How is your rawhide?

[SlowJet]

Mine is ok.
Haven't updated any of the new Xorg stuuf because it kills the mouse and keyboard or hangs after login.

On koji, the new libcanberra package really ticks off several packages after statup. Lots of cpu usage. hangs on background screen with mouse pointer.

Other than that it is running great with
newest kernel 2.6.31 24
evolution
openoffice 3.1.1-13.3
a new libselinux and selinuxy policy
several memory leak fixes.
new openssl
new nss, poppler, upstart
yum 3.2.23-8

So it only a matter of what is actually broke, not the end of the world.

SJ

[scottro]

Well, both myself and one other person found the latest updates killed X. He fixed his, he has an NVidia (a NVidia?) and updating drivers fixed it for him--mine's an onboard Intel in an Asus EEE.

Aside from that, nothing horrible--there's a coredump on boot, but it continues to boot.

I haven't had time to investigate any of it though.

[youknowwho]

Well, latest update killed X on me, so, yeah, I can't say is going fine with me, I have an Intel onboard card, so no clue other than to ask how to update to rawhide without updating the X.org packages?

Thanks in advance!

[scottro]

Looks like tonight's update fixes it, at least for me. :)

[SlowJet]

Last night's updates worked but I have a task called rpc.ldmapsd hogging cpu.
I can't releate that to any package of late.
Fx would not start.

After rawhide today 24th, in level 3 with yum, then a reboot, I get a spinning mouse on the backgroud after logon.
Tasks rpc.ldmapd, canberra, Xorg, and 2 or 3 others are in a cpu fight.

SJ

[sonoran]

How is my Rawhide? Funny you should ask. Last night's updates rendered it unworkable. Lots of amusing messages like "Fixing recursive fault but reboot is needed..." and "rsyslogd used greatest stack depth 2984 bytes left" and the cpu was running at 99%.

Also Gnome problems - "Cannot find session."

Finally after 3 or 4 reboots Gnome started and system-monitor revealed rpc.idmapd was using 95% cpu. I started yumex and found a single x86_64 update: rpcbind.x86_64 0.1.7-3.fc12!

Installed that, but problems persist. X only runs for 30 seconds or so before it segfaults.

I think I will wait until there is an Alpha rawhide to reinstall from - installing F11 and then updating to (possibly) unusable F12 again would require more time than I have. I should have cloned my working F12 while I had the chance, but it's too late now.

As someone said on the Test List: "Rawhide eats babies." Waaaaaaaa:(

[SlowJet]

It will be fixed in a few days.

SJ

P. S. The is not going to be an Alpha for F12. Beta is duw in Aug.

[SlowJet]

Rawhide is workking again.
update libcanberra libcanberra-gtk version 12 --oldpackage - that allow gnome to startup
kill -s 9 pid# rpc-ldmapd - free up the cpu
D/L newest xulrunner from koji on a working machine and sftp to the rawhide - that allowed Fx to start up.

OK, bring on dracut. :)

SJ

[sonoran]

Thanks for the news, sj. I'll have a go tonight.
Yeah, why does that dracut sound so ominous? They should have given it a name like Fuzzy, or Kitten.

[SlowJet]

Thanks for the news, sj. I'll have a go tonight.
Yeah, why does that dracut sound so ominous? They should have given it a name like Fuzzy, or Kitten.

I figured out what rpc.idmapd was about. (I was reading it ldmapd)
It is part of nfs-utils. This is for nfs4.
So just --oldpackage to the previous nfs-utils f12 rpm
rpm -Uvh nfs-utils....f12.i586.rpm --oldpackage --test

A reboot is required because SELinux-policy deneys rpc-statd to run after the fact.

So everything is normal again.

SJ

[leigh123linux]

Rawhide works great :) apart from the 64bit repo fiasco :(


[leigh@localhost ~]$ rpm -q xorg-x11-server-Xorg
xorg-x11-server-Xorg-1.6.99-3.20090618.fc12.x86_64
[leigh@localhost ~]$ rpm -qa |grep nvidia
nvidia-x11-drv-32bit-185.18.14-2.fc12.x86_64
nvidia-x11-drv-185.18.14-2.fc12.x86_64
[leigh@localhost ~]$ uname -r
2.6.31-0.24.rc0.git18.fc12.x86_64
[leigh@localhost ~]$ glxinfo
name of display: :0.0
display: :0 screen: 0
direct rendering: Yes
server glx vendor string: NVIDIA Corporation
server glx version string: 1.4
server glx extensions:
GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_SGIX_fbconfig,
GLX_SGIX_pbuffer, GLX_SGI_video_sync, GLX_SGI_swap_control,
GLX_EXT_texture_from_pixmap, GLX_ARB_create_context, GLX_ARB_multisample,
GLX_NV_float_buffer, GLX_ARB_fbconfig_float
client glx vendor string: NVIDIA Corporation
client glx version string: 1.4
client glx extensions:
GLX_ARB_get_proc_address, GLX_ARB_multisample, GLX_EXT_visual_info,
GLX_EXT_visual_rating, GLX_EXT_import_context, GLX_SGI_video_sync,
GLX_NV_swap_group, GLX_NV_video_out, GLX_SGIX_fbconfig, GLX_SGIX_pbuffer,
GLX_SGI_swap_control, GLX_ARB_create_context, GLX_NV_float_buffer,
GLX_ARB_fbconfig_float, GLX_EXT_fbconfig_packed_float,
GLX_EXT_texture_from_pixmap, GLX_EXT_framebuffer_sRGB,
GLX_NV_present_video, GLX_NV_multisample_coverage
GLX version: 1.3
GLX extensions:
GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_SGIX_fbconfig,
GLX_SGIX_pbuffer, GLX_SGI_video_sync, GLX_SGI_swap_control,
GLX_EXT_texture_from_pixmap, GLX_ARB_create_context, GLX_ARB_multisample,
GLX_NV_float_buffer, GLX_ARB_fbconfig_float, GLX_ARB_get_proc_address
OpenGL vendor string: NVIDIA Corporation
OpenGL renderer string: GeForce 7600 GT/PCI/SSE2
OpenGL version string: 2.1.2 NVIDIA 185.18.14

[scottro]

I'm only running it on an x86 machine right now, so have missed Leigh's luck.

[leigh123linux]

I'm only running it on an x86 machine right now, so have missed Leigh's luck.

It looks like they are fixing it now :cool:


http://ftp.tudelft.nl/download.fedora.redhat.com/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
http://ftp.crc.dk/fedora/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
http://mirror.karneval.cz/pub/linux/fedora/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
http://ftp.fi.muni.cz/pub/linux/fedora/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
http://ftp.linux.cz/pub/linux/fedora/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
http://ftp.df.lth.se/pub/fedora/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
http://mirrors.nl.eu.kernel.org/fedora/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
http://mirrors.se.eu.kernel.org/fedora/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
http://fedora.nano-box.net/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 14] HTTP Error 404: Not Found
Trying other mirror.
ftp://alviss.et.tudelft.nl/pub/fedora/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 4] IOError: <urlopen error ftp error: [Errno ftp error] 550 Failed to change directory.>
Trying other mirror.
ftp://ftp6.linux.cz/pub/linux/fedora/linux/development/x86_64/os/repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2: [Errno 4] IOError: <urlopen error [Errno -2] Name or service not known>
Trying other mirror.
Error: failure: repodata/5a7fbc9ddacca1ba6160ed3cd8cc286d4c80629ae8e67914e1 8e3dbf1055953f-primary.sqlite.bz2 from rawhide: [Errno 256] No more mirrors to try.


At least it isn't offering to install i586 packages any more :)

[SlowJet]

Well, I updated to the next kernel, mkintrd, grubby, libselinux, selinux-policy.
And now Fx crashes. lol

Time to eat :)

SJ

[typerlc]

Well, I updated to the next kernel, mkintrd, grubby, libselinux, selinux-policy.
And now Fx crashes. lol

Time to eat :)

SJ

You could try replacing your xulrunner again, or do what I did and update to 3.5rc3 from the mozilla website. NB: make sure you disable the adobe flash plugin ... it causes the browser to die. :(

[Jake]

To answer question in title..

The reply from rawhide was:

"Yeah I'm doing fine mate, how are you? I just been updating and updating!"

:-)

My Rawhide seems to be fine.

[SlowJet]

You could try replacing your xulrunner again, or do what I did and update to 3.5rc3 from the mozilla website. NB: make sure you disable the adobe flash plugin ... it causes the browser to die. :(

I don't think it is crashing because of rawhide, as it only crashes when I try to use this site!?:)
Other sites work ok so far.
Probably a definition spec changed from all the on/off glitz?


Otherwise rawhide is fine, save libcanberra, nfs-utils, just slow due to the debugging kernel.


SJ

[typerlc]

Otherwise rawhide is fine, save libcanberra, nfs-utils, just slow due to the debugging kernel.

I'm glad it's working for you. It's unstable for me. For unknown reasons (maybe too much load on the system), X will just suddenly die. But then it will go for a long period without problems. Yesterday it crashed 5 times in an hour. Now its been running for 12+ hours without a problem.

Keyboard drivers for X are still dodgy for me. After a reboot, certain keys (e.g. 'a', backspace, control, etc) cause the PC speaker to beep. And for no reason some of the lights on the keyboard keep toggling. And then that problem disappears.

But, right now, the system is usable. Anyone relying on their box would be well to avoid/be cautious about upgrading rawhide. Not that I listen to my own advice ;)

[youknowwho]

I'm glad it's working for you. Me too. For me it's been quite unstable since Sunday, anyhow, wish me the best, I'll be updating in a few instants, there's a new intel drivers, so, I'll trying that out and tell you about it!

[scottro]

@typeric, that one made me laugh. Been there, done that, got the tshirt.

Now Adam really impresses me, he runs it on his actual work box.

[typerlc]

I'll be updating in a few instants, there's a new intel drivers, so, I'll trying that out and tell you about it!

If you can login to your box to post a new message ;)

Now Adam really impresses me, he runs it on his actual work box.
I really shouldn't be proud of the fact, but so do I. It's caught me out a few times this last week. But, hey, I've got no one to blame but myself.

[youknowwho]

Now Adam really impresses me, he runs it on his actual work box.

I really shouldn't be proud of the fact, but so do I. It's caught me out a few times this last week. But, hey, I've got no one to blame but myself.

Well, I thought I was the only one...Yay, I'm not the only that's careless and irresponsible!

If you can login to your box to post a new message

You know what? I got the Live USB, so no problem, it only takes me 10 minutes to have a working system, but then again, I update to rawhide once again and...well...the circle keeps on going so...you get what I mean, although rawhide worked pretty for me since Fedora 11 got released until this past Sunday...

[NoEffex]

My rawhide is sexy as ever. Running my own custom compiled kernel(uses optimizations for various hardware and the like), and a really gnarley theme. I haven't had any real problems, because I'm a nazi over what updates and what doesn't.

[youknowwho]

If you can login to your box to post a new message

Just reboot, and at first it wouldn't boot up, I thought, it has to be the kernel, and indeed it was, just use the last kernel before the update and everything just works now! At last!

[scottro]

My only major issue was for those two or so upgrades when X was broken. Even that wasn't a show stopper as I don't use X all that much, though if I'd been using it on my real workstation, it would have been a showstopper.

In my case, I hoped it was the kernel, but it wasn't.

The other, really minor issue that I have is that on boot, or if running dhclient on a card, it will spew out a lot of debugging code. I'm 95 percent sure this is simply due to the Rawhid debugging code, because I remember when I had the same issue with F11's rawhide, one of the developers on the testing list gave me an rpm for a kernel without debugging and that fixed that issue.

It's minor though, it happens during boot, then it continues to boot, it happens when I run dhclient and then the card, whether wired or wireless gets an address.

[youknowwho]

My only major issue was for those two or so upgrades when X was broken. Even that wasn't a show stopper as I don't use X all that much, though if I'd been using it on my real workstation, it would have been a showstopper.

Yeah, for me since Sunday, it wouldn't boot up to the gui, of at least it would boot up to the log in screen, but then it would just kept on logging in or just wouldn't log me in at all, so my question, what do I do if it ever happens to me again?, how can I update without gui?, if I remember correctly even changing vt's wouldn't work for me, didn't even display them

[typerlc]

I had the same problem. Once X attempted to start, it failed, and then couldn't shutdown X, or switch to vts. But actually, the box is still running. So you can ssh into it.

Alternatively, I would recommend booting up into single-user mode, and changing the default runlevel in /etc/inittab to 3. Next time you boot, X won't automatically start. Then you can test X, and if happy it's working, do 'init 5'

[scottro]

I always immediately change my default runlevel to three. (I do this on working systems too, it's what I prefer.)

So, that would be my recommendation. Boot into runlevel 3 all the time, and use startx, rather than booting into the GUI. Among other things, this way, if the GUI breaks, you're not stuck. and can continue to update until the next update fixes it. :)

[typerlc]

Because of singie user mode, you're never really stuck. And even without this, you could ues a live-cd to boot and edit your inittab. Whatever you personally prefer though.

[Demz]

my question is hows the new PulseAudio in rawhide going? 0.9.16

[youknowwho]

Alternatively, I would recommend booting up into single-user mode, and changing the default runlevel in /etc/inittab to 3. Next time you boot, X won't automatically start. Then you can test X, and if happy it's working, do 'init 5'

I always immediately change my default runlevel to three. (I do this on working systems too, it's what I prefer.)

So, that would be my recommendation. Boot into runlevel 3 all the time, and use startx, rather than booting into the GUI. Among other things, this way, if the GUI breaks, you're not stuck. and can continue to update until the next update fixes it.

Because of singie user mode, you're never really stuck. And even without this, you could ues a live-cd to boot and edit your inittab. Whatever you personally prefer though.

Thanks! Those are wise solutions! I'll do that from now on...and as far as editing my inittab to default runlevel 3 from the live-cd, I have a live-usb, so everytime that I encounter this problem again, I hope not, I'll just edit my inittab from the live-usb.
Thanks again for the wise advises.

And regarding...
my question is hows the new PulseAudio in rawhide going? 0.9.16
Well, when I attempt to play a video file, a notification comes out and says that the device Pulse Audio stopped working or something like that so it defaults to the HDA Intel Audio card, but then again, I'm using KDE, so it may just be a Phonon issue conflicting with Pulse Audio.

[SlowJet]

dhclient 21 , kernel 2.6.31 33, openssh.
Bootup locks at pci irq 10 - the network.

reboot with old kernel-2.6.31 32 - same thing
reboot with old kernel inti 3 - comes upt to logon
dhclient 20 --oldpackage
reboot to kernel 2.6.31.33 - hangs up at pci irq 10 again
reboot to kernel-2.6.31 32 comes up as normal
with messages /sys/kernel/debug/kmem? (which is always empty)
However, I am where I was when I started except for openssh.
So the kernel-2.6.31 32 is now detecting
131,351 new suspected memory leaks. (The only diff is openssh packages)
openssh packages were final of several changes.
So logic has stopped working, and koji has created a blackhole. :)


SJ

P.S after posting this, I start a gnome-terminal and did su -
entered the password and the whole system froose up.
So maybe it's gamma rays.

[zaitcev]

my question is hows the new PulseAudio in rawhide going? 0.9.16
It works ok here (on an HDA-compatible chiset, although by AMD). However, there is a couple issues:

#1 The biggest problem that Lennart's juggernaut continues to roll over everything and ~/.libao stopped working. So there is no way to bypass Pulse for most applications anymore. None that I know anyway.

#2 The new Pulse attempts to find the lowest latency playback parameters. Which is probably good for Skype and Ekiga users, but it causes significant skipping right after the stream is opened, that abates gradually as Pulse learns the situation. Unfortunately, VLC reopens audio stream every time it goes to the next track, so it's pretty annoying.

Add to it the expected stuff with the volume applet which we just cannot leave enough alone. It's not new in F12 though. Was the same for about 3 releases now: always changing, always crashing, always coming up muted. I see some bugs fixed: the single slider sometimes stops working, then gets fixed; double click ceased to work at some point, works currently. So, the usual.

-- Pete

[SlowJet]

xorg-x11-server-1.6.99-12-20090714 doesn't work (flashing _ prompt.)
#rpm -e libxf85config --nodeps
rpm -Uvh xorg-x11-xorg-server..99.9-20090706 --oldpackage

readline-6 breaks about 20 packages.

Otherwise most of the main package are installing and the system boots, as of now.

SJ

[Demz]

It works ok here (on an HDA-compatible chiset, although by AMD). However, there is a couple issues:

#1 The biggest problem that Lennart's juggernaut continues to roll over everything and ~/.libao stopped working. So there is no way to bypass Pulse for most applications anymore. None that I know anyway.

#2 The new Pulse attempts to find the lowest latency playback parameters. Which is probably good for Skype and Ekiga users, but it causes significant skipping right after the stream is opened, that abates gradually as Pulse learns the situation. Unfortunately, VLC reopens audio stream every time it goes to the next track, so it's pretty annoying.

Add to it the expected stuff with the volume applet which we just cannot leave enough alone. It's not new in F12 though. Was the same for about 3 releases now: always changing, always crashing, always coming up muted. I see some bugs fixed: the single slider sometimes stops working, then gets fixed; double click ceased to work at some point, works currently. So, the usual.

-- Pete

thanks for that, i'll be interested to see how it works on the x-fi cards when i get it installed

[SlowJet]

After some missing updates
I got selinux-policy 22 installed, the gnome-volume-co and notification stopped fighting over cpu,
using the 81 kernel-PAE still, the 86 kernel works but nouveau drv garbles the screen (First time I got the Plymouth bubble by default.)

samba-3.4.0 works for shared printing, and windows share (except cp from windows share gets an invalid argument and truncates the file.)
Have not tried to set up a printer again, it has NOT authorized popup at the very last step.)
SSH and sftp work for Gnome (to the other machine, Fedora 11.)

SJ

[SlowJet]

Interesting development - a package (hal, I think) has turned off ConsoleKit. polkit and ACLS.
All the auth is set to console yes.

This allowed the printer to be setup via root auth (test print worked.)

But when trying OOo 16.1, it opened up in document recovery (caused from a crash a few package versions earlier.)
The file in question was copied from the windows share with the first try of samba-client-3.4.0(meaning it was probably truncated) so the file does not exist or is garbage.
OOo froze up the computer, no mouse or keyboard available.
I don't see a way around this.

SJ

[SlowJet]

OOo 16.1 hangs the system. nouveau drv bad for kerenl 86 samba 3.4.0 cp from W/Share bad.
So that is a wrap for pre-Alpha rawhide until next week.

Linux 2.6.31-0.81.rc3.git4.fc12.i686.PAE #1 SMP Mon Jul 20 19:34:01 EDT 2009 i686 i686 i386 GNU/Linux
xorg-x11-server-Xorg-1.6.99-14.20090715.fc12.i586
xorg-x11-drv-nouveau-0.0.14-3.20090717gitb1b2330.fc12.i586
openssh-5.2p1-15.fc12.i686
hpijs-3.9.2-8.fc12.i686
setuptool-1.19.5-2.fc11.i586
setroubleshoot-server-2.2.16-1.fc12.i686
selinux-policy-3.6.22-2.fc12.noarch
hplip-libs-3.9.2-8.fc12.i686
cpp-4.4.1-2.i686
mesa-libGL-7.6-0.4.fc12.i686
dbus-1.2.16-3.fc12.i686
setools-console-3.3.6-1.fc12.i686
openoffice.org-core-3.1.1-16.1.fc12.i686
setup-2.8.7-1.fc12.noarch
mesa-libGLU-7.6-0.4.fc12.i686
dhclient-4.1.0-25.fc12.i686
setroubleshoot-2.2.16-1.fc12.i686
curl-7.19.5-8.fc12.i686
udev-145-2.fc12.i686
setroubleshoot-plugins-2.1.11-1.fc12.noarch
dhcpv6-client-1.2.0-2.fc12.i586
ibus-1.2.0.20090722-1.fc12.i686
hplip-3.9.2-8.fc12.i686
parted-1.9.0-6.fc12.i686
firefox-3.5.1-3.fc12.i686
samba-winbind-3.4.0-0.40.fc12.i586
samba-client-3.4.0-0.40.fc12.i586
samba-common-3.4.0-0.40.fc12.i586
setools-libs-3.3.6-1.fc12.i686


SJ

[scottro]

Hope you realize that even though it seems you might be talking to yourself, lots of people, well me at least, are watching and knowing when to not upgrade.

This is, IMHO, helpful to all, and those of us using rawhide, but not having the time (or courage) to constantly update, are finding this valuable. So, in case no one else says it, Thank you.

[Demz]

people with heaps of time on there hands use a pre-alpha rawhide , you should go visit the Strippers SJ or u'll become a hermit :)

[SlowJet]

Hope you realize that even though it seems you might be talking to yourself, lots of people, well me at least, are watching and knowing when to not upgrade.

This is, IMHO, helpful to all, and those of us using rawhide, but not having the time (or courage) to constantly update, are finding this valuable. So, in case no one else says it, Thank you.

Your are welcome. And thanks for the comment.

Here is today's news.
After today's updates things still looked bad.
But all that was needed was kernel 100 (lost track of koji)

It boots with out "nomodeset" parm
OO0 16.1 works - the popup screen was trying to tell me that OOo was already running, but had no text.
Today I read the text and deleted the lock file in "dotOOo/3".
After that the usual doc recovery came up, I canceled that and OOo cmae up.
Windows shared printing works.
Using xorg-x11-drv-nouveau 14.3
Fx-3.5.1-3 works

Sound is as usual for this machine - N/A

SJ

[phoenixpb]

how is my rawhide ?
hummmmmm fine :p

display is oki :) compiz works very well
sound is oki, i can listen my mp3s and my dvds

i have just one problem bluetooth is broken but it's not very important :p i can access my phone throw usb

any news about FC13 ??

thanks guys

............. yum update junky ...........................

[Demz]

how is my rawhide ?
hummmmmm fine :p

display is oki :) compiz works very well
sound is oki, i can listen my mp3s and my dvds

i have just one problem bluetooth is broken but it's not very important :p i can access my phone throw usb

any news about FC13 ??

thanks guys

............. yum update junky ...........................
yeah, Gnome3 is ugly http://forums.fedoraforum.org/showthread.php?t=227585

[phoenixpb]

yeah, Gnome3 is ugly http://forums.fedoraforum.org/showthread.php?t=227585

any link to try this gnome 3 ? :p

[Demz]

yeah click the link i gave you http://forums.fedoraforum.org/showpost.php?p=1249497&postcount=2

http://www.youtube.com/watch?v=9_B7VpIwdiY

http://www.youtube.com/watch?v=VyoD2v7QGCE

[phoenixpb]

humm i have

[root@localhost Download]# /bin/bash gnome-shell-build-setup.sh
gnome-shell-build-setup.sh: line 98: gpk-install-package-name: command not found
Updating jhbuild ... done
Installing jhbuild...
Writing ~/.jhbuildrc ... done
PATH does not contain /root/bin, it is recommended that you add that.

any idea ?

thanks :)

[Ulat]

I updated F11 to rawhide (last time few days ago) and I still have mess in my system, here is a list:

- mouse is jumping everywhere
- sound blaster x-fi gamer still not working (I know, I know... kernel driver is not ready yet ;) )
- I have new kernel problem:
"pci 0000:00:00.0: BAR 3: address space collision on of device [0xe0000000-0xffffffff]
pci 0000:00:00.0: BAR 3: can't allocate resource"
- icons in panel are not always displayed correctly, especially in notice area (probably because of default driver for Radeon 48xx cards)
- I still have two (old) "ata(number): softreset failed (device not ready)" reports from kernel
- I have a lot of usage from pulseaudio when there is no sound card in system (blacklisted snd-cxfi module)

At this moment I have all new updates.

[typerlc]

I don't have all your problems, but I am having my own. The nouveau display driver is still a bit dodgy. In particular, now the hardware mouse cursor doesn't work. Changing to software cursor works fine and may help you problem Ulat (edit /etc/xorg.conf and look for SWcursor).

I hope everything is still compiled with lots of debugging, because things are still sluggish.

On the plus side, the updates yesterday moved my system from unusable with the nouveau driver to useable. Hopefully the updates today don't break it :)

I updated F11 to rawhide (last time few days ago) and I still have mess in my system, here is a list:

- mouse is jumping everywhere
- sound blaster x-fi gamer still not working (I know, I know... kernel driver is not ready yet ;) )
- I have new kernel problem:
"pci 0000:00:00.0: BAR 3: address space collision on of device [0xe0000000-0xffffffff]
pci 0000:00:00.0: BAR 3: can't allocate resource"
- icons in panel are not always displayed correctly, especially in notice area (probably because of default driver for Radeon 48xx cards)
- I still have two (old) "ata(number): softreset failed (device not ready)" reports from kernel
- I have a lot of usage from pulseaudio when there is no sound card in system (blacklisted snd-cxfi module)

At this moment I have all new updates.

[SlowJet]

http://koji.fedoraproject.org/koji/builds

If you already have rawhide, try rawhid as there no need to wait for some fixes.

SJ

[vallimar]

humm i have

[root@localhost Download]# /bin/bash gnome-shell-build-setup.sh
gnome-shell-build-setup.sh: line 98: gpk-install-package-name: command not found
Updating jhbuild ... done
Installing jhbuild...
Writing ~/.jhbuildrc ... done
PATH does not contain /root/bin, it is recommended that you add that.

any idea ?

thanks :)

Maybe too late to mention, but gnome-shell and such are in Koji now.
Download and install at least these packages:
gnome-shell, clutter, mutter, gjs, gir-repository, gobject-introspection

After you have it installed, open a terminal under your regular account and do:
> gnome-shell --replace

Play around, when you are done being underwhelmed, go back to that terminal and
issue a ctrl-c to cancel. Your regular gnome-panel and window manager should restart.

[Ulat]

I don't want to start new topic so I will post a new info here.

I tried again update from F11 to F12 via yum (rawhide) and it looks that some of previous issues are gone and I noticed some new too. Here is a list:

- sound blaster x-fi gamer still not working (is better but it seams that snd-cxfi driver is still far from perfect)
- now I don't need to force radeonhd driver in xorg.conf (radeon driver now works fine with my 4850 - no xorg.conf needed)
- kernel mode settings works (correct resolution etc)
- I still have two "ata(number): softreset failed (device not ready)" reports from kernel
- I still have kernel reports:
"pci 0000:00:00.0: BAR 3: address space collision on of device [0xe0000000-0xffffffff]"
"pci 0000:00:00.0: BAR 3: can't allocate resource"
- icons in panel now are displayed correctly
- PulseAudio now works fine with two sound drivers: snd-cxfi and snd_hwdep (integrated Realtek)
- icon of VM applet is not transparent (I have dark background and icon always have grey background)
- and i still have a problem with mouse but only with specific model - A4-TECH X-710 FS (with F11 mouse works fine): pointer is not moving or jumping everywhere and xinput test-xi2 reports strange values when moving:

EVENT type 17
device: 2
detail: 0
valuators:
0: 0.00 (1.00)
1: 0.00 (32767.00)
2: 7271.00 (11.00)

EVENT type 17
device: 2
detail: 0
valuators:
0: 0.00 (-1.00)
1: 0.00 (-1.00)
2: 7282.00 (11.00)

EVENT type 17
device: 2
detail: 0
valuators:
1: 0.00 (-1.00)
2: 11.00 (11.00)


That's all for now.

[Finalzone]

+ plymouth now boot with graphic on nvidia system (Geforce 7800 GS)
+ stable so far
+ packagekit appears to improve because it automatically detect missing drivers from printer
- no full 3D support from nouveau driver
- gnome-shell is useless with too many flicking going on screen

[RahulSundaram]

Hi,

The gnome shell issue is a lot dependent on the hardware. Would be helpful perhaps to file a bug report with the hardware details

[Finalzone]

Hi,

The gnome shell issue is a lot dependent on the hardware. Would be helpful perhaps to file a bug report with the hardware details

Nvidia Geforce 78OOGS in this case. I could not screencapture desktop (spec. on my signature). I will check the log to see if nouveau driver is the culprit before submitting bugs and make sure there is no duplicata.
Edit
Bug report (https://bugzilla.redhat.com/show_bug.cgi?id=525248) has been submitted against mesa. You can view the nice result running gnome-shell on rawhide 20090924. If that requires hardware acceleration, I am afraid nVidia users are screwed because nouveau driver does not support 3D. (T_T)