc14.i686 from rawhide


But XFCE works, GNOME works even I have some errors in ".xsession-errors". I notice that compiz loads it self with gnome and works also (I noticed "fade in/out")
So it is safe to update. It is all about i686.

[R3353]

I wasn't so lucky...

--> Processing Conflict: firefox-3.6.3-4.fc14.i686 conflicts xulrunner >= 1.9.2.4
--> Finished Dependency Resolution
Skip-broken could not solve problems:eek:
Error: Package: perl-Git-1.7.1-1.fc14.noarch (@rawhide/13)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.10.1-118.fc14.i686 (@rawhide/13)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-123.fc14.i686 (rawhide)
Not found
Error: firefox conflicts with xulrunner
Error: Package: perl-DBD-SQLite-1.29-1.fc14.i686 (@rawhide)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.10.1-118.fc14.i686 (@rawhide/13)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-123.fc14.i686 (rawhide)
Not found
Error: Package: 2:vim-enhanced-7.2.444-1.fc14.i686 (@rawhide)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.10.1-118.fc14.i686 (@rawhide/13)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-123.fc14.i686 (rawhide)
Not found
Error: Package: 1:net-snmp-libs-5.5-15.fc14.i686 (@rawhide)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.10.1-118.fc14.i686 (@rawhide/13)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-123.fc14.i686 (rawhide)
Not found
You could try running: rpm -Va --nofiles --nodigest

Can't really complain though, what I use works.

[nimnull22]

You can try like this:
yum update --skip-broken -x xulrunner.i686

[R3353]

yes, thanks nimnull. evince, gimp, and poppler updated without errors. Other than that, your list of updates skipped for dependency problems is about the same as mine - extremely perly.

[dd_wizard]

For now,[gene@Mobile-PC RPMS]$ sudo yum --nogpgcheck --exclude=xulrunner --exclude=perl* --exclude=gnome-power-manager update is my default update command. Seems to complete without complaining. :)

How's the 2.6.35 kernel working? So far, so good here.

dd_wizard

---------- Post added at 04:13 PM CDT ---------- Previous post was at 03:45 PM CDT ----------

Oh, the new dconf in koji fixes the evince and gedit startup crashes, but gedit still seg faults on exit. :(

Has anybody else lost there theme colors in nautilus? This happened after updating yesterday.

dd_wizard

[kurtdriver]

How's the 2.6.35 kernel working?
I just booted into it and some new gnome stuff. I can't keep track of all the gnome crashes, and whether or not that's the kernel's fault, I couldn't tell you. It's a crash a second on my laptop, a bit slow too. At least the pointing device has returned on a consistent basis.


Has anybody else lost there theme colors in nautilus? This happened after updating yesterday.

Is that what happened? I thought it was deliberate
(and horrible), it reminds me of RH 5.1.

[nimnull22]

OK, update of 24.06.10.
Firefox and xulrunner were updated, but firefox still gives:
$ firefox
/usr/lib/firefox-3.6/run-mozilla.sh: line 131: 2066 Segmentation fault "$prog" ${1+"$@"}
And: SELinux is preventing firefox from making its memory writable and executable.
Looks like it is not really important - try Opera.

What was skipped:
Skipped (dependency problems):
libebml.i686 0:1.0.0-1.fc14 libmatroska.i686 0:1.0.0-1.fc14
perl.i686 4:5.12.1-123.fc14 perl-Compress-Raw-Bzip2.i686 0:2.027-1.fc14
perl-Compress-Raw-Zlib.i686 0:2.027-1.fc14 perl-File-BaseDir.noarch 0:0.03-7.fc14
perl-File-DesktopEntry.noarch 0:0.04-11.fc14 perl-HTML-Parser.i686 0:3.64-3.fc14
perl-HTML-Tagset.noarch 0:3.20-5.fc14 perl-IO-Compress.noarch 0:2.027-1.fc14
perl-Module-Pluggable.noarch 1:3.90-123.fc14 perl-Pod-Escapes.noarch 1:1.04-123.fc14
perl-Pod-Simple.noarch 1:3.13-123.fc14 perl-URI.noarch 0:1.54-2.fc14
perl-XML-Parser.i686 0:2.36-9.fc14 perl-libs.i686 4:5.12.1-123.fc14
perl-libwww-perl.noarch 0:5.834-1.fc14 perl-version.noarch 3:0.82-123.fc14
vim-common.i686 2:7.2.444-2.fc14 vim-enhanced.i686 2:7.2.444-2.fc14
xchat.i686 1:2.8.8-3.fc14


New kernel gives me segmentation fault on Powertop also.
Did not check Gnome.

[dd_wizard]

Odd, on my x86_64 install I'm seeing the AVC denials from FF, but no seg faults with either 2.6.34 or 2.6.35. Also, powertop works correctly with the 2.6.35 kernel.

dd_wizard

[nimnull22]

Try to start FF on console.
I can wait with powertop for a few weeks, it is not very important.

But FF, it is different.

[R3353]

I have been running firefox -safe-mode for almost 3 weeks now. You don't have to disable any custom settings, themes, or extensions (on my setup anyway....).

[GoinEasy9]

From the posts in the bugzilla's, the AVC denial in Firefox is due to one of the extensions, so, running it in safe-mode will work because it doesn't load up the addons. I really wish I had the extra time to disable the extensions one by one to see which one(s) is causing the problem. The Firefox AVC denial isn't really a problem for me, got tired of the memory leaks etc., and switched to Chrome.

[R3353]

ah, yes, I see. I was electing to not make these changes permanent.

[dd_wizard]

Firefox run from a CLI:
[gene@Mobile-PC ~]$ firefox
(child won, so we're not deferring)
(child won, so we're deferring)
(processing deferred in-call)
(child won, so we're deferring)
(child won, so we're not deferring)
(processing deferred in-call)
NOTE: child process received `Goodbye', closing down
[gene@Mobile-PC ~]$

I'm not sure what the child and processing messages are about, but they occurred while the tabs were loading. The NOTE: message occurred when I manually closed the FF window. No crashes occurred. I'm starting to think the AVC denials may be related to LSOs in general, or BetterPrivacy in particular. They seem to occur when I get an LSO loaded warning from BetterPrivacy.

dd_wizard

[GoinEasy9]

I just did a check-update, long list, but, nothing is coming up as broken. Been about 3 days since I updated, any advice on if I should still hold perl back? I might just let the whole thing ride and see what happens. OK, no answer in a couple of minutes and I pull the switch...LOL.

[dd_wizard]

Perl is still blocked by vim* and gnome-power-manager is still the broken one. I suppose you could erase vim if you really want the perl update. However, erasing vim might get you banned from the forums, and perhaps cause physical damage! :p :D

dd_wizard

[GoinEasy9]

What's wrong with gnome-power-manager? That's the new version that the BZ said would fix my notification popup during login. Geez.

---------- Post added at 04:33 PM CDT ---------- Previous post was at 04:32 PM CDT ----------

BTW - keep it a secret, but I only have vim-minimal on this system..I never use it.

---------- Post added at 04:34 PM CDT ---------- Previous post was at 04:33 PM CDT ----------

And vim-minimal is in the updates...could they have fixed everything????

[dd_wizard]

See posts 82 and 83 in this thread. The new version can't find a required shared library now that it uses GTK3. So it seg faults on start up.

dd_wizard

---------- Post added at 01:36 PM CDT ---------- Previous post was at 01:35 PM CDT ----------

Apparently not:
---> Package vim-common.x86_64 2:7.2.444-2.fc14 set to be updated
---> Package vim-enhanced.x86_64 2:7.2.444-2.fc14 set to be updated
---> Package vim-minimal.x86_64 2:7.2.444-2.fc14 set to be updated
--> Running transaction check
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: perl-Git-1.7.1-1.fc14.noarch
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: 1:net-snmp-libs-5.5-15.fc14.x86_64
---> Package perl-CPAN.noarch 0:1.9402-123.fc14 set to be installed
---> Package perl-Compress-Raw-Bzip2.x86_64 0:2.027-1.fc14 set to be installed
--> Finished Dependency Resolution
Error: Package: 1:net-snmp-libs-5.5-15.fc14.x86_64 (@rawhide)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.10.1-118.fc14.x86_64 (@rawhide)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-123.fc14.x86_64 (rawhide)
Not found
Error: Package: perl-Git-1.7.1-1.fc14.noarch (@rawhide)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.10.1-118.fc14.x86_64 (@rawhide)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-123.fc14.x86_64 (rawhide)
Not found
You could try using --skip-broken to work around the problem
You could try running: rpm -Va --nofiles --nodigest

:(

---------- Post added at 01:37 PM CDT ---------- Previous post was at 01:36 PM CDT ----------

HAHA! I just read the error messages I posted. It looks like vim is fixed, but not net-snmp.

dd_wizard

[R3353]

From the posts in the bugzilla's, the AVC denial in Firefox is due to one of the extensions, so, running it in safe-mode will work because it doesn't load up the addons. I really wish I had the extra time to disable the extensions one by one to see which one(s) is causing the problem. The Firefox AVC denial isn't really a problem for me, got tired of the memory leaks etc., and switched to Chrome.

I disabled them all, made permanent, and restarted FF. Same seg fault.

[GoinEasy9]

I disabled them all, made permanent, and restarted FF. Same seg fault.

OK, so now we know they are on the wrong track @ BZ.

I'm gonna hold back gnome-power-manager and see what happens with update....keeping fingers crossed.

[dd_wizard]

@R3353: Do you have selinux set to permissive? If not, change the line SELINUX=enforcing to SELINUX=permissive in /etc/selinux/config. I think that stopped the crashes for one of the posters to this thread. You'll still get the AVC denial messages, but they won't be enforced.

dd_wizard

[GoinEasy9]

Let's see, I still have a desktop.

Installing gtk3-engines did make OOO look a whole lot better than it did after new version came in.
gedit still segfaults when closed.
System-->Preferences-->Appearance doesn't come up.
Yes, nautilus looks funky...interesting, that's the way OOO looked before i installed gtk3-engines...gonna reboot.

---------- Post added at 06:00 PM CDT ---------- Previous post was at 05:34 PM CDT ----------

After reboot, nautilus looks ok (logout/login would have worked also). Guess one needs gtk3-engines.
gnome-commander root mode not opening files with F4. (need to file a BZ about that).
Add//Remove Software got a gpk-update-icon crash the first time, not the second time, hmmm.
KDE apps like Konversation & k3b still are wrapped up in beige. gtk3-engines not helping there yet.
Evince is working and WOW so is Google Earth.
Clicking on the Update icon on the panel flashes something but doesn't bring up software update.

OK, looking at more later.

[R3353]

@R3353: Do you have selinux set to permissive? If not, change the line SELINUX=enforcing to SELINUX=permissive in /etc/selinux/config. I think that stopped the crashes for one of the posters to this thread. You'll still get the AVC denial messages, but they won't be enforced.

dd_wizard

Yes, selinux is almost always enforcing on my boxes. One thing that is quite odd is that I am able to install a FF theme in permissive mode, then set back to enforcing, reboot, start FF in safe-mode, and the theme remains. I can even install new ones despite safe-mode claiming themes are disabled.

[dd_wizard]

Hmm, my nautilus still looks funky. :( I installed the gtk3 engines, maybe I missed something else? Here's what I have:
Installed Packages
gtk3.x86_64 2.90.3-1.fc14 @rawhide
gtk3-debuginfo.x86_64 2.90.3-1.fc14 @rawhide-debuginfo
gtk3-engines.x86_64 2.90.1-4.fc14 @rawhide
Available Packages
gtk3-devel.x86_64 2.90.3-1.fc14 rawhide
gtk3-devel-docs.x86_64 2.90.3-1.fc14 rawhide
gtk3-engines-debuginfo.x86_64 2.90.1-4.fc14 rawhide-debuginfo
gtk3-engines-devel.x86_64 2.90.1-4.fc14 rawhide
gtk3-immodule-xim.x86_64 2.90.3-1.fc14 rawhide
gtk3-immodules.x86_64 2.90.3-1.fc14 rawhide

Do I need one of the available packages?

On a different note, something is definitely broken in the 2.6.35 kernel's acpi driver. I have a script that runs whenever the laptop changes between ac power and batteries. The script switches to the conservative governor and lowers the maximum cpu frequency to extend battery life on batteries. It automatically goes back to normal governor and maximum frequency on ac power. But on the 35 kernel, the script doesn't fire and "sudo service acpid start|stop|restart" fail. I'll have to look into that.

dd_wizard

[GoinEasy9]

Hey dd_wizard
You have the same gtk3 packages as I do installed. I'm on 32 bit, although I don't know why that would make a difference. Before I installed gtk3-engines, I brought up OOO and nautilus to see how they looked, and, they were swimming in beige. After I installed, they looked more like the gnome theme, with the blue head and off-white background that is the default. I looked to see if there were any other gtk3 packages that might be missing, and, so far, I only came up with the Available packages that you came up with, and didn't install any of those. The only apps that look funky, for me now, are the qt based kde apps.

---------- Post added at 10:19 PM CDT ---------- Previous post was at 10:03 PM CDT ----------

What's very interesting is the admission on the BZ by one person that gnome-power-manager-2.31.3-1.fc14 solved the notification popup during login, yet, there are numerous BZ's that claim that that new version is segfaulting just by booting up. LOL...maybe he noticed that the notification did not appear before it blew up...heh.

[nimnull22]

After latest update Firefox works as usual. On my i686.

How can I find out how is doing or managing XFCE for rawhide, I want to tell some suggestions?

[R3353]

What's very interesting is the admission on the BZ by one person that gnome-power-manager-2.31.3-1.fc14 solved the notification popup during login, yet, there are numerous BZ's that claim that that new version is segfaulting just by booting up. LOL...maybe he noticed that the notification did not appear before it blew up...heh.

the notification has been very inconsistent for me on the last couple gnome-power-managers. I've thought it's been fixed a couple times, only to get prompted on the next reboot.

[GoinEasy9]

the notification has been very inconsistent for me on the last couple gnome-power-managers. I've thought it's been fixed a couple times, only to get prompted on the next reboot.
Interesting, because for me it pops up every time. Once at the login screen, and then again on the desktop after login.

---------- Post added at 11:36 PM CDT ---------- Previous post was at 09:39 PM CDT ----------


How can I find out how is doing or managing XFCE for rawhide, I want to tell some suggestions?
If you look at the package in koji, doesn't it list the maintainers?

---------- Post added at 11:53 PM CDT ---------- Previous post was at 11:36 PM CDT ----------

Update tonight brought in new Firefox, still gets AVC denials. Abrt still can't find debuginfos, gedit still segfaults when closing, SSDD. Still holding back gnome-power-manager. Geez.

[nimnull22]

Have to say sorry for confuse you all.
Last update did not solve Firefox problem.
I did. I was trying to fix selinux skype denial. And execute "setsebool -P allow_execstack 1"
It allows skype to start and work, and also fix firefox problem.
Later I decided to return policy back - because it is not a solution. The same way we can just turn selinux off.
Sorry about it.

About XFCE, I filled a bug about some xkb-plugin.

[gid99]

I just got an email from the devel mailing list. Since it is public, I'll post it:


I guess it's fair warning, maybe one should wait till tomorrow to do any updates.

---------- Post added at 05:01 PM CDT ---------- Previous post was at 03:23 PM CDT ----------

The next message explains what happens if you do an update:


In order words....WAIT...LOL

I'm not a fan of KDE - it all started when I was fortunate enough to go to Red Hat training (and I am sure I need more), and there I learned that certain things in command line do not work in KDE (not well known, only a few things, most not readily apparent especially in a normal non-server setting). When I called the Red Hat instructor over and showed him how the commands in shell were NOT working, he noticed I was the only one in the class using KDE (I diverged from the book), he said "try following the book and use gnome with this and see what happens" (and the book had said to use gnome). And (drum roll) it worked in gnome. odd. He didn't/wouldn't elaborate further.

Unfortunately I dumped gnome-panel in my zeal to get my weekly yum update, thinking I could bring it back. and now I can't, and I have no gnome panels. The only way I could get a terminal was to launch dangermouse's autoten and then init 3/init 5 --> switch to kde.

So for me, I'm going to "wait it out using KDE" <palming my face>. I'm not a fan of KDE. Mercifully things work for me in this environment and I still run my virtual systems in gnome.

waiting.
I heard Linus Torvalds switched to gnome (http://www.computerworld.com/s/article/9126619/Q_A_Linux_founder_Linus_Torvalds_talks_about_open_ source_identity?taxonomyId=18&pageNumber=1&taxonomyName=Software) a while back.

.

[GoinEasy9]

@gid99
Why don't you go into /var/log/yum.log - see what apps were removed, then, reinstall them along with gnome-panel, which has problems, but is definitely installable.

[gid99]

@gid99
Why don't you go into /var/log/yum.log - see what apps were removed, then, reinstall them along with gnome-panel, which has problems, but is definitely installable.

Thanks,

I'll check it out.

I did try reinstalling it with --skip-broken

I'll try your suggestion

[GoinEasy9]

I have 3 updates pending today.
dconf
glib2
and gnome-power-manager that I've been holding back.
LOL...could be really good, could be really bad....waiting for a call back from my astrologer, I'll let you know what happens.

---------- Post added at 08:52 PM CDT ---------- Previous post was at 06:35 PM CDT ----------

Installed dconf and glib2...the stars said it was to risky to play with gnome-power-manager.

[R3353]

GoinEasy9, you might go ahead and let power manager update and see if the authentication prompts pre/post boot persist.

[dd_wizard]

I updated gnome-power-manager yesterday, and it's working correctly. However, this error in /etc/log/boot.log isn't good.
/etc/rc5.d/S26udev-post: line 21: /etc/sysconfig/udev: No such file or directory
Retrigger failed udev events [ OK ]

[gene@Mobile-PC init.d]$ head -n21 /etc/rc.d/init.d/udev-post | tail -n1
. /etc/sysconfig/udev

So the udev init script is sourcing a file that no longer exists. There is a file named /etc/udev/udev.conf that has a creation date/time that matches the last update to udev. It looks like initscripts didn't get updated for the new udev package. I wonder which package should get the bug report?

dd_wizard

[nimnull22]

Update 28.06.10


Installed:
perl-IO-Compress.noarch 0:2.027-1.fc14

Dependency Installed:
perl-Compress-Raw-Bzip2.i686 0:2.027-1.fc14 python-markupsafe.i686 0:0.9.2-3.fc14

Updated:
cairo.i686 0:1.9.10-1.fc14 glib2.i686 0:2.25.10-2.fc14
glib2-devel.i686 0:2.25.10-2.fc14 gstreamer.i686 0:0.10.29.2-2.fc14
gstreamer-plugins-base.i686 0:0.10.29.2-1.fc14 gstreamer-plugins-good.i686 0:0.10.23.2-1.fc14
gstreamer-tools.i686 0:0.10.29.2-2.fc14 libXfont.i686 0:1.4.2-1.fc14
net-snmp.i686 1:5.5-16.fc14 net-snmp-libs.i686 1:5.5-16.fc14
perl.i686 4:5.12.1-123.fc14 perl-Compress-Raw-Zlib.i686 0:2.027-1.fc14
perl-File-BaseDir.noarch 0:0.03-7.fc14 perl-File-DesktopEntry.noarch 0:0.04-11.fc14
perl-HTML-Parser.i686 0:3.64-3.fc14 perl-HTML-Tagset.noarch 0:3.20-5.fc14
perl-Module-Pluggable.noarch 1:3.90-123.fc14 perl-Pod-Escapes.noarch 1:1.04-123.fc14
perl-Pod-Simple.noarch 1:3.13-123.fc14 perl-URI.noarch 0:1.54-2.fc14
perl-XML-Parser.i686 0:2.36-9.fc14 perl-libs.i686 4:5.12.1-123.fc14
perl-libwww-perl.noarch 0:5.834-1.fc14 perl-version.noarch 3:0.82-123.fc14
psacct.i686 0:6.5.4-5.fc14 python-beaker.noarch 0:1.5.3-3.fc14
python-mako.noarch 0:0.3.4-1.fc14 transmission-common.i686 0:2.01-2.fc14
transmission-gtk.i686 0:2.01-2.fc14 un-core-dotum-fonts.noarch 0:1.0.2-0.15.080608.fc14
un-core-fonts-common.noarch 0:1.0.2-0.15.080608.fc14 vim-common.i686 2:7.2.444-2.fc14
vim-enhanced.i686 2:7.2.444-2.fc14 xchat.i686 1:2.8.8-3.fc14

Skipped (dependency problems):
atlas.i686 0:3.8.3-15.fc13 libebml.i686 0:1.0.0-1.fc14 libgfortran.i686 0:4.4.4-9.fc14 libmatroska.i686 0:1.0.0-1.fc14
opencv.i686 0:2.1.0-2.fc14

Replaced:
perl-Compress-Zlib.i686 0:2.020-118.fc14 perl-IO-Compress-Base.i686 0:2.020-118.fc14
perl-IO-Compress-Zlib.i686 0:2.020-118.fc14



Some errors in xsession-errors on gnome, I tried to start "appearance".

(gsettings-data-convert:1650): GLib-GIO-CRITICAL **: g_settings_new_with_path: assertion `path != NULL' failed
(gsettings-data-convert:1650): GLib-GIO-CRITICAL **: g_settings_delay: assertion `G_IS_SETTINGS (settings)' failed

(gnome-settings-daemon:1659): GdkPixbuf-CRITICAL **: gdk_pixbuf_format_get_name: assertion `format != NULL' failed
Gtk-Message: Failed to load module "pk-gtk-module": libpk-gtk-module.so: cannot open shared object file: No such file or directory
** Message: NumLock remembering disabled because hostname is set to "localhost"

(gnome-settings-daemon:1659): GdkPixbuf-CRITICAL **: gdk_pixbuf_format_get_name: assertion `format != NULL' failed
gnome-session[1620]: WARNING: Application 'gsettings-data-convert.desktop' failed to register before timeout
Gtk-Message: Failed to load module "pk-gtk-module": libpk-gtk-module.so: cannot open shared object file: No such file or directory

(polkit-gnome-authentication-agent-1:1686): GLib-GObject-WARNING **: cannot register existing type `_PolkitError'

(polkit-gnome-authentication-agent-1:1686): GLib-CRITICAL **: g_once_init_leave: assertion `initialization_value != 0' failed
Gtk-Message: Failed to load module "pk-gtk-module": libpk-gtk-module.so: cannot open shared object file: No such file or directory

MCS->Xfconf settings migration complete

Initializing nautilus-gdu extension
compiz (video) - Warn: No 8 bit GLX pixmap format, disabling YV12 image format

(nautilus:1684): GdkPixbuf-CRITICAL **: gdk_pixbuf_format_get_name: assertion `format != NULL' failed

Gtk-ERROR **: GTK+ 2.x symbols detected. Using GTK+ 2.x and GTK+ 3 in the same process is not supported
aborting...

Gtk-ERROR **: GTK+ 2.x symbols detected. Using GTK+ 2.x and GTK+ 3 in the same process is not supported
aborting...

[dd_wizard]

I think the libpk-gtk-module.so errors are related to the fact that the module has never been compiled for gtk3. So the more apps that get ported to gtk3, the more errors we'll see. I filed one bug about it, but it was against a particular app, not gtk3. Hopefully this gets some attention soon, since there isn't a gtk3 component in bugzilla to file against. :rolleyes: :p

dd_wizard

[GoinEasy9]

OK, finally let gnome-power-manager update, and it did fix the notifications, and it also didn't cause any other problems, but, then again, it's late, just got home from work, so I'll play with it more tomorrow.

[nimnull22]

Update 29.06.10

Strange, I got:
--> Finished Dependency Resolution
Error: Package: vlc-1.1.0-0.11.rc2.fc14.i686 (@rpmfusion-free-rawhide)
Requires: libcxcore.so.2.0
Removing: opencv-2.0.0-10.fc14.i686 (@rawhide)
libcxcore.so.2.0
Updated By: opencv-2.1.0-2.fc14.i686 (rawhide)
Not found
Error: Package: 1:control-center-2.31.3-2.fc14.i686 (@rawhide)
Requires: libgdk_pixbuf-3.0.so.0
Removing: gtk3-2.90.3-1.fc14.i686 (@rawhide)
libgdk_pixbuf-3.0.so.0
Updated By: gtk3-2.90.4-1.fc14.i686 (rawhide)
Not found
Error: Package: vlc-1.1.0-0.11.rc2.fc14.i686 (@rpmfusion-free-rawhide)
Requires: libcvaux.so.2.0
Removing: opencv-2.0.0-10.fc14.i686 (@rawhide)
libcvaux.so.2.0
Updated By: opencv-2.1.0-2.fc14.i686 (rawhide)
Not found
Error: Package: vlc-core-1.1.0-0.11.rc2.fc14.i686 (@rpmfusion-free-rawhide)
Requires: libmatroska.so.0
Removing: libmatroska-0.9.0-1.fc14.i686 (@rawhide)
libmatroska.so.0
Updated By: libmatroska-1.0.0-1.fc14.i686 (rawhide)
Not found
Error: Package: vlc-1.1.0-0.11.rc2.fc14.i686 (@rpmfusion-free-rawhide)
Requires: libml.so.2.0
Removing: opencv-2.0.0-10.fc14.i686 (@rawhide)
libml.so.2.0
Updated By: opencv-2.1.0-2.fc14.i686 (rawhide)
Not found
Error: Package: 1:openoffice.org-core-3.2.1-19.18.fc14.i686 (@rawhide)
Requires: libgdk_pixbuf-3.0.so.0
Removing: gtk3-2.90.3-1.fc14.i686 (@rawhide)
libgdk_pixbuf-3.0.so.0
Updated By: gtk3-2.90.4-1.fc14.i686 (rawhide)
Not found
Error: Package: 1:java-1.6.0-openjdk-1.6.0.0-41.b18.fc13.i686 (@rawhide)
Requires: libjpeg = 6b
Removing: libjpeg-6b-46.fc12.i686 (@anaconda-InstallationRepo-200911081854.i386/12)
libjpeg = 6b-46.fc12
Obsoleted By: libjpeg-turbo-utils-0.0.93-9.fc14.i686 (rawhide)
Not found
Error: Package: vlc-1.1.0-0.11.rc2.fc14.i686 (@rpmfusion-free-rawhide)
Requires: libcv.so.2.0
Removing: opencv-2.0.0-10.fc14.i686 (@rawhide)
libcv.so.2.0
Updated By: opencv-2.1.0-2.fc14.i686 (rawhide)
Not found
Error: Package: vlc-1.1.0-0.11.rc2.fc14.i686 (@rpmfusion-free-rawhide)
Requires: libhighgui.so.2.0
Removing: opencv-2.0.0-10.fc14.i686 (@rawhide)
libhighgui.so.2.0
Updated By: opencv-2.1.0-2.fc14.i686 (rawhide)
Not found
Error: Package: libwmf-0.2.8.4-22.fc14.i686 (@rawhide/13)
Requires: /usr/bin/update-gdk-pixbuf-loaders
Removing: gtk2-2.21.2-2.fc14.i686 (@rawhide)
Not found
Updated By: gtk2-2.21.3-1.fc14.i686 (rawhide)
Not found
Error: Package: vlc-core-1.1.0-0.11.rc2.fc14.i686 (@rpmfusion-free-rawhide)
Requires: libebml.so.0
Removing: libebml-0.8.0-1.fc14.i686 (@rawhide)
libebml.so.0
Updated By: libebml-1.0.0-1.fc14.i686 (rawhide)
Not found

It helps:
yum update --skip-broken --exclude=gdk-pixbuf2.i686 --exclude=libjpeg-turbo.i686 --exclude=systemd-units.i686 --exclude=libgfortran.i686

At least I can check new kernel 35-0.13.


And result is:
Installed:
kernel-PAE.i686 0:2.6.35-0.13.rc3.git2.fc14

Dependency Installed:
dialog.i686 0:1.1-11.20100428.fc14

Updated:
ModemManager.i686 0:0.4-2.git20100628.fc14 alsa-utils.i686 0:1.0.23-3.fc14 at-spi2-core.i686 0:0.3.4-1.fc14
avahi-autoipd.i686 0:0.6.26-3.fc14 bind-libs.i686 32:9.7.1-1.fc14 bind-utils.i686 32:9.7.1-1.fc14
curl.i686 0:7.21.0-2.fc14 elfutils.i686 0:0.148-1.fc14 elfutils-libelf.i686 0:0.148-1.fc14
elfutils-libs.i686 0:0.148-1.fc14 jpackage-utils.noarch 0:1.7.5-3.11.fc14 kernel-headers.i686 0:2.6.35-0.13.rc3.git2.fc14
libcurl.i686 0:7.21.0-2.fc14 logrotate.i686 0:3.7.9-1.fc14 mtr.i686 2:0.79-1.fc14
openldap.i686 0:2.4.22-3.fc14 perf.noarch 0:2.6.35-0.13.rc3.git2.fc14 pyatspi.noarch 0:0.3.4-1.fc14
ql23xx-firmware.noarch 0:3.03.28-1.fc14 ql2400-firmware.noarch 0:5.03.07-1.fc14 ql2500-firmware.noarch 0:5.03.07-1.fc14
selinux-policy.noarch 0:3.8.6-1.fc14 selinux-policy-targeted.noarch 0:3.8.6-1.fc14 telepathy-gabble.i686 0:0.9.14-1.fc14
xsane-common.i686 0:0.997-7.fc14

Skipped (dependency problems):
atlas.i686 0:3.8.3-15.fc13 avahi.i686 0:0.6.26-3.fc14 avahi-glib.i686 0:0.6.26-3.fc14
avahi-gobject.i686 0:0.6.26-3.fc14 avahi-libs.i686 0:0.6.26-3.fc14 avahi-ui.i686 0:0.6.26-3.fc14
cups.i686 1:1.4.4-5.fc14 cups-libs.i686 1:1.4.4-5.fc14 evince.i686 0:2.31.3-5.20100621git.fc14
evince-djvu.i686 0:2.31.3-5.20100621git.fc14 evince-libs.i686 0:2.31.3-5.20100621git.fc14 gcalctool.i686 0:5.31.4-1.fc14
gimp.i686 2:2.6.9-3.fc14 gimp-help-browser.i686 2:2.6.9-3.fc14 gimp-libs.i686 2:2.6.9-3.fc14
gnome-color-manager.i686 0:2.31.3-3.fc14 gnome-desktop3.i686 0:2.90.2-1.fc14 gnome-packagekit.i686 0:2.31.3-2.fc14
gnome-power-manager.i686 0:2.31.3-2.fc14 gnome-settings-daemon.i686 0:2.31.3-3.fc14 gtk2.i686 0:2.21.3-1.fc14
gtk2-immodule-xim.i686 0:2.21.3-1.fc14 gtk3.i686 0:2.90.4-1.fc14 ibus.i686 0:1.3.5-3.fc14
ibus-gtk2.i686 0:1.3.5-3.fc14 ibus-gtk3.i686 0:1.3.5-3.fc14 ibus-libs.i686 0:1.3.5-3.fc14
libcanberra.i686 0:0.25-2.fc14 libcanberra-gtk2.i686 0:0.25-2.fc14 libcanberra-gtk3.i686 0:0.25-2.fc14
libebml.i686 0:1.0.0-1.fc14 libjpeg-turbo-utils.i686 0:0.0.93-9.fc14 libmatroska.i686 0:1.0.0-1.fc14
libnotify.i686 0:0.5.0-1.fc14 librsvg2.i686 0:2.31.0-0.1.20100628git.fc14 mousetweaks.i686 0:2.31.4-1.fc14
nautilus.i686 0:2.31.4-1.fc14 nautilus-extensions.i686 0:2.31.4-1.fc14 nspluginwrapper.i686 0:1.3.0-12.fc14
opencv.i686 0:2.1.0-2.fc14 rhythmbox.i686 0:0.12.90-2.fc14 xsane.i686 0:0.997-7.fc14
xsane-gimp.i686 0:0.997-7.fc14

[dd_wizard]

OK, this dependency problem really sucks. The blocked packages really need to be updated, but I don't think it's safe to erase the blockers. Oh well, when the automagic dep solver is finished, this shouldn't happen any more. ;)

dd_wizard

[nimnull22]

DO NOT UPDATE kernel to 2.6.35-0.13.rc3.git2

My keyboard on laptop doesn't work, so I can't type password and press enter, change tty, ....

Booted from previous kernel 0.2.rc3.git0

Be careful everyone.

[dd_wizard]

Thanks for the warning. I'm pretty sure the kernel you're using has problems with acpi, so I'm staying away from 2.6.35 until something newer works.

dd_wizard

[tox]

just turn acpi off that way it shouldnt be a problem

it also culd be a keyboard mapping issue?

[nimnull22]

The real question now is how to update and safe previous good kernel?
Yum usually installs new one and deletes third. I do not want it.

Any suggestions please.

[dd_wizard]

Use:
sudo yum --exclude=kernel --exclude=kernel-* --exclude=perf update
If you have never installed kernel-devel, kernel-headers, or perf you can skip the last two excludes.

dd_wizard

---------- Post added at 08:54 AM CDT ---------- Previous post was at 08:38 AM CDT ----------

Hopefully, Bug 609560 (https://bugzilla.redhat.com/show_bug.cgi?id=609560) and Bug 609564 (https://bugzilla.redhat.com/show_bug.cgi?id=609564) will let us update gtk3 soon.

dd_wizard

[kurtdriver]

The real question now is how to update and safe previous good kernel?
Yum usually installs new one and deletes third. I do not want it.

Any suggestions please.

Many people set yum.conf to exclude or protect the kernel. Just cause you want bleeding edge doesn't mean that you want a computer that won't boot.
More at man yum.conf.

[nimnull22]

Many people set yum.conf to exclude or protect the kernel. Just cause you want bleeding edge doesn't mean that you want a computer that won't boot.
More at man yum.conf.


I solve this problem much easy - I removed it by myself to a safe place.

To all.
Next kernel: 2.6.35-0.15.rc3.git3.fc14 has the same problem - keyboard is dead.

Most of yesterday's "excluded" packages, today was successfully updated.

[kurtdriver]

Avahi came today! It's been ages.

[GoinEasy9]

Has anyone else had bad experiences with 2.6.35? The original one seems not to be causing any problems. Could it be Nim's Lappy that has taken a dislike to the new kernel? - I've got to wait till I get home tonight to see for myself.

[dd_wizard]

A lot of stuff made it today, about all I'm missing is the gtk3 update. I think koji may have a working 2.6.35 kernel now. Here's the change log for 2.6.35-0.19.rc3.git4:
Changelog
* Wed Jun 30 2010 Kyle McMartin <kyle@redhat.com> 2.6.35-0.19.rc3.git4
- 2.6.35-rc3-git4

* Tue Jun 29 2010 Chuck Ebbert <cebbert@redhat.com>
- Disable Intel Moorestown support -- it breaks PC keyboard controllers.

dd_wizard

---------- Post added at 11:31 AM CDT ---------- Previous post was at 11:30 AM CDT ----------

[/COLOR]Did I answer your question, GE9? I saw yours after I typed mine! :D

dd_wizard[COLOR="Silver"]

---------- Post added at 12:27 PM CDT ---------- Previous post was at 11:31 AM CDT ----------

[gene@Mobile-PC ~]$ uname -sr
Linux 2.6.35-0.19.rc3.git4.fc14.x86_64

Seems to be working ok with respect to the keyboard, but acpid is still borked. :(

dd_wizard

[kurtdriver]

Has anyone else had bad experiences with 2.6.35? The original one seems not to be causing any problems. Could it be Nim's Lappy that has taken a dislike to the new kernel? - I've got to wait till I get home tonight to see for myself.

My keyboard is fine, a Satellite L450.
The only thing I had to exclude today, on the laptop and desktop as well was libjeg.

[dd_wizard]

The openoffice.org build in koji lets gtk3 update. :) But there are a couple of new broken deps. Oh well, it's progress!

dd_wizard

[fonya]

DO NOT UPDATE kernel to 2.6.35-0.13.rc3.git2
My keyboard on laptop doesn't work, so I can't type password and press enter, change tty, ....


I have this problem too. But my intel kms working right now :) External USB keyboard, and mouse working too.

edit: make quote

[GoinEasy9]

New kernel and gtk3 came in today. Getting a notification window on boot up saying "Could not switch the monitor configuration", doesn't seem to affect anything. Nothing broken tonight, heh, except all the usual stuff that hasn't been fixed yet like abrt, gedit, FF.
Opened OpenOffice writer and wound up with the right monitor mirroring the left monitor, but, not really, was able to erase desktop icons, panels and dup of OOO-writer from right monitor, it was sort of like a picture of the left monitor appeared on the right, very interesting effect. It's probably not a bug, it's a feature :) brought to us by nouveau..lol.
I must say, after seeing gnome-shell working on the Intel based lappy running F13 (it's really usable), I'd love to upgrade to rawhide, but I'm going to wait till rawhide calms down. At least I can play with gnome-shell on a really stable system.

[kurtdriver]

I've lost the sound in my Rawhide laptop sometime over the last few days. I haven't changed anything, could it have been something in the updates?

[nimnull22]

About sound:
I use XFCE with alsa on
00:1e.2 Multimedia audio controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Audio Controller (rev 04), sound does exist. Player: "xterm -e mplayer"

So check everything.

[kurtdriver]

A reboot solved my sound problem. The device was not listed in the volume control preferences. I also didn't have a shutdown option in the System Menu. It seems strange that one boot should be different from another, although I've seen an awful lot over the years. Thanks anyway.

[dd_wizard]

Along those lines, I used to use telinit 1 after updating. But I've had glitches with that trick lately. So far, if things behave oddly after an update, a reboot usually seems to clean things up.

dd_wizard

[GoinEasy9]

Thanks to dd I installed pcsc-lite-1.6.1-4.fc14.i686.rpm & pcsc-lite-libs-1.6.1-4.fc14.i686.rpm from koji before the update, so I missed the problems he explained in the other thread.
--skip-broken worked for the update, missing dependencies libwebkit-1.0.so.2 and libcelt.so.0 prevented the update of celt, webkitgtk, yelp & yelp-libs.
Another new kernel, but, same old, same old. Hmmm, gedit is not blowing up on exit, how'd they sneek that in...LOL

[dd_wizard]

Odd, gedit still seg faults on exit for me. I'm on 64 bit Rawhide, maybe that makes a difference?

dd_wizard

[nimnull22]

I've got it since the first time on 32 bit.

[GoinEasy9]

OK, the gedit fix was a false alarm, but I discovered that by right clicking the Bug Notification Icon (after the first exit from gedit) and choosing quit, exiting gedit doesn't error out on closing until computer is rebooted. This is what fooled me last night. Hey, but at least it's only a PIA once a session.
Now, to do tonight's update, which contains many Xorg packages, LOL, I may or may not BRB.

[kurtdriver]

I'll be darned! Once a session, thanks for relaying that one. Works on 64bit.

[GoinEasy9]

@kurtdriver Interesting find on gedit, which gets me wondering why that shuts the crash off. I restarted the Bug Notification app from the menu, then tried gedit, and it still doesn't crash on exit. So, what exactly is happening then we quit from the icon that's fixing the gedit crash. Considering that abrt is still not finding debuginfo files, it may be abrt that's causing the problem and not gedit. It's not like abrt itself hasn't caused problems before. Oh well.

New Xorg packages bring a non-functioning gnome-shell with nouveau <sigh>. Still have one missing dependency libwebkit-1.0.so.2, --skip-broken works. Ahh, what's a rawhide update without a missing dependency...LOL.
Packages skipped because of dependency problems:
webkitgtk-1.3.2-2.fc14.i686 from rawhide
yelp-2.31.6-2.fc14.i686 from rawhide
yelp-libs-2.31.6-2.fc14.i686 from rawhide

[kurtdriver]

@kurtdriver Interesting find on gedit, which gets me wondering why that shuts the crash off. I restarted the Bug Notification app from the menu, then tried gedit, and it still doesn't crash on exit. So, what exactly is happening then we quit from the icon that's fixing the gedit crash. Considering that abrt is still not finding debuginfo files, it may be abrt that's causing the problem and not gedit. It's not like abrt itself hasn't caused problems before. Oh well.

It might be clever to put it in the startup applications,

gedit
killall gedit

and get it over with at the beginning. That would be neglecting our testing role, however.:)

I've been thinking that it might be ABRT. It's reported crashes in graphical programs that I hadn't tried to run. Gnote, for instance, a program I've never used. Hang on, is that the sound of a yum remove that I hear? It is! It is Gedit gone!

[GoinEasy9]

@kurtdriver
Aww, that's too easy. Besides, I use gedit a lot. Abrt would be the one to delete, but we need it to report bugs. Such is rawhide...LOL.

[kurtdriver]

Oh no, I meant gnote. I got rid of it and then threw in packagekit for good measure. Gedit is useful. In the code tags I should have also run abrt and then killed it too. A joke mind you, but I was bleary eyed and that's the excuse I'm sticking with.

[nimnull22]

Ok, if it is interesting, mine i915GM works fine (I would say - the same) after X update. Have not notice anything unusual, Gnome works as well (I should say - the same).
Unfortunately do not have anything with nvidia to check.

[GoinEasy9]

I can't believe I'm considering buying some ATI cards so I can use them with the Open Source drivers. I've been lucky and the Intel drivers on this lappy aren't problematic like some. But the ATI card in the one desktop was dancing around the Nvidia cards when I was testing it.

[GoinEasy9]

New clutter, new mutter, and a shiny new kernel. Almost tempted to remove abrt and disable SELinux to see how F14 is actually doing. But, with all the transitions going on, I might as well wait. I did see in the one BZ re: FF AVC errors, more and more users have removed addons and have reported that the errors were still there. Another BZ shows Appearances is looking for the wrong version of gtk, and after I found an app that will change the colors of the theme icons :/
It's dangerous not to be able to do so much with rawhide, I've started looking at building a USB3 AMD 6 core box, and if things don't improve faster, it's gonna cost me a whole lot of money.... heh.
Back to F13 and it's boring stability.

[nimnull22]

New clutter, new mutter, and a shiny new kernel. Almost tempted to remove abrt and disable SELinux to see how F14 is actually doing. But, with all the transitions going on, I might as well wait. I did see in the one BZ re: FF AVC errors, more and more users have removed addons and have reported that the errors were still there. Another BZ shows Appearances is looking for the wrong version of gtk, and after I found an app that will change the colors of the theme icons :/
It's dangerous not to be able to do so much with rawhide, I've started looking at building a USB3 AMD 6 core box, and if things don't improve faster, it's gonna cost me a whole lot of money.... heh.
Back to F13 and it's boring stability.

I partially agree with you. Actually I disable abrt daemon long time ago, cause I found easy to check dmesg. I do not like "automatic" stuff anyway.
What for selinux - it is just a question of time, it will work.
And we all have to remember - we are discussing the Rawhide, not even alpha and beta.
Anyway my fxce works just fine, I can do most general things already.
So by now, I forgot about boring stability.

[GoinEasy9]

Can't debug the debugger if the debugger has bugs. Heh. As for SELinux, it's working really well in F13, I feel so safe and secure.
BTW- Word is that a recent update in Sid caused the removal of dhcp client and resolveconf. I watched a couple of chat rooms light up in the past 2 days because no ones networks will work anymore. So it could be worse...LOL. Boring stability is nice when it's on your production machines. I like rawhide as an experimental install.

[nimnull22]

You said about dhcp and reminded me that Fedora doesn't have very nice client/server=dhcpd, which only can work in test mode, to show dhcp offer and change nothing. I couldn't enforce dhclent to just to show without changes.
As long as I do not use any kind of NM, I use my small script for "up and configure" interface, I use a lot dhcpd - just to see what actually I can have. I compiled it for Fedora, it was easy, except for selinux, it needs to be taught.
I also very happy with YUM - it shows what it is going to remove, install and update. So every one who can read and think have a choice.
Actually, after OpenSuse, Fedora like fresh air, I have 13, but I do not use it - I'm waiting for 14.

[nimnull22]

update 09.07.10:

Installed:
kernel-PAE.i686 0:2.6.35-0.31.rc4.git4.fc14

Updated:
acl.i686 0:2.2.49-8.fc14 deltarpm.i686 0:3.6-0.1.20100708git.fc14
farsight2.i686 0:0.0.20-2.fc14 file.i686 0:5.04-11.fc14
file-libs.i686 0:5.04-11.fc14 finger.i686 0:0.17-40.fc14
glx-utils.i686 0:7.9-0.5.fc14 gnome-session.i686 0:2.31.2-2.fc14
gnome-session-xsession.i686 0:2.31.2-2.fc14 gutenprint.i686 0:5.2.5-9.fc14
gutenprint-cups.i686 0:5.2.5-9.fc14 iproute.i686 0:2.6.34-5.fc14
kernel-headers.i686 0:2.6.35-0.31.rc4.git4.fc14 libacl.i686 0:2.2.49-8.fc14
libart_lgpl.i686 0:2.3.21-1.fc14 libvorbis.i686 1:1.3.1-2.fc14
libwmf.i686 0:0.2.8.4-26.fc14 libwmf-lite.i686 0:0.2.8.4-26.fc14
mesa-dri-drivers.i686 0:7.9-0.5.fc14 mesa-libGL.i686 0:7.9-0.5.fc14
mesa-libGLU.i686 0:7.9-0.5.fc14 pcre.i686 0:7.8-4.fc14
pcsc-lite.i686 0:1.6.1-5.fc14 pcsc-lite-libs.i686 0:1.6.1-5.fc14
perf.i686 0:2.6.35-0.31.rc4.git4.fc14 pkgconfig.i686 1:0.25-2.fc14
python-deltarpm.i686 0:3.6-0.1.20100708git.fc14 qt.i686 1:4.7.0-0.26.beta2.fc14
qt-sqlite.i686 1:4.7.0-0.26.beta2.fc14 qt-webkit.i686 1:4.7.0-0.26.beta2.fc14
qt-x11.i686 1:4.7.0-0.26.beta2.fc14 redhat-lsb.i686 0:4.0-5.fc14
sendmail.i686 0:8.14.4-9.fc14 systemd-units.i686 0:2-0.fc14
xorg-x11-drv-evdev.i686 0:2.4.0-3.20100406.fc14 xorg-x11-drv-synaptics.i686 0:1.2.99.1-3.20100617.fc14
xorg-x11-drv-wacom.i686 0:0.10.7-4.20100705.fc14 xorg-x11-font-utils.i686 1:7.4-2.fc14
xorg-x11-server-Xorg.i686 0:1.8.99.904-3.20100702.fc14 xorg-x11-server-common.i686 0:1.8.99.904-3.20100702.fc14
xorg-x11-xkb-utils.i686 0:7.4-8.fc14 ypbind.i686 3:1.32-1.fc14

Skipped (dependency problems):
atlas.i686 0:3.8.3-15.fc13 empathy.i686 0:2.31.3-3.fc14 gimp.i686 2:2.6.9-5.fc14
gimp-help-browser.i686 2:2.6.9-5.fc14 gimp-libs.i686 2:2.6.9-5.fc14 libchamplain.i686 0:0.6.1-1.fc14
libchamplain-gtk.i686 0:0.6.1-1.fc14 libebml.i686 0:1.0.0-1.fc14 libgfortran.i686 0:4.5.0-1.fc14
libmatroska.i686 0:1.0.0-1.fc14 opencv.i686 0:2.1.0-2.fc14 pywebkitgtk.i686 0:1.1.6-4.fc14
webkitgtk.i686 0:1.3.2-2.fc14 yelp.i686 0:2.31.6-2.fc14 yelp-libs.i686 0:2.31.6-2.fc14

I do not know what could be a problem, but this "skipped" bunch is here for a half a month I think.

So after update video works as it was before.
Xfce works, but Gnome mostly doesn't, it is impossible to do most things through GUI, you all can try. Looks like it is closer to downgrade.
Kernel works, "segmentation faults" as usual.

[GoinEasy9]

Hey nimnull22
I just updated also, but I don't seem to have serious problems working in the GUI in Gnome. New Mesa experimental drivers came in, I thought I's get gnome-shell back with nouveau, but that didn't improve.
It's funny, all AVC denials now point to FF. I tried to bring up Blender (I knew it wouldn't work, but figured I'd try) and it got an AVC denial, but when I brought up the report, it was pointing to Firefox, not Blender.
Compiz is working, but Gnome-shell just dies until it brings back the standard graphics.
libchamplain-gtk-0.4.so.0 & libwebkit-1.0.so.2 are the missing dependencies, I've only seen yum complain about them since July 4th.
Oh well, time to do some other things, will look at it again later.

[dd_wizard]

@nimnull22,

I fixed the dependency problems late last week or early this week by deleting a package or two. Looking at yum history, I'm pretty sure deleting shotwell fixed them. Try updating without --skip-broken and we can figure out what's causing the dependency problems. If it's a package you can live without, you can erase it and get a full update.

Also, gnome is working for me on x86_64, but gnome-shell broke on the last mutter update. :(

dd_wizard

[nimnull22]

Ok, I will try to figure out what I can to clean up.
When I said about Gnome, I meant that GUI works incorrectly, for example, "startup application" window (or something like it, I do not remember, as I do not use it for long time, just for a test), I can't switch between two tabs with ONE click, also if I choose some application from the list, buttons on the right are not active, I had to move mouse over them to make them pushable. When I open "Application" drop-down-menu, I see list of application without icons.
All of it was better recently.

[R3353]

my dirty laundry list is still pretty long, but the drop-down menus and what not look okay inside GNOME, and the guis I have tried are okay, although "appearance" is not. Note, many of these depend on perl* stuff.

Skipped (dependency problems):
ImageMagick.i686 0:6.6.2.1-11.fc14
ImageMagick-perl.i686 0:6.6.2.1-11.fc14
cpan-upload.noarch 0:2.2-7.fc14
cpp.i686 0:4.5.0-1.fc14
foomatic.i686 0:4.0.4-12.fc14
foomatic-filters.i686 0:4.0.4-12.fc14
gcc.i686 0:4.5.0-1.fc14
gcc-c++.i686 0:4.5.0-1.fc14
gcc-gfortran.i686 0:4.5.0-1.fc14
gcc-java.i686 0:4.5.0-1.fc14
git.i686 0:1.7.1.1-1.fc14
git-cpan-patch.noarch 0:0.2.1-3.fc14
git-email.noarch 0:1.7.1.1-1.fc14
kdeartwork.i686 0:4.4.92-1.fc14
kdebase.i686 6:4.4.92-1.fc14
kdebase-libs.i686 6:4.4.92-1.fc14
kdebase-runtime.i686 0:4.4.92-1.fc14
kdebase-runtime-flags.noarch 0:4.4.92-1.fc14
kdebase-runtime-libs.i686 0:4.4.92-1.fc14
kdebase-workspace.i686 0:4.4.92-1.fc14
kdebase-workspace-devel.i686 0:4.4.92-1.fc14
kdebase-workspace-libs.i686 0:4.4.92-1.fc14
kdegames.i686 6:4.4.92-2.fc14
kdegames-devel.i686 6:4.4.92-2.fc14
kdegames-libs.i686 6:4.4.92-2.fc14
kdegames-minimal.i686 6:4.4.92-2.fc14
kdegraphics.i686 7:4.4.92-1.fc14
kdegraphics-libs.i686 7:4.4.92-1.fc14
kdemultimedia.i686 6:4.4.92-1.fc14
kdemultimedia-devel.i686 6:4.4.92-1.fc14
kdemultimedia-libs.i686 6:4.4.92-1.fc14
kdenetwork.i686 7:4.4.92-1.fc14
kdenetwork-libs.i686 7:4.4.92-1.fc14
kdesdk.i686 0:4.4.92-2.fc14
kdesdk-libs.i686 0:4.4.92-2.fc14
kdesdk-utils.i686 0:4.4.92-2.fc14
kio_msits.i686 7:4.4.92-1.fc14
ksysguardd.i686 0:4.4.92-1.fc14
ktorrent.i686 0:4.0.2-1.fc14
ktorrent-libs.i686 0:4.0.2-1.fc14
libgcj.i686 0:4.5.0-1.fc14
libgcj-devel.i686 0:4.5.0-1.fc14
libgfortran.i686 0:4.5.0-1.fc14
libgomp.i686 0:4.5.0-1.fc14
libktorrent.i686 0:1.0.2-1.fc14
libmpc.i686 0:0.8.1-1.fc13
libstdc++.i686 0:4.5.0-1.fc14
libstdc++-devel.i686 0:4.5.0-1.fc14
libtool.i686 0:2.2.10-2.fc14
net-snmp.i686 1:5.5-16.fc14
net-snmp-libs.i686 1:5.5-16.fc14
qt.i686 1:4.7.0-0.26.beta2.fc14
qt-devel.i686 1:4.7.0-0.26.beta2.fc14
qt-mysql.i686 1:4.7.0-0.26.beta2.fc14
qt-sqlite.i686 1:4.7.0-0.26.beta2.fc14
qt-webkit.i686 1:4.7.0-0.26.beta2.fc14
qt-webkit-devel.i686 1:4.7.0-0.26.beta2.fc14
qt-x11.i686 1:4.7.0-0.26.beta2.fc14
vim-common.i686 2:7.2.445-1.fc14
vim-enhanced.i686 2:7.2.445-1.fc14

[dd_wizard]

Try to figure out what's blocking perl. I recognize quite a few of those packages, and managed to update them by erasing only a couple of non-essential packages. Do an update without --skip-broken and figure out which packages need the old version of perl. If you can get along without them until they get rebuilt, you can erase them. Even better, file a bug report saying they need to be rebuilt against the new package that they're blocking.

dd_wizard

[R3353]

okay, removing frysk-0.4-26.fc14.i386 and perl-DBD-SQLite-1.29-1.fc14.i686 enabled a --skip-brokenless update to run without error. First in a couple moons.

[dd_wizard]

@R3353: It would be really helpful for the devs if you filed bugs against those packages so they get rebuilt against their new dependencies.

Thanks,
dd_wizard

[R3353]

yes, tried to get to this before leaving home-base for a bit but won't be able to get to it until the 20th or so now.

[dd_wizard]

Since yum thinks it's OK to update gdb today, but the rpm transaction check fails:
Downloading Packages:
Running rpm_check_debug
Running Transaction Test


Transaction Check Error:
file /usr/share/gdb/auto-load/usr/lib64/libstdc++.so.6.0.14-gdb.py from install of gdb-7.1-28.fc14.x86_64 conflicts with file from package libstdc++-4.5.0-2.fc14.x86_64
file /usr/share/gdb/auto-load/usr/lib64/libstdc++.so.6.0.14-gdb.pyc from install of gdb-7.1-28.fc14.x86_64 conflicts with file from package libstdc++-4.5.0-2.fc14.x86_64
file /usr/share/gdb/auto-load/usr/lib64/libstdc++.so.6.0.14-gdb.pyo from install of gdb-7.1-28.fc14.x86_64 conflicts with file from package libstdc++-4.5.0-2.fc14.x86_64

Error Summary
-------------

using --skipbroken may not work. However, this works fine:
[gene@Mobile-PC ~]$ sudo yum --nogpgcheck --exclude=gdb update


dd_wizard

[nimnull22]

Yeah, it works, --exclude=gdb helps.
But still, selinux denials skype and firefox. So, let's wait.

[AdamW]

https://bugzilla.redhat.com/show_bug.cgi?id=613710 (gdb conflict)

[kurtdriver]

Yeah, it works, --exclude=gdb helps.
But still, selinux denials skype and firefox. So, let's wait.

I haven't had that problem with Firefox on either of my machines. Haven't tried Skype, though.
Excluding libstdc++ worked for me today.
I didn't exclude gdb.

[dd_wizard]

@kurtdriver: Check out Adam's bug. The real culprit is gdb, so you might want to download the previous gdb from koji and downgrade it, then update libstdc++. Although it should all get straightened out in a few days.

dd_wizard

[GoinEasy9]

I was able to upgrade libstdc++, but had to exclude gdb. I guess if you excluded libstdc++ it held back gdb.
Updated: libstdc++-4.5.0-2.fc14.i686
Still used --skip-broken libwebkit-1.0.so.2 is still missing.

[dd_wizard]

Today's update works cleanly if you download the newest gnome-shell from koji, then in the folder you put it in, run:

sudo yum --nogpgcheck update mutter ./gnome-shell*

then for the rest of the fedora updates...

sudo yum --nogpgcheck --disablerepo=rpm* --exclude=gdb update

I'm still working on getting the RPMFusion updates, but I may just wait for gstreamer-plugins-ugly to be rebuilt.

dd_wizard

[SlowJet]

Hi guys,

I had to drop my isp but I'm borrowing a wireless router for a bit to update things.

I wanted to do
yum install 'Web Server' 'MySQL Database'
to get all the parts but perl is in the way.
My question is
What the status of perl-5.12.1 and the perl(compat) 5.10.0 5.10.1

I can get everything for 5.12.1 except DBI and compat stuff but then several programs (installed are asking for compat.

Any comments or thoughts would be helpful.
Otherwise rawhide boots and runs good less the broken programs.

thanks,
SJ

[nimnull22]

For dd_wizard

I have noticed, that after recent update udev doesn't show error during the boot.
BZ 608838. At least I do not see it anymore.


Additionally:
I decide to check "gedit", as usual I start it from xterm and got:
$ gedit

** (gedit:3130): WARNING **: Private typelib 'Gedit-3.0' not found
libpeas-Message: Error opening directory '/home/vas/.local/share/gedit/plugins': No such file or directory

(gedit:3130): libpeas-WARNING **: Type not found in introspection: 'GeditWindowActivatable'

...

(gedit:3130): libpeas-WARNING **: Type not found in introspection: 'GeditViewActivatable'

(gedit:3130): libpeas-WARNING **: Type not found in introspection: 'GeditWindowActivatable'

...

(gedit:3130): libpeas-WARNING **: Type not found in introspection: 'GeditWindowActivatable'

(gedit:3130): libpeas-WARNING **: Type not found in introspection: 'GeditWindowActivatable'


(gedit:3130): libpeas-WARNING **: Type not found in introspection: 'GeditViewActivatable'

(gedit:3130): libpeas-WARNING **: Type not found in introspection: 'GeditWindowActivatable'


But, there is no any AVC errors on exit. (in xfce)

[dd_wizard]

@SlowJet:
I remember that issue a while ago. If I recall correctly, someone filed a bug against the package(s) blocking perl. I'm thinking that something you have installed wasn't covered and you'll need to file a bug. What's the dependency error look like?

@nimnull22:
I'm seeing the same errors in gnome/openox. At least gedit isn't seg faulting on exit anymore. Smile at the good things. :)

dd_wizard

[SlowJet]

@SlowJet:
I remember that issue a while ago. If I recall correctly, someone filed a bug against the package(s) blocking perl. I'm thinking that something you have installed wasn't covered and you'll need to file a bug. What's the dependency error look like?

@nimnull22:
I'm seeing the same errors in gnome/openox. At least gedit isn't seg faulting on exit anymore. Smile at the good things. :)

dd_wizard

Well, this is what is on rawhide for now, but there is a newer perl on koji today.
This does not look to bad but trying to get perl-5.12.1 with this as a llocalgroupinstall breaks more perl.
But even so with out the compat stuff it doesn't seem like it is going to work?

# yum groupinstall 'MySQL Database' 'Web Server'
Loaded plugins: presto, refresh-packagekit
Setting up Group Process
Package mysql-5.1.48-3.fc14.i686 already installed and latest version
Package httpd-2.2.15-3.fc14.i686 already installed and latest version
Package 1:mod_ssl-2.2.15-3.fc14.i686 already installed and latest version
Package distcache-1.4.5-21.i686 already installed and latest version
Package httpd-manual-2.2.15-3.fc14.noarch already installed and latest version
Resolving Dependencies
--> Running transaction check
---> Package MySQL-python.i686 0:1.2.3-0.4.c1.fc13 set to be installed
---> Package crypto-utils.i686 0:2.4.1-27 set to be installed
--> Processing Dependency: perl(Newt) for package: crypto-utils-2.4.1-27.i686
---> Package libdbi-dbd-mysql.i686 0:0.8.3-5.fc12 set to be installed
--> Processing Dependency: libdbi-drivers = 0.8.3-5.fc12 for package: libdbi-dbd-mysql-0.8.3-5.fc12.i686
--> Processing Dependency: libdbi.so.0 for package: libdbi-dbd-mysql-0.8.3-5.fc12.i686
---> Package mod_perl.i686 0:2.0.4-11.fc14 set to be installed
--> Processing Dependency: perl(BSD::Resource) for package: mod_perl-2.0.4-11.fc14.i686
--> Processing Dependency: perl(ExtUtils::MakeMaker) for package: mod_perl-2.0.4-11.fc14.i686
---> Package mod_python.i686 0:3.3.1-11 set to be installed
---> Package mysql-connector-odbc.i686 0:5.1.5r1144-7.fc13 set to be installed
---> Package mysql-server.i686 0:5.1.48-3.fc14 set to be installed
--> Processing Dependency: perl-DBI for package: mysql-server-5.1.48-3.fc14.i686
--> Processing Dependency: perl(DBI) for package: mysql-server-5.1.48-3.fc14.i686
---> Package perl-DBD-MySQL.i686 0:4.016-1.fc14 set to be installed
--> Processing Dependency: perl(:MODULE_COMPAT_5.12.1) for package: perl-DBD-MySQL-4.016-1.fc14.i686
---> Package php.i686 0:5.3.2-3.fc14 set to be installed
--> Processing Dependency: php-cli = 5.3.2-3.fc14 for package: php-5.3.2-3.fc14.i686
--> Processing Dependency: php-common = 5.3.2-3.fc14 for package: php-5.3.2-3.fc14.i686
---> Package php-ldap.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-mysql.i686 0:5.3.2-3.fc14 set to be installed
--> Processing Dependency: php-pdo for package: php-mysql-5.3.2-3.fc14.i686
---> Package squid.i686 7:3.1.4-2.fc14 set to be installed
---> Package unixODBC.i686 0:2.2.14-11.fc14 set to be installed
---> Package webalizer.i686 0:2.21_02-3 set to be installed
--> Running transaction check
---> Package libdbi.i686 0:0.8.3-3.fc12 set to be installed
---> Package libdbi-drivers.i686 0:0.8.3-5.fc12 set to be installed
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: 1:net-snmp-libs-5.5-15.fc14.i686
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: ack-1.92-1.fc14.noarch
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: perl-WWW-Curl-4.11-3.fc14.i686
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: 2:vim-enhanced-7.2.444-1.fc14.i686
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: 2:vim-X11-7.2.444-1.fc14.i686
--> Processing Dependency: perl = 4:5.12.0-122.fc14 for package: 1:perl-Pod-Escapes-1.04-122.fc14.noarch
--> Processing Dependency: perl = 4:5.12.0-122.fc14 for package: 4:perl-libs-5.12.0-122.fc14.i686
--> Processing Dependency: perl = 4:5.12.0-122.fc14 for package: 1:perl-Pod-Simple-3.13-122.fc14.noarch
--> Processing Dependency: perl = 4:5.12.0-122.fc14 for package: 1:perl-Module-Pluggable-3.90-122.fc14.noarch
--> Processing Dependency: perl = 4:5.12.0-122.fc14 for package: 3:perl-version-0.82-122.fc14.noarch
---> Package perl.i686 4:5.12.1-126.fc14 set to be updated
---> Package perl-BSD-Resource.i686 0:1.29.04-1.fc14 set to be installed
---> Package perl-DBI.i686 0:1.611-1.fc14 set to be installed
---> Package perl-ExtUtils-MakeMaker.noarch 0:6.56-126.fc14 set to be installed
--> Processing Dependency: perl(Test::Harness) for package: perl-ExtUtils-MakeMaker-6.56-126.fc14.noarch
--> Processing Dependency: perl-devel for package: perl-ExtUtils-MakeMaker-6.56-126.fc14.noarch
---> Package perl-Newt.i686 0:1.08-26.fc14 set to be installed
---> Package php-cli.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-common.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-pdo.i686 0:5.3.2-3.fc14 set to be installed
--> Running transaction check
--> Processing Dependency: net-snmp-libs = 1:5.5-15.fc14 for package: 1:net-snmp-5.5-15.fc14.i686
---> Package net-snmp-libs.i686 1:5.5-16.fc14 set to be updated
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: ack-1.92-1.fc14.noarch
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: perl-WWW-Curl-4.11-3.fc14.i686
---> Package perl-Module-Pluggable.noarch 1:3.90-126.fc14 set to be updated
---> Package perl-Pod-Escapes.noarch 1:1.04-126.fc14 set to be updated
---> Package perl-Pod-Simple.noarch 1:3.13-126.fc14 set to be updated
---> Package perl-Test-Harness.noarch 0:3.17-126.fc14 set to be installed
--> Processing Dependency: perl(CPAN) for package: perl-Test-Harness-3.17-126.fc14.noarch
---> Package perl-devel.i686 4:5.12.1-126.fc14 set to be installed
--> Processing Dependency: perl(ExtUtils::ParseXS) for package: 4:perl-devel-5.12.1-126.fc14.i686
---> Package perl-libs.i686 4:5.12.1-126.fc14 set to be updated
---> Package perl-version.noarch 3:0.82-126.fc14 set to be updated
---> Package vim-X11.i686 2:7.2.446-1.fc14 set to be updated
--> Processing Dependency: vim-common = 2:7.2.446-1.fc14 for package: 2:vim-X11-7.2.446-1.fc14.i686
---> Package vim-enhanced.i686 2:7.2.446-1.fc14 set to be updated
--> Running transaction check
---> Package net-snmp.i686 1:5.5-16.fc14 set to be updated
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: ack-1.92-1.fc14.noarch
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: perl-WWW-Curl-4.11-3.fc14.i686
---> Package perl-CPAN.noarch 0:1.9402-126.fc14 set to be installed
--> Processing Dependency: perl(Digest::SHA) for package: perl-CPAN-1.9402-126.fc14.noarch
---> Package perl-ExtUtils-ParseXS.noarch 1:2.21-126.fc14 set to be installed
---> Package vim-common.i686 2:7.2.446-1.fc14 set to be updated
--> Running transaction check
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: ack-1.92-1.fc14.noarch
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: perl-WWW-Curl-4.11-3.fc14.i686
---> Package perl-Digest-SHA.i686 1:5.47-126.fc14 set to be installed
--> Finished Dependency Resolution
Error: Package: perl-WWW-Curl-4.11-3.fc14.i686 (@rawhide)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.12.0-122.fc14.i686 (installed)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-126.fc14.i686 (rawhide)
Not found
Error: Package: ack-1.92-1.fc14.noarch (installed)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.12.0-122.fc14.i686 (installed)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-126.fc14.i686 (rawhide)
Not found
You could try using --skip-broken to work around the problem
You could try running: rpm -Va --nofiles --nodigest
[root@Ruthie-07 ~]#

SJ

[dd_wizard]

Error: Package: perl-WWW-Curl-4.11-3.fc14.i686 (@rawhide)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.12.0-122.fc14.i686 (installed)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-126.fc14.i686 (rawhide)
Not found
Error: Package: ack-1.92-1.fc14.noarch (installed)
Requires: perl(:MODULE_COMPAT_5.10.1)
Removing: 4:perl-5.12.0-122.fc14.i686 (installed)
perl(:MODULE_COMPAT_5.10.1)
Updated By: 4:perl-5.12.1-126.fc14.i686 (rawhide)
Not found
You could try using --skip-broken to work around the problem
You could try running: rpm -Va --nofiles --nodigest
SJ
It looks like filing bugs against perl-WWW-Curl and ack will get the problem fixed. They haven't been rebuilt against the new perl yet. That kind of thing seems to be a common problem still.

dd_wizard

---------- Post added at 04:28 PM CDT ---------- Previous post was at 04:20 PM CDT ----------

And there's a new compiz on koji that fixes:

Error: Package: compiz-gnome-0.8.6-1.fc14.x86_64 (rawhide)
Requires: libgnome-window-settings.so.1()(64bit)

dd_wizard

[SlowJet]

This is the same but with --skip-broken (easier to read the results.)
Note MySQL-sever is part of the skipped group.

Also, this is perl-5.12.1 but it is 126 and koji is 128

# yum groupinstall 'Web Server' 'MySQL Database' --skip-broken
Loaded plugins: presto, refresh-packagekit
Setting up Group Process
Package httpd-2.2.15-3.fc14.i686 already installed and latest version
Package httpd-manual-2.2.15-3.fc14.noarch already installed and latest version
Package mysql-5.1.48-3.fc14.i686 already installed and latest version
Resolving Dependencies
--> Running transaction check
---> Package MySQL-python.i686 0:1.2.3-0.4.c1.fc13 set to be installed
<snip>


--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: 1:net-snmp-libs-5.5-15.fc14.i686
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: ack-1.92-1.fc14.noarch
--> Processing Dependency: perl(:MODULE_COMPAT_5.10.1) for package: perl-WWW-Curl-4.11-3.fc14.i686

<snip>


---> Package php-ldap.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-mysql.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-pdo.i686 0:5.3.2-3.fc14 set to be installed
---> Package squid.i686 7:3.1.4-2.fc14 set to be installed
---> Package unixODBC.i686 0:2.2.14-11.fc14 set to be installed
---> Package webalizer.i686 0:2.21_02-3 set to be installed
--> Finished Dependency Resolution
--> Running transaction check
---> Package mod_perl.i686 0:2.0.4-11.fc14 set to be installed
---> Package mysql-server.i686 0:5.1.48-3.fc14 set to be installed
---> Package net-snmp.i686 1:5.5-16.fc14 set to be updated
---> Package perl-BSD-Resource.i686 0:1.29.04-1.fc14 set to be installed
---> Package perl-DBI.i686 0:1.611-1.fc14 set to be installed
--> Processing Dependency: perl(DBI) for package: 7:squid-3.1.4-2.fc14.i686
--> Processing Dependency: perl(DBI) for package: 7:squid-3.1.4-2.fc14.i686
--> Running transaction check
---> Package perl-DBI.i686 0:1.611-1.fc14 set to be installed
--> Finished Dependency Resolution
--> Running transaction check
---> Package MySQL-python.i686 0:1.2.3-0.4.c1.fc13 set to be installed
---> Package compat-db47.i686 0:4.7.25-3.fc14 set to be installed
---> Package crypto-utils.i686 0:2.4.1-27 set to be installed
---> Package distcache.i686 0:1.4.5-21 set to be installed
---> Package libdbi.i686 0:0.8.3-3.fc12 set to be installed
---> Package libdbi-dbd-mysql.i686 0:0.8.3-5.fc12 set to be installed
---> Package libdbi-drivers.i686 0:0.8.3-5.fc12 set to be installed
---> Package mod_python.i686 0:3.3.1-11 set to be installed
---> Package mod_ssl.i686 1:2.2.15-3.fc14 set to be installed
---> Package mysql-connector-odbc.i686 0:5.1.5r1144-7.fc13 set to be installed
---> Package perl-DBI.i686 0:1.611-1.fc14 set to be installed
---> Package perl-Newt.i686 0:1.08-26.fc14 set to be installed
---> Package php.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-cli.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-common.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-ldap.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-mysql.i686 0:5.3.2-3.fc14 set to be installed
---> Package php-pdo.i686 0:5.3.2-3.fc14 set to be installed
---> Package squid.i686 7:3.1.4-2.fc14 set to be installed
---> Package unixODBC.i686 0:2.2.14-11.fc14 set to be installed
---> Package webalizer.i686 0:2.21_02-3 set to be installed
--> Finished Dependency Resolution

Packages skipped because of dependency problems:
mod_perl-2.0.4-11.fc14.i686 from rawhide
mysql-server-5.1.48-3.fc14.i686 from rawhide
1:net-snmp-5.5-16.fc14.i686 from rawhide
1:net-snmp-libs-5.5-16.fc14.i686 from rawhide
4:perl-5.12.1-126.fc14.i686 from rawhide
perl-BSD-Resource-1.29.04-1.fc14.i686 from rawhide
perl-CPAN-1.9402-126.fc14.noarch from rawhide
perl-DBD-MySQL-4.016-1.fc14.i686 from rawhide
perl-DBI-1.611-1.fc14.i686 from rawhide
1:perl-Digest-SHA-5.47-126.fc14.i686 from rawhide
perl-ExtUtils-MakeMaker-6.56-126.fc14.noarch from rawhide
1:perl-ExtUtils-ParseXS-2.21-126.fc14.noarch from rawhide
1:perl-Module-Pluggable-3.90-126.fc14.noarch from rawhide
1:perl-Pod-Escapes-1.04-126.fc14.noarch from rawhide
1:perl-Pod-Simple-3.13-126.fc14.noarch from rawhide
perl-Test-Harness-3.17-126.fc14.noarch from rawhide
4:perl-devel-5.12.1-126.fc14.i686 from rawhide
4:perl-libs-5.12.1-126.fc14.i686 from rawhide
3:perl-version-0.82-126.fc14.noarch from rawhide
2:vim-X11-7.2.446-1.fc14.i686 from rawhide
2:vim-common-7.2.446-1.fc14.i686 from rawhide
2:vim-enhanced-7.2.446-1.fc14.i686 from rawhide

Dependencies Resolved

================================================== ================================================== =========
Package Arch Version Repository Size
================================================== ================================================== =========
Installing:
MySQL-python i686 1.2.3-0.4.c1.fc13 rawhide 85 k
crypto-utils i686 2.4.1-27 rawhide 74 k
distcache i686 1.4.5-21 rawhide 108 k
libdbi-dbd-mysql i686 0.8.3-5.fc12 rawhide 14 k
mod_python i686 3.3.1-11 rawhide 260 k
mod_ssl i686 1:2.2.15-3.fc14 rawhide 84 k
mysql-connector-odbc i686 5.1.5r1144-7.fc13 rawhide 116 k
php i686 5.3.2-3.fc14 rawhide 1.1 M
php-ldap i686 5.3.2-3.fc14 rawhide 32 k
php-mysql i686 5.3.2-3.fc14 rawhide 71 k
squid i686 7:3.1.4-2.fc14 rawhide 1.6 M
unixODBC i686 2.2.14-11.fc14 rawhide 382 k
webalizer i686 2.21_02-3 rawhide 120 k
Installing for dependencies:
compat-db47 i686 4.7.25-3.fc14 rawhide 1.2 M
libdbi i686 0.8.3-3.fc12 rawhide 37 k
libdbi-drivers i686 0.8.3-5.fc12 rawhide 364 k
perl-DBI i686 1.611-1.fc14 rawhide 727 k
perl-Newt i686 1.08-26.fc14 rawhide 71 k
php-cli i686 5.3.2-3.fc14 rawhide 2.2 M
php-common i686 5.3.2-3.fc14 rawhide 514 k
php-pdo i686 5.3.2-3.fc14 rawhide 69 k
Skipped (dependency problems):
mod_perl i686 2.0.4-11.fc14 rawhide 3.2 M
mysql-server i686 5.1.48-3.fc14 rawhide 8.3 M
net-snmp i686 1:5.5-16.fc14 rawhide 293 k
net-snmp-libs i686 1:5.5-16.fc14 rawhide 1.5 M
perl i686 4:5.12.1-126.fc14 rawhide 11 M
perl-BSD-Resource i686 1.29.04-1.fc14 rawhide 34 k
perl-CPAN noarch 1.9402-126.fc14 rawhide 246 k
perl-DBD-MySQL i686 4.016-1.fc14 rawhide 135 k
perl-DBI i686 1.611-1.fc14 rawhide 727 k
perl-Digest-SHA i686 1:5.47-126.fc14 rawhide 58 k
perl-ExtUtils-MakeMaker noarch 6.56-126.fc14 rawhide 289 k
perl-ExtUtils-ParseXS noarch 1:2.21-126.fc14 rawhide 40 k
perl-Module-Pluggable noarch 1:3.90-126.fc14 rawhide 35 k
perl-Pod-Escapes noarch 1:1.04-126.fc14 rawhide 28 k
perl-Pod-Simple noarch 1:3.13-126.fc14 rawhide 207 k
perl-Test-Harness noarch 3.17-126.fc14 rawhide 239 k
perl-devel i686 4:5.12.1-126.fc14 rawhide 427 k
perl-libs i686 4:5.12.1-126.fc14 rawhide 607 k
perl-version noarch 3:0.82-126.fc14 rawhide 48 k
vim-X11 i686 2:7.2.446-1.fc14 rawhide 983 k
vim-common i686 2:7.2.446-1.fc14 rawhide 6.0 M
vim-enhanced i686 2:7.2.446-1.fc14 rawhide 845 k

Transaction Summary
================================================== ================================================== =========
Install 21 Package(s)

Total download size: 9.2 M
Installed size: 28 M
Is this ok [y/N]: N
Exiting on user Command
Complete!

[dd_wizard]

I must have grabbed net-snmp-libs-1:5.5-16.fc14 from koji. It no longer blocks that perl compat update. Maybe ack and perl-WWW-Curl have updates on koji?
[gene@Mobile-PC RPMS]$ yum list installed net-snmp-libs
Loaded plugins: auto-update-debuginfo, fastestmirror, presto, refresh-packagekit, remove-with-
: leaves, show-leaves, verify
Installed Packages
net-snmp-libs.x86_64 1:5.5-16.fc14 @rawhide

dd_wizard

[SlowJet]

Thanks, I'll check things out tomorrow and see what left. :)

SJ

[nimnull22]

To All.
Do not update to kernel 0.41 - I couldn't boot with it.
And also I did not find any logs related to it.
It just hands in the middle of boot.
I am talking about i686.

[GoinEasy9]

I thought I already got -41 a couple of nights ago. I'll check when I get home. It wouldn't be the first kernel to REEK HAVOC.

[kurtdriver]

This is the second boot off it, it works on on my laptop 64bit.

[kurt@Heimdal ~]$ uname -a
Linux Heimdal 2.6.35-0.41.rc5.git1.fc14.x86_64 #1 SMP Fri Jul 16 21:29:45 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux
[kurt@Heimdal ~]$

[dd_wizard]

Odd, I updated to it before I saw your post, nimnull22, and...
[makerpm@Mobile-PC build]$ uname -sr
Linux 2.6.35-0.41.rc5.git1.fc14.x86_64

So it's either 32 bit or hardware specific. Where in the boot process did it fail?

dd_wizard

[nimnull22]

Check them out.

Sorry for the large zip, I just want it to be readable.

Here was an attachment, I deleted it. Problems was solved.

[dd_wizard]

Do you need to use the PAE kernel? I'm not if that would make any difference, though. That's worth posting a bug against the kernel. The [ OK ] never displayed for udevd, so the oops occurred while starting udevd. Which may point to a hardware specific issue, instead of the PAE kernel.

dd_wizard

[GoinEasy9]

I have the PAE kernel, and there doesn't seem to be anything wrong with it. Then again, except for gedit, everything that was broken still is.

[nimnull22]

On koji there is two new kernel: 43 and 44, but both of them has state - "failed"

So I think this is "well known" problem, I am going to wait for a new kernel, however I reported to BZ.

[SlowJet]

$ uname -a
Linux Ruthie-07.WinProxy 2.6.35-0.47.rc5.git2.fc14.i686.PAE #1 SMP Sun Jul 18 14:06:39 UTC 2010 i686 i686 i386 GNU/Linux


47 finally did it.
Near the top of koji so easy to find.

SJ

[nimnull22]

Ooops.

But kernel 2.6.35-0.47, has another "oops" in different place.
So - do not update, or be prepared.

[GoinEasy9]

Kernel 2.6.35-0.47 is not causing me any problems. Still can't play with Gnome shell with nouveau, this time mutter is the problem. https://bugzilla.redhat.com/show_bug.cgi?id=616248

[dd_wizard]

@nimnull22: The change log for

[gene@Mobile-PC ~]$ uname -sr
Linux 2.6.35-0.49.rc5.git2.fc14.x86_64

says it's a fix for an oops. You might want to give it a try.

@GoinEasy9: I've filed 2 bugs for SIGABRT (6) and one for SIGSEGV (11) against mutter since I started testing rawhide. It seems every new mutter still has a problem. :( I hope they get the point! :p

dd_wizard

[nimnull22]

To whom is interested:

Kernel 2.6.35-0.49 works without any errors.
Mesa works on my i915GM
Compiz works awfully in gnome (terminals start to write with artifacts, menus items can't be chosen at ones and so so), but it does something.

[GoinEasy9]

So, kernel 2.6.35-0.49 came in tonight, along with new mesa experimental drivers, and, now I have gnome-shell working again. It even has blender working, which, never worked for me with nouveau before.

Damn, every AVC denial, even though I'm not using Firefox, still points to Firefox saying SELinux is preventing Firefox from making its memory writable and executable.

Got scared, when reverting Desktop Effects from gnome-shell to standard, my Menus were redone and left in the gnome-shell format until I rebooted. At least they reverted back to the standard format again, though I may just leave gnome-shell on if it doesn't blow up like before. I'm hoping this is a turning point for nouveau and gnome-shell, but I'm not holding my breath. It's the first time I've got a working gnome-shell with nouveau in over a month.

I'll play with it more tomorrow, maybe look for some documentation to see if I can make some changes to the default gnome-shell. I have to admit, the alphabetical cell phone like menu is starting to bug me. I need to make better use of the left side of the screen when activities are up, and, I've got to see if I can get my most used apps displayed better. And I've got to reduce the size of the recent activities.

OK, one step at a time, tomorrow.

[nimnull22]

Hi all.

I hope it is not a secret, so if anyone want to run firefox up with all its plugins without selinux complains, one need to type:

runcon -t unconfined_execmem_t -- bash -c "/usr/bin/firefox"

As I was told it gives to bash the right to execute memory, and as long the bash starts firefox, it gets this right also. It works on my rawhide. So you and anyone else can try, it doesn't change default rules.

To: GoinEasy9,

I do not use gnome a lot, so I do not know where to change Desktop Effects. I only can turn compiz on and off.
But where do you change Desktop Effects? I want to check it.

[GoinEasy9]

@nimnull22 System-->Preferences-->Desktop Effects In gnome with the correct compiz and also with gnome-shell installed I have 3 choices: a)standard b)compiz c)gnome-shell.
AND
It was a secret to me, what's still a secret is why all AVC denials come up saying its Firefox's fault. BTW, thanks for the runcon line.

[kyryder]

Ran F14 for the first time in about a month. Only took a hour or so to work out all the dependency problems and get it updated. :)

On the runcon command, I changed it up a bit to only allow execmem on firefox instead of bash:

runcon -t unconfined_execmem_t -- /usr/bin/firefox

Seems to be working fine so far.

Ky

[dd_wizard]

@GE9:
I still can't get gnome-shell to run with the mesa drivers and Intel graphics. And compiz has terrible artifacts, even in a terminal window. So I decided to try openbox/AWN and I like it! I'm still figuring out how to get gnome-power-manager and gnome-volume-control-applet started automatically at login. But after manually starting them, everything seems to work fine. No desktop effects, but I didn't have any with gnome/openbox anyway. ;)

dd_wizard

[nimnull22]

As I said, "runcon" is not my idea. I do not have enough experience with selinux.
My main shield is "iptables". So, I was told about it by Daniel Walsh in BZ story about skype, which is also did not work normal way. But the main idea came from this conversation: http://www.nsa.gov/research/selinux/list-archive/0708/22072.shtml.

To kyryder.
Thanks for the suggestion, I will check this. Because skype also needs some medicine.

To dd_wizard.
Thanks for screen shots, I will read about that openbox. Also I really await of Enlightenment (E17).

[GoinEasy9]

i just installed AWN. I was surprised to see that Firefox is considered an open window. I've only been using chrome for a long time. I looked in top and see no Firefox commands listed. I haven't used the chcon line yet, now I'm more interested why FF is showing up, and maybe its open presence is what's causing the AVC errors. (or am I totally wrong and AWN puts FF into it's WM for convenience sake?)

I see dd uses AWN, I'm going to try and cleanup gnome-shell and see if I can make the activities screen likable. Openbox looks interesting, but, I need to make gnome work,

I have a few links to gnome-shell documentation, but, if anyone has a really good link, I'd appreciate seeing it.

---------- Post added at 03:22 PM CDT ---------- Previous post was at 03:21 PM CDT ----------

LOL...I'm working against time here...one more update and I might not have gnome-shell again....Heh

BTW- First yum update without --skip-broken in a long time. \o/ yay \o/

[dd_wizard]

I had to add a launcher for FF to AWN, so I think that means you have a copy of FF running, GE9. Do you see it with "ps -ef | grep -i firefox"?

dd_wizard

[GoinEasy9]

[GoinEasy9@Fedora14dw32 ~]$ ps -ef | grep -i firefox
500 2755 2737 0 16:38 pts/0 00:00:00 grep --color=auto -i firefox

Funny, it's not showing anywhere else. Even if I do "ps -ef" it doesn't appear in the terminal. You've got wizardly ways, don't you? :)

[GoinEasy9]

I finally got AWN up again, and before it blew up, I noticed that AWN placed FF into the list of apps that would appear automatically. Removing it, solved the problem...heh..until I tried to configure AWN some more and it blew up. It may not be nouveau, it may be all the transitioning from gnome 2 to 3 thats causing all the problems. Gnome apps that work in control-center that don't have links to icons, icons that point to apps that have disappeared. To much going on to experiment with it seriously. At least I can play with it on F13.

[GoinEasy9]

Systemd will now be the default in rawhide. I'm assuming this will happen automatically upon update when systemd-4-3 hits the repo, it's now in koji. This ought to be interesting.

[nimnull22]

I expect a lot of excitement after todays update:
systemd-4-3.fc14
----------------
* Sat Jul 24 2010 Lennart Poettering <lpoetter at redhat.com> - 4-3
- Add libselinux to build dependencies

* Sat Jul 24 2010 Lennart Poettering <lpoetter at redhat.com> - 4-2
- Use the right tarball

* Sat Jul 24 2010 Lennart Poettering <lpoetter at redhat.com> - 4-1
- New upstream release, and make default


upstart-0.6.5-7.fc14
--------------------
* Sat Jul 24 2010 Casey Dahlin <cdahlin at redhat.com> - 0.6.5-7
- Make upgrades work correctly now that -sysvinit is split off

* Sat Jul 24 2010 Lennart Poettering <lpoetter at redhat.com> - 0.6.5-6
- Split off -sysvinit to make Upstart parallel installable with systemd

I will update later today and will tell how things go.

[dd_wizard]

Today's update failed with conflicts between the systemd and upstart updates. :rolleyes: Excluding systemd* would have worked, but I noticed that the new upstart has some functionality split out into the new systemd. So, for safety I updated with "--exclude systemd* --exclude=upstart*". I'm pretty sure you don't want to update either one until the conflicts are resolved.

dd_wizard

[GoinEasy9]

That conflict was noticed last night on the devel list, did you try removing then re-installing it? Once again I have to wait till I get home to play with it.

[dd_wizard]

If you mean removing systemd, then installing it with the upstart update, I haven't tried that. Is that what they suggested? Rebooting without upstart might be fun. :p

Thanks,
dd_wizard

[GoinEasy9]

Well, I spoke too soon before asking if you had experimented with it earlier. They gave no fix yet on the list. I was thinking that recent changes they made to systemd might be responsible for the dependency error. So, if you experimented with it earlier just removing it and reinstalling would fix the dependency. I'm at work and can't try it. I figure, since I didn't experiment with it earlier, I wouldn't get the dependency error when I updated.
Did that make sense?

[dd_wizard]

I think so. :p But I've never tried systemd yet, so the conflict should happen to you, too. And as I said, the change log for upstart says part of the functionality has split out into systemd. So, I'd think it would be risky to exclude systemd and let upstart update. For now, I'm excluding both of them.

dd_wizard

[nimnull22]

Yes, "--skip-broken" deprives me of some fun. I am upset.
I think developers have to be more attentive, may be.

Skipped (dependency problems):
upstart.i686 0:0.6.5-7.fc14
systemd.i686 0:4-3.fc14

But we have some time to explore something, for example:
/lib/systemd
/usr/share/doc/systemd
/etc/systemd

[GoinEasy9]

I guess that's one theory shot down. One post on the list blamed it on recent changes, Oh well, guess we'll have to wait. Interesting round robin dependency:
> ---> Package upstart-sysvinit.x86_64 0:0.6.5-7.fc14 set to be installed
> --> Processing Conflict: systemd-sysvinit-4-3.fc14.x86_64 conflicts
> upstart-sysvinit
> --> Processing Conflict: upstart-sysvinit-0.6.5-7.fc14.x86_64 conflicts
> systemd-sysvinit
(Not my printout, from the list)

Trying to figure out...if upstart is the chicken, is systemd the egg???? :)

[nimnull22]

I think this is also important (from the list):
...
- You can boot into either of them by setting the "init=" kernel cmdline
option according to your wishes. If you pass "init=/bin/systemd" you
will boot into systemd, if you pass "init=/sbin/upstart" you will boot
into upstart (note the /sbin vs. /bin!)
...
So, to put this in shorter words:

If systemd does not work for you and you need a temporary fix, pass
"init=/bin/upstart" on the kernel command line.

If systemd does not work for you and you need a non-temporary fix,
install "upstart-sysvinit".

[GoinEasy9]

By George I think you've got it....LOL...now all they have to do is get them to install.

---------- Post added at 08:38 PM CDT ---------- Previous post was at 04:48 PM CDT ----------

Hey, no one even mentioned the pulsating fedora bubble, or do you all have plymouth turned off? It's worth turning it back on just to see it once. :)

[nimnull22]

Examining todays "rawhide report" from mail list I found it has update of grab2 for fedora14:
grub2-1.98-3.fc14
-----------------
* Sat Jul 17 2010 Dennis Gilmore <dennis at ausil.us> - 1:1.98-3
- correctly generate a grub.cfg on kernel update

I got rawhide from Fedora 13, and I do not have grub2 installed.
Also my boot partition is ext3.
If there is update for grub2, my be I need somehow to install it, otherwise when I will get it.
Of course the question is do I (or one) need grub2, but it looks like it will be on fedora 14 and grub won't update it self up to grub2.

---------- Post added at 07:08 AM CDT ---------- Previous post was at 06:34 AM CDT ----------

Attention please !!!!!!!!!!!!!!.

I decided to do some experiments with upstart-0.6.5-7.fc14, and I update it with upstart-sysvinit without systemd-4-3.fc14

So now I can't even log into linux from Level 3.
I see login prompt, but when I type user login/password or root login/password, system just asks for it again. Looks like upstart-0.6.5 broke something.

Any suggestions, please.

I will open a BUG against it.

[GoinEasy9]

No update for grub2, if it's on the rawhide report, it's usually because something went wrong when they were trying to build it. I don't see it in the 7/25 changes. Grub2 is available in the repos, so, if you want it, you can install it, but, they're not forcing it on us yet.
@nimnull22 When you experiment with the init's, you get to keep both pieces. Have you tried Level 1? If that doesn't work you might have to use the DVD for rescue. Did you try "init=/bin/upstart" on the kernel line? I really don't know what upstart-sysvinit does, it says it's a non-temporary fix if systemd doesn't work. I don't know how ominous non-temporary is. Good luck.

---------- Post added at 11:30 PM CDT ---------- Previous post was at 11:29 PM CDT ----------

BTW - except for the systemd/upstart problem, tonights updates just included a lot of KDE apps, since I use k3b and Konversation I get the bloat that comes with them.

[nimnull22]

Temporary solution:
one needs to add "enforcing=0" to the kernel boot options

[nimnull22]

Today is 27.07.
And today in accordance with https://fedoraproject.org/wiki/Releases/14/Schedule should be made new branch "Fedora 14". Where is it?

Second, which repository we need to use to update Fedora 14? I think rawhide is not the option.
I checked recently and fedora, rawhide, updates and updates-testing has the same list (from yesterday, because there is no new updates yet)

Thanks.

[GoinEasy9]

Rawhide will always be rawhide, and if the date is correct for branching today, it will probably be much later today.

[kurtdriver]

Rawhide is now a rolling release, to downgrade to f14 you'll have to install the fedora-release appropriate for it, although the fedora-release for rawhide has 14 in it's name, which is a little confusing.

[GoinEasy9]

Well, the release has 14 in it's name because it hasn't been forked yet. Are there repos with 14 as $relver yet? I'll have to look later.

[AdamW]

This is rumour control, here are the facts :)

The branch was delayed to try and give us some time to fix up the Python 2.7 migration which landed just prior.

The branch is now happening. At the same time, we're migrating from CVS to git for package maintenance, which obviously makes it a much bigger and scarier change which is going to take a day or so.

Today's Rawhide has landed Python 2.7 and Boost 1.44, which causes breakage in a few packages which have not yet been rebuilt against these versions. We'll be working through to fix those up once the branch is done and repos re-opened for submissions.

On the systemd/upstart thing, Lennart just muffed up the deps. He's at GUADEC at present which is why he hasn't fixed it yet, but he is aware of the issue.

[GoinEasy9]

Thanks AdamW, it's always nice to know what's ACTUALLY going on. Trying to find out by reading the devel mailing list or trying to search through bugzilla isn't always helpful.

[nimnull22]

Thanks for the information.
When I saw today update with Python 2.7 I thought that now, we have to wait for sure.

With new selinux policy at least upstart works, so when Lennart will make systemd, one of them will be ready to use.

[dd_wizard]

Just for grins, I ran this command:
[gene@Mobile-PC ~]$ sudo yum --nogpgcheck --exclude=systemd* --exclude=upstart* update 2>&1 | grep '^Error' | wc -l
31

I wonder how long it will take for the error count to reach zero. ;)

dd_wizard

[GoinEasy9]

Big update, but I had to use --skip-broken, many python apps that still haven't got through the 2.7 rebuild. Along with that, the infamous systemd came in, they fixed the dependency, but it locks up the boot. As I said in the other thread, you can use "init=/sbin/upstart" on the kernel line to get back into your machine.

[kurtdriver]

As I said in the other thread, you can use "init=/sbin/upstart" on the kernel line to get back into your machine.

My laptop wouldn't boot all the way today.
the last little bit of the "F" was real slow and then it just stopped at a blue screen. Is this what you're referring too here?
Anyway, I nuked it this afternoon. I did yum history undue 155 in the belief that it would just undue the last update. Being in init 1 at the time and therefore disconnected from the internet, it erased programs but didn't download the new ones. I guess you can't believe everything you read on the internet.:(

Backing up as we speak.

[GoinEasy9]

@kurtdriver Yes, same symptoms here. Last lines during boot read failed to load default.service, then failed to start default.service. Wondering if the install of systemd didn't initialize everything correctly. BTW - no need to downgrade, just use init=/sbin/upstart on the kernel line till it's fixed. Oh, just reread your post, sorry, I almost nuked mine before I used the init line also.

---------- Post added at 11:55 PM CDT ---------- Previous post was at 10:13 PM CDT ----------

There is a fix to allow boot to complete with systemd. (WARNING: Boot completes but there are more problems)

rpm -e --nodeps systemd-units
yum install systemd-units

Which created the correct symlinks...maybe. One core of my AMD Phenom II X4 955 was at 99% even though I tried cold booting a few times. Shutdown/Restart locks up. And the one app I tried to open (Chrome) never opened.

I cold booted back with init=/sbin/upstart and system was usable again.

I added comment and am following https://bugzilla.redhat.com/show_bug.cgi?id=618315. I just hope I didn't bork my system applying the fix.

[nimnull22]

This is interesting enough.
First time when (systemd and systemd-sysvinit) appear in update I excluded them with "yum -x" and never saw it again.
Now "systemd-units" was skipped by "--skip-broken".
So now to get it all I need to download it from koji I guess. I will try when new version appears.
I was really bad idea to change them together.

[kurtdriver]

So I've reinstalled on the laptop after nuking it yesterday. Tonight, I installed the rawhide repos, including rpmfusion, and then did yum upgrade kernel yum rpm python kernel-headers. Rebooted and yum update --skip-broken produces


Skip-broken could not solve problems
Error: Package: notify-python-0.1.1-8.fc12.x86_64 (installed)
Requires: python(abi) = 2.6
Removing: python-2.6.4-25.fc13.x86_64 (@fedora/13)
Available: compat-python24-2.4.6-1.fc13.x86_64 (rpmfusion-free-rawhide)
Available: python-2.7-7.fc14.i686 (rawhide)
Available: python3-3.1.2-12.fc14.i686 (rawhide)
Error: Package: gnome-dvb-daemon-0.1.20-1.fc14.x86_64 (@rawhide)
Requires: python(abi) = 2.6
Removing: python-2.6.4-25.fc13.x86_64 (@fedora/13)
Available: compat-python24-2.4.6-1.fc13.x86_64 (rpmfusion-free-rawhide)
Available: python-2.7-7.fc14.i686 (rawhide)
Available: python3-3.1.2-12.fc14.i686 (rawhide)
Error: Package: wireshark-1.2.9-3.fc14.x86_64 (@rawhide)
Requires: python(abi) = 2.6
Removing: python-2.6.4-25.fc13.x86_64 (@fedora/13)
Available: compat-python24-2.4.6-1.fc13.x86_64 (rpmfusion-free-rawhide)
Available: python-2.7-7.fc14.i686 (rawhide)
Available: python3-3.1.2-12.fc14.i686 (rawhide)
Error: Package: totem-2.30.2-2.fc14.x86_64 (@rawhide)
Requires: libpython2.6.so.1.0()(64bit)
Removing: python-libs-2.6.4-25.fc13.x86_64 (@fedora/13)
Error: Package: audacious-plugins-freeworld-mp3-2.2-3.fc13.x86_64 (@rpmfusion-free/13)
Requires: libaudid3tag.so.2()(64bit)
Removing: audacious-libs-2.2-14.fc13.x86_64 (@fedora/13)
You could try running: rpm -Va --nofiles --nodigest



I can exclude those programs and still get these errors. Thing is, I've had them previously and even with google I can't figure out these dependencies. Is this something that will fix itself in a few days? Is there some way of excluding these packages? --exclude=totem doesn't do it.

[GoinEasy9]

The python 2.7 rebuild is in progress, I would assume that's the cause of missing dependencies and it will fix itself in the coming days. I'm surprised --skip-broken didn't work. When I get to my rawhide box later I'll see if I have the same problem.

[kurtdriver]

GE9, thanks for a quick reply, so you think this may be related to SJ's thread of a couple days ago? At least it booted today:)

[GoinEasy9]

My whole update tonight, which is quite a few packages, is all being skipped because of python dependencies. SJ's thread was not the only one that showed problems with python dependencies. BTW - I'm amazed that you were able to reinstall rawhide in the state that it's in, and the fact that it booted after systemd became the default init without using init=/sbin/upstart on the kernel line is curious, unless somehow that update wasn't included in your reinstall.
Last night, I had a few of those same dependency errors, but, I was able to work around them with --skip-broken. I guess I'll try again tomorrow. There are way to many things in flux in rawhide. Hah, and I think the branch for F14 and F15(rawhide) is also happening.
I'm going back to play with my stable soothing F13 tonight, I'll see how rawhide looks again tomorrow.

[nimnull22]

As I thought, I installed systemd from koji and surprisingly it works. You can type "systemd --test" on root console, redirect output to a file and see what it does.
So systemd works.
There is "systemd-gtk" also in koji, how to use this graphical frontend
Name : systemd-gtk
Arch : i686
Version : 4
Release : 4.fc14
Size : 213 k
Repo : installed
From repo : /systemd-gtk-4-4.fc14.i686
Summary : Graphical frontend for systemd
URL : http://www.freedesktop.org/wiki/Software/systemd
License : GPLv2+
Description : Graphical front-end for systemd.

---------- Post added at 01:20 AM CDT ---------- Previous post was at 01:13 AM CDT ----------


Ok, I did as suggested before:
rpm -e --nodeps systemd-units
yum install systemd-units

Then added "init=/bin/systemd" to grub and got:
[root@localhost ~]# dmesg | grep systemd
Kernel command line: ro root=/dev/mapper/VolGroup-lv_root edd=off init=/bin/systemd
systemd[1]: Succesfully opened /dev/kmsg for logging.
systemd[1]: systemd 4 running in system mode.
systemd[1]: Your kernel apparently lacks built-in autofs4 support. Please fix that. We'll now try to work around this by calling '/sbin/modprobe autofs4'...
systemd[1]: Your kernel apparently lacks built-in ipv6 support. Please fix that. We'll now try to work around this by calling '/sbin/modprobe ipv6'...
systemd[1]: Set hostname to <localhost.localdomain>.
systemd[1]: Failed to load configuration for distcache.service: No such file or directory
systemd[1]: Failed to load configuration for openct.service: No such file or directory
systemd[1]: systemd-initctl.socket failed to listen on sockets: No such file or directory
systemd[1]: Unit systemd-initctl.socket entered maintenance state.
systemd[1]: Successfully connected to system D-Bus bus 7eaa2d8874b0235100e1994400000018 as :1.6
systemd[1]: Succesfully opened syslog for logging.
systemd[1]: Job dev-mapper-VolGroup\x1dlv_root.device/start timed out.
systemd[1]: Job dev-disk-by\x1duuid-923c8122\x1ddb4d\x1d49a0\x1d9409\x1d7cab5993b221.d evice/start timed out.


Not bad, it works
"systemctl" works
tree -d /cgroup/systemd/ - works also.

Interesting, but by default I have "upstart".


---------- Post added at 01:28 AM CDT ---------- Previous post was at 01:20 AM CDT ----------

Also, there is no Fedora 14 branch yet.
You can check on http://mirrors.fedoraproject.org/publiclist/.

---------- Post added at 02:34 AM CDT ---------- Previous post was at 01:28 AM CDT ----------

Additionally.
When I use systemd I got in dmesg:
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev cgroup, type cgroup), uses genfs_contexts
SELinux: initialized (dev autofs, type autofs), uses genfs_contexts
SELinux: initialized (dev autofs, type autofs), uses genfs_contexts
SELinux: initialized (dev autofs, type autofs), uses genfs_contexts
SELinux: initialized (dev autofs, type autofs), uses genfs_contexts
SELinux: initialized (dev autofs, type autofs), uses genfs_contexts


This is not good, what do you think?

---------- Post added at 02:51 AM CDT ---------- Previous post was at 02:34 AM CDT ----------

Also under XFCE doesn't do appropriate action for "restart" and "shut down", both of them lead to "logout".

[GoinEasy9]

@nimnull22
Once you add systemd, there should be no need for init=/bin/systemd. It should be the default init.
Many reports, including mine, of unresponsive restart & shutdown.
I have "cpu gone wild" after booting up using systemd, seems I'm not alone.
I'm following https://bugzilla.redhat.com/show_bug.cgi?id=618315 and devel mailing list to watch for more info.

[nimnull22]

We have new branch for Fedora 14.
http://mirrors.fedoraproject.org/publiclist/

So I suppose we also need a new thread in this forum about Fedora 14.

---------- Post added 2010-07-31 at 05:14 AM CDT ---------- Previous post was 2010-07-30 at 10:58 PM CDT ----------

Stupid question, how can I change repository to update from Fedora 14 branch?

[GoinEasy9]

There's now a fedora-release-14 AND a fedora-release-rawhide, so, pick your poison. I'm staying with rawhide ... how much worse can it get? LOL

---------- Post added at 10:41 AM CDT ---------- Previous post was at 10:35 AM CDT ----------

And here's your repo link....mirrors are working.
http://mirror.us.as6453.net/fedora/linux//development/

[nimnull22]

Ok, thanks, I did it like you suggested.
Now I download updates from Fedora and updates-testing.
I checked many time - buy default I got "upstart" somehow, and I am happy with it. Because systemd really need to be repaired.

But, I am going to left Rawhide by now. I want to get 14 first, and then I start to play with new rawhide.

Thanks.

[kyryder]

Today I got
rpm -qa|grep fedora-release*
fedora-release-15-0.1.noarch
but still have many Packages skipped because of dependency problems and still using "init=/sbin/upstart"

Ky

[dd_wizard]

@GoinEasy9:

Memory tells me you have the same issue with the gnome-volume-control-applet not showing an icon in the notification area. The solution, from https://bugzilla.redhat.com/show_bug.cgi?id=616543, is to install gnome-icon-theme-symbolic. I can't seem to reply to the F14 thread, so I'll reply to your question about who's staying with rawhide here. Once most of my open bugs against F14 are resolved, I'll probably move back to rawhide. Of course, that means I can't find any more F14 bugs! :p

dd_wizard

---------- Post added at 12:46 PM CDT ---------- Previous post was at 12:15 PM CDT ----------

Oh, I was having issues with icons in avant-window-navigator, too. Installing gnome-icon-theme-legacy fixed that problem.

dd_wizard

[GoinEasy9]

When I get home tonight I'll try to install both. Rawhide report doesn't look any shorter than it did when it was last run, but, I'm still hoping the python rebuild concludes so we can start updating again.

[dd_wizard]

When you do, check out koji/kernel... 2.6.35 went gold! :D
[gene@Mobile-PC ~]$ uname -sr
Linux 2.6.35-1.fc14.x86_64

Funny though, it's not available for F15, yet. :p

dd_wizard

[GoinEasy9]

Well, boost-system came in tonight, which allowed some updates, but the python problems persist, so there are many packages just sitting broken. Tomorrow's another day.
BTW - gnome-icon-theme-symbolic restores the icon, but the functionality of gnome volume control from that applet is still the same. Not that I thought an icon would fix it. The fix might be in that long list of python packages, we'll see.

[tox]

Well, boost-system came in tonight, which allowed some updates, but the python problems persist, so there are many packages just sitting broken. Tomorrow's another day.
BTW - gnome-icon-theme-symbolic restores the icon, but the functionality of gnome volume control from that applet is still the same. Not that I thought an icon would fix it. The fix might be in that long list of python packages, we'll see.

https://fedoraproject.org/wiki/Features/Python_2.7 22 Packages remaining to be rebuilt

[GoinEasy9]

Thanks dexta, only 22 left is a good sign.

[dd_wizard]

What functionality are you missing from gnome-volume-control, GoinEasy9? I restored mine by deleting gnome-volume-control.desktop from wherever it was in my home directory.

dd_wizard

[GoinEasy9]

Gnome-volume-control if accesses from control center or menu has all functionality. Trying to click the icon, left click brings a slider that triggers abrt if clicked, right click allows a checkmark for mute, or preferences, and preferences bring up gnome volume control. I couldn't find any .desktop files in /home, I'll look again. The desktop files I found were buried in the /usr folder.

[dd_wizard]

OUCH! I just discovered the left-click bug you're describing with headphones on and a rough blues song playing! :( Looks like we're in the same state.

dd_wizard

[kurtdriver]

My desktop machine has the disease mentioned above. Adding init=/sbin/upstart does the same as not adding it. It dies at the line that says

init[1]: Failed to load configuration for default.target: No such file or directory
init[1]: Failed to start default target: Unit default.target failed to load.


This leaves me unable to reinstall systemd. Is there something I can do with a livecd?