Bug 619461

Same for by-uuid links.

(In reply to comment #1)
> Same for by-uuid links.

Hi,

  Would you try my latest submit in Base:System ? I just fixed an ancient issue
in SuSE specific patch, which is related to this bug.

  Thanks.

>   Would you try my latest submit in Base:System ? I just fixed an ancient issue
> in SuSE specific patch, which is related to this bug.

which RPM(s) shall I update for testing ?

I'd like to stay as close to factory/final as possible, so I'd like to avoid getting all new stuff from Base:System...

(In reply to comment #2)
>   Would you try my latest submit in Base:System ? I just fixed an ancient issue
> in SuSE specific patch, which is related to this bug.

fyi, with:

rpm -qa \*udev\* | sort
 libgudev-1_0-0-157-44.1.x86_64
 libudev0-157-44.1.x86_64
 libudev0-32bit-157-44.1.x86_64
 udev-157-44.1.x86_64
 udev-configure-printer-1.2.0-1.11.x86_64

from,

grep base /etc/zypp/repos.d/BaseSystem.repo 
 baseurl=http://download.opensuse.org/repositories/Base:/System/openSUSE_Factory

now mounts,

 lvs | grep HOME
  LVD0_HOME VG0 -wi-ao  50.00g                                      
 grep HOME /etc/fstab
  LABEL=LVD0_HOME  /home  ext4 relatime,acl,user_xattr,journal_checksum,barrier=1  1 2

OK @ boot.

(In reply to comment #4)
> OK @ boot.

NAK -- this udev update does *not* work for me:

I did update those udev packages from Base:System -- from "rpm -qa --last" :

libgudev-1_0-devel-157-44.1                   Tue Jul  6 23:38:53 2010
libgudev-1_0-0-157-44.1                       Tue Jul  6 23:38:46 2010
libgudev-1_0-0-32bit-157-44.1                 Tue Jul  6 23:38:18 2010
libudev-devel-157-44.1                        Tue Jul  6 23:38:12 2010
libudev0-157-44.1                             Tue Jul  6 23:38:08 2010
libudev0-32bit-157-44.1                       Tue Jul  6 23:37:43 2010
udev-157-44.1                                 Tue Jul  6 23:36:33 2010
gpg-pubkey-5d053b4f-479655f4                  Tue Jul  6 23:33:00 2010

and did reboot -- still only real partitions in /dev/disk/by-label and .../by-uuid/  (but reboot into runlevel 5 was ok!).


ok, next guess: it might be some fix which depends on initrd ?!
so I run "mkinitrd" and did reboot again -- and now I'm in *big* trouble
with the nouveau drm/KMS module driver:  
4.5 secs after boot (with udev/mouveau msgs) the screen display first goes a bit nuts and then black and hangs for 1-5 minutes before startup continues :-(

so I've stepped back to the udev version from factory (again output from "rpm -qa --last"  which shows that no Base:System packages are left):


libgudev-1_0-devel-157-1.2                    Wed Jul  7 02:02:02 2010
libgudev-1_0-0-157-1.2                        Wed Jul  7 02:02:01 2010
libgudev-1_0-0-32bit-157-1.2                  Wed Jul  7 02:02:00 2010
libudev-devel-157-1.2                         Wed Jul  7 02:01:59 2010
libudev0-157-1.2                              Wed Jul  7 02:01:58 2010
libudev0-32bit-157-1.2                        Wed Jul  7 02:01:53 2010
udev-157-1.2                                  Wed Jul  7 02:01:52 2010
gpg-pubkey-5d053b4f-479655f4                  Tue Jul  6 23:33:00 2010

and run mkinitrd (multiple times;) but my nouveau/initrd problems remain :-(

for more details about the remaining problem of nouveau/kms/initrd I opened another bug:  #620384


do you have any idea if #620384 might be related to that udev update and where I should look for any leftover changes ?
or do you think this was "only" triggered by mkinitrd and udev versions do not matter ?


anyway, help and ideas now to test/fix #620384 are *very* welcome because this severly broke my 11.3 setup on my notebook and I think about stopping any further tests for 11.3 as this is *the* major usability problem for me right now :-(

Hi Harald,

  It's critical to update both udev and device-mapper.
  Could you update both and retest ?

  Hope this helps.

  Thanks.

(In reply to comment #6)
> Hi Harald,
> 
>   It's critical to update both udev and device-mapper.
>   Could you update both and retest ?
> 
>   Hope this helps.

ok, I'll try it later... 
is it neccessary tu run mkinitrd after update ?

(In reply to comment #7)
> (In reply to comment #6)
> > Hi Harald,
> > 
> >   It's critical to update both udev and device-mapper.
> >   Could you update both and retest ?
> > 
> >   Hope this helps.
> 
> ok, I'll try it later... 
> is it neccessary tu run mkinitrd after update ?

Yes, you have to run mkinitrd after update.

Thanks.

Have observed the same problem with openSUSE 11.3 RC2, when using MD raid with LVM and file systems mounted using labels -- for more details, refer to bug #586837 comments 16, 17 and 18.

(In reply to comment #8)
> Yes, you have to run mkinitrd after update.

ACK: with udev-159-45.1.x86_64 and device-mapper-1.02.49-38.1.x86_64 all my fs labels show up again!

but again a nouveau mks hang for 3 minutes with X11 blocked later on:-(


thanks!

(In reply to comment #10)
> (In reply to comment #8)
> > Yes, you have to run mkinitrd after update.
> 
> ACK: with udev-159-45.1.x86_64 and device-mapper-1.02.49-38.1.x86_64 all my fs
> labels show up again!
> 
> but again a nouveau mks hang for 3 minutes with X11 blocked later on:-(
> 
> 
> thanks!

Thanks for the feedback.

I'm using nouveau too, but X11's running fine.
Wondering if this worth another bug report ?