Mandriva Cooker

Adam Williamson wrote: Noticed it hadn't been raised yet today, so for anyone who didn't
already know, today is the 10th anniversary of the release of the very
first Mandrake, 5.1.

A bit of poking finds the official announcement (in French):

http://groups.google.com/group/fr.comp.os.linux/browse_thread/thread/fb737882216cd303/7a88da659a13278b?lnk=st&q=#7a88da659a13278b

Happy birthday!
--
adamw
Congratulations, keep improving one of the best Linux distributión

> -----Original Message-----
> From: Tomasz Paweł Gajc [mailto:phenomenal@...]
> Sent: 24. juli 2008 00:31
> To: cooker@...
> Subject: Re: [Cooker] Happy 10th birthday Mandriva
>
> Dnia 2008-07-23, o godz. 12:15:09
> Adam Williamson <awilliamson@...> napisał(a):
>
> > Noticed it hadn't been raised yet today, so for anyone who didn't
> > already know, today is the 10th anniversary of the release of the very
> > first Mandrake, 5.1.
> >
> > A bit of poking finds the official announcement (in French):
> >
> > http://groups.google.com/group/fr.comp.os.linux/browse_thread/thread/f
> > b737882216cd303/7a88da659a13278b?lnk=st&q=#7a88da659a13278b
> >
> > Happy birthday!
>
> Happy birthday to Mandriva, i hope that next 10 years would be more
> sucessfull ;)

Woah that means that I have been running this fantastic distribution
for almost 10 years now.
                HAPPY HAPPY HAPPY BIRTHDAY TO
  THE ENTIRE TEAM OF DEVELOPERS, CONTRIBUTERS, TESTERS, AND USERS

I got my first version 5.2 CD set during the early winter of 1998 and
I am pretty sure that I still have the CD's on my attic somewhere.
Before that I had been running Redhat, slackware and one or two other
distributions for some years, but Mandrake Linux caught my attention
right away. I was hooked and no matter what distribution I have tried
since then, I have stuck with Mandrake/mandriva on my main computer ever
since that winter (with Gnome as GUI since late 1999).

I am sure that Mandriva Linux will bring me many many more joyful hours
in front of my computer in the future.


Kim Schulz

On Wed, Jul 23, 2008 at 9:15 PM, Adam Williamson
<awilliamson@...> wrote:
> Happy birthday!

So strictly speaking, it's Mandriva Linux birthday and not Mandriva... ;-)

Dnia 2008-07-23, o godz. 12:15:09
Adam Williamson <awilliamson@...> napisał(a):

> Noticed it hadn't been raised yet today, so for anyone who didn't
> already know, today is the 10th anniversary of the release of the very
> first Mandrake, 5.1.
>
> A bit of poking finds the official announcement (in French):
>
> http://groups.google.com/group/fr.comp.os.linux/browse_thread/thread/fb737882216cd303/7a88da659a13278b?lnk=st&q=#7a88da659a13278b
>
> Happy birthday!

Happy birthday to Mandriva, i hope that next 10 years would be more
sucessfull ;)

--
Regards,
TPG - MandrivaLinux's contributor

www.mandriva.com
  _
 ( )
  X ascii ribbon campaign - against html e-mail
 / \ www.asciiribbon.org   - against proprietary attachments


signature.asc (204 bytes) Download Attachment

Dnia 2008-07-23, o godz. 21:04:35
Frederik Himpe <fhimpe@...> napisał(a):

> On wo, 2008-07-23 at 19:21 +0200, Tomasz Paweł Gajc wrote:
> > Dnia 2008-07-23, o godz. 18:52:34
> > Frederik Himpe <fhimpe@...> napisał(a):
> >
> > > When trying to rebuild things like Firefox or Xulrunner with bm
> > > on one of the build nodes, this fails with this message:
> > >
> > > configure: error: The header jni.h was not found.  Set $JAVA_HOME
> > > to your java sdk directory, use
> > > --with-java-bin-path={java-bin-dir}, or reconfigure with
> > > --disable-javaxpcom. error: Bad exit status
> > >
> > > Now these pacakages already buildrequire java-rpmbuild and hance
> > > openjdk-devel, but still they don't find jni.h. Should not the
> > > java-rpmbuild or openjdk set the $JAVA_HOME environment variable
> > > by default?
> > >
> >
> > They find jni.h with:
> >
> > --with-java-include-path=%{java_home}/include \
> > --with-java-bin-path=%{java_home}/bin \
>
> Right, I had not noticed this, and there was another problem I
> introduced myself which was causing the problem, sorry.
>
> Still, the question remains: if openjdk would set this variable, we
> would not even need these configure paramters. As lots of java
> applications actually use this variable, why not set it by default in
> openjdk?
> Most of java stuff uses %{java_home} and there is no need to provide
usbdirectories of it. It is upstream issue of e.g firefox/xulrunner.

--
Regards,
TPG - MandrivaLinux's contributor

www.mandriva.com
  _
 ( )
  X ascii ribbon campaign - against html e-mail
 / \ www.asciiribbon.org   - against proprietary attachments


signature.asc (204 bytes) Download Attachment

> That patch went into 2.6.24.6 so it's in the above mentioned testkernels...
> Do they work for you?

Oh cool and sorry for the noise. I have so far only tested some older
2.6.24 tmb kernel and I think the latest nmb kernel.

I will test your kernel on sunday once I get back to computer which has
that dvb card.

Mika

Mika Laitio skrev:
>> If you are running 2008.1 and want to help, please test the:
>>
>> kernel-2.6.24.7-1.uc2mnb1
>
> Hi
>
> Would it be possible to get also the fix from
> https://qa.mandriva.com/show_bug.cgi?id=41140
>
> to be included in next tmb and nmb 2.6.24 kernels for 2008.1?
> 2.6.25/2.6.26 kernels in cooker do not need patching for this bug.
>
That patch went into 2.6.24.6 so it's in the above mentioned testkernels...

Do they work for you?

--
Thomas

> If you are running 2008.1 and want to help, please test the:
>
> kernel-2.6.24.7-1.uc2mnb1

Hi

Would it be possible to get also the fix from
https://qa.mandriva.com/show_bug.cgi?id=41140

to be included in next tmb and nmb 2.6.24 kernels for 2008.1?
2.6.25/2.6.26 kernels in cooker do not need patching for this bug.

Mika

Felix Miata wrote:
> On 2008/07/22 07:15 (GMT+1000) Steve Morris apparently typed:
>
>  
>> Felix Miata wrote:
>>    
>
>  
>>> On 2008/07/21 21:56 (GMT+1000) Steve Morris apparently typed:
>>>      
>
>  
>>>> Felix, I issued the command mkinitrd --preload via82cxxx
>>>> initrd-2.6.24.4-desktop-1mnb.img 2.6.24.4-desktop-1mnb to recreate the
>>>> initrd and when I rebooted, the boot process failed complaining it
>>>> couldn't find /dev/root. Have I done something wrong?
>>>>        
>
>  
>>> I'd boot Knoppix and do lsmod to find out if via82cxxx is the right driver.
>>> If it is, I'd suspect either you're not loading the right initrd, or a
>>> mismatch between root= in Grub and fstab. Did you ever set a label on your /
>>> partition? One other possibility might be that the newer driver versions are
>>> objecting to your slave HD that has no master, and giving different results
>>> between Grub and kernel.
>>>      
>
>  
>> Felix, I issued lsmod on my running system and received to following
>> portion but I am not sure I understand what it is telling me.
>>    
>
>  
>> ide_generic             1824  0 [permanent]
>> via82cxxx               8068  0 [permanent]
>> it821x                  6724  0 [permanent]
>> pata_via                9732  4
>> libata                138000  1 pata_via
>> sd_mod                 26336  6
>> scsi_mod              129836  4 sr_mod,sg,libata,sd_mod
>> ide_disk               14176  3
>> ide_core               98700  5 ide_cd,ide_generic,via82cxxx,it821x,ide_disk
>> reiserfs              236324  1
>>    
>
>  
>> Is this telling me that the hard disks are using ide_disk?
>>    
>
> I really don't know exactly what it's telling you, but, based on what's
> showing as permanent, I believe it's telling you that the IDE subsystem is in
> use; while the libata subsystem is loaded, but not currently necessary to
> system operation.
>
> I don't currently have any VIA systems running any kernel or distro that
> prefers to use libata instead of IDE that I have configured to use IDE
> instead. I do have that for several Intel and NForce. A check of an Intel
> shows there is nothing generic loaded. Maybe generic is grabbing the / HD
> before via82cxxx can get it. I don't know.
>
>   I tried an initrd with a preload of via82cxxx and the machine failed to boot because libATA couldn't talk to the root device.
> Since almost no one but you and me are on this thread, maybe you should try
> irc://freenode/mandriva-cooker and see what you can get.
>
>  
I will try here if I don't get any response to my new thread around this
where using an initrd that is preloading ide_disk has highlighted that
libATA and PATA_VIA appear to be force loaded and abort the boot process
because they can't talk to the root device.

But in any event, thankyou for all your help.
>> I have not put a label on the root partition as tune2fs can't touch it
>>    
>
> It can if you boot a live CD and don't mount that / before running tune2fs on it.
>
>  
>> and I can't remember at the moment what the equivalent for reiserfs is.
>>    
>
> I have no idea, as I've only once ever created a reiserfs partition, and it's
> long gone.
>
>  
>> I am using uuid on all hard disk entries and the uuid in fstab and grub
>> both match.
>>    
>
> I don't know how anyone can tell without going cross-eyed trying.
>  
[samorris.vcf]
begin:vcard
fn:Steve Morris
n:Morris;Steve
email;internet:samorris@...
x-mozilla-html:TRUE
version:2.1
end:vcard

Andrey Borzenkov wrote:
> Steve Morris wrote:
>
>  
>> Felix, I issued the command mkinitrd --preload via82cxxx
>> initrd-2.6.24.4-desktop-1mnb.img 2.6.24.4-desktop-1mnb to recreate the
>> initrd and when I rebooted, the boot process failed complaining it
>> couldn't find /dev/root. Have I done something wrong?
>>
>>    
>
> You need to
>
>   Thanks Andrey,
> - load ide-disk driver
>   I have built an initrd that preloads ide_disk which causes the hard
disks on the motherboard to be mounted as /dev/hdx but it appears that
libaATA and PATA_VIA are being force loaded somewhere and abort the boot
process because they can't communicate with the hard disks. I will raise
another thread about why this is happening.
> - tell kernel to look for root on IDE disk (root=/dev/hda1 or whatever); or
> use LABELs/UUIDs as already discussed
> - finally you also need to adapt fstab otherwise rc.sysinit likely blows out
> early checking file systems
>
>  
I am already using UUID in both fstab and grub.

regards,
Steve


[samorris.vcf]
begin:vcard
fn:Steve Morris
n:Morris;Steve
email;internet:samorris@...
x-mozilla-html:TRUE
version:2.1
end:vcard

On Wed, Jul 23, 2008 at 10:36 PM, Jean-Joseph THIBAULT <jeanjo@...> wrote:
> Downgrading gvfs to 0.99.1 works around this.
> Ecxatly, thank you :-)
Upgrading to 0.99.2-3mdv will help too, it is in the queue.


--
AL I:40: Do what thou wilt shall be the whole of the Law.

Götz Waschk a écrit : Downgrading gvfs to 0.99.1 works around this. Ecxatly, thank you :-)

Noticed it hadn't been raised yet today, so for anyone who didn't
already know, today is the 10th anniversary of the release of the very
first Mandrake, 5.1.

A bit of poking finds the official announcement (in French):

http://groups.google.com/group/fr.comp.os.linux/browse_thread/thread/fb737882216cd303/7a88da659a13278b?lnk=st&q=#7a88da659a13278b

Happy birthday!
--
adamw

On wo, 2008-07-23 at 19:21 +0200, Tomasz Paweł Gajc wrote:
> Dnia 2008-07-23, o godz. 18:52:34
> Frederik Himpe <fhimpe@...> napisał(a):
>
> > When trying to rebuild things like Firefox or Xulrunner with bm on one
> > of the build nodes, this fails with this message:
> >
> > configure: error: The header jni.h was not found.  Set $JAVA_HOME to
> > your java sdk directory, use --with-java-bin-path={java-bin-dir}, or
> > reconfigure with --disable-javaxpcom.
> > error: Bad exit status
> >
> > Now these pacakages already buildrequire java-rpmbuild and hance
> > openjdk-devel, but still they don't find jni.h. Should not the
> > java-rpmbuild or openjdk set the $JAVA_HOME environment variable by
> > default?
> >
>
> They find jni.h with:
>
> --with-java-include-path=%{java_home}/include \
> --with-java-bin-path=%{java_home}/bin \
Right, I had not noticed this, and there was another problem I
introduced myself which was causing the problem, sorry.

Still, the question remains: if openjdk would set this variable, we
would not even need these configure paramters. As lots of java
applications actually use this variable, why not set it by default in
openjdk?

--
Frederik Himpe <fhimpe@...>

Le mercredi 23 juillet 2008 à 19:14 +0200, Götz Waschk a écrit :
> Downgrading gvfs to 0.99.1 works around this.

It is under investigation by gvfs people upstream
( http://bugzilla.gnome.org/show_bug.cgi?id=544399 )

ATM, only cooker is seeing this.

And it is not a dbus bug.

--
Frederic Crozat <fcrozat@...>
Mandriva

Dnia 2008-07-23, o godz. 18:52:34
Frederik Himpe <fhimpe@...> napisał(a):

> When trying to rebuild things like Firefox or Xulrunner with bm on one
> of the build nodes, this fails with this message:
>
> configure: error: The header jni.h was not found.  Set $JAVA_HOME to
> your java sdk directory, use --with-java-bin-path={java-bin-dir}, or
> reconfigure with --disable-javaxpcom.
> error: Bad exit status
>
> Now these pacakages already buildrequire java-rpmbuild and hance
> openjdk-devel, but still they don't find jni.h. Should not the
> java-rpmbuild or openjdk set the $JAVA_HOME environment variable by
> default?
> They find jni.h with:

        --with-java-include-path=%{java_home}/include \
        --with-java-bin-path=%{java_home}/bin \


--
Regards,
TPG - MandrivaLinux's contributor

www.mandriva.com
  _
 ( )
  X ascii ribbon campaign - against html e-mail
 / \ www.asciiribbon.org   - against proprietary attachments


signature.asc (204 bytes) Download Attachment

Le mercredi 23 juillet 2008 à 19:08 +0200, Götz Waschk a écrit :
> Hi,
>
> I can reproduce the problem. D-Bus is running for system and user.
> Just click on file and on save, select a directory, press OK and
> firefox will crash:
> # firefox
> process 13421: arguments to
> dbus_connection_send_with_reply_and_block() were incorrect, assertion
> "(error) == NULL || !dbus_error_is_set ((error))" failed in file
> dbus-connection.c line 3301.
> This is normally a bug in some application using the D-Bus library.
>   D-Bus not built with -rdynamic so unable to print a backtrace
> /usr/lib/firefox-2.0.0.16/run-mozilla.sh: line 131: 13421 Abgebrochen
>            /usr/bin/soundwrapper "$prog" ${1+"$@"}
Ok, I was able to get a crash..

I'll investigate (it is probably one of the patches I added in dbus
which is causing the problem).

--
Frederic Crozat <fcrozat@...>
Mandriva

Downgrading gvfs to 0.99.1 works around this.


--
AL I:40: Do what thou wilt shall be the whole of the Law.

Here's the backtrace:
process 15877: arguments to
dbus_connection_send_with_reply_and_block() were incorrect, assertion
"(error) == NULL || !dbus_error_is_set ((error))" failed in file
dbus-connection.c line 3301.
This is normally a bug in some application using the D-Bus library.
  D-Bus not built with -rdynamic so unable to print a backtrace

Program received signal SIGABRT, Aborted.
0xffffe424 in __kernel_vsyscall ()
Missing debug package(s), you should install: gamin-debug libbeagle-debug
(gdb) bt
#0  0xffffe424 in __kernel_vsyscall ()
#1  0xb74edd90 in raise (sig=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
#2  0xb74ef7f8 in abort () at abort.c:88
#3  0xb69ec3e5 in ?? () from /lib/libdbus-1.so.3
#4  0xb69e7b81 in ?? () from /lib/libdbus-1.so.3
#5  0xb69cf05e in dbus_connection_send_with_reply_and_block ()
   from /lib/libdbus-1.so.3
#6  0xb69c866e in ?? () from /lib/libdbus-1.so.3
#7  0xb69c874a in dbus_bus_remove_match () from /lib/libdbus-1.so.3
#8  0xaf3377ac in ?? ()
   from /usr/lib/gio/modules/libgioremote-volume-monitor.so
#9  0xb7834698 in IA__g_object_unref (_object=0xa385600) at gobject.c:2411
#10 0xb734deee in g_union_volume_monitor_finalize (object=0xa439380)
    at gunionvolumemonitor.c:73
#11 0xb7834698 in IA__g_object_unref (_object=0xa439380) at gobject.c:2411
#12 0xb79f9e6a in gtk_file_system_dispose (object=0xa941f80)
    at gtkfilesystem.c:184
#13 0xb7834600 in IA__g_object_unref (_object=0xa941f80) at gobject.c:2383
#14 0xb79f7852 in free_async_data (async_data=0xb8fc7c8) at gtkfilesystem.c:672
#15 0xb79f7d95 in query_info_callback (source_object=0xb153a90,
    result=0xab4b000, user_data=0xb8fc7c8) at gtkfilesystem.c:783
#16 0xb734b2eb in IA__g_simple_async_result_complete (simple=0xab4b000)
    at gsimpleasyncresult.c:554
#17 0xb734b31c in complete_in_idle_cb (data=0xab4b000)
    at gsimpleasyncresult.c:564
#18 0xb71e37d0 in g_idle_dispatch (source=0xb14ab88, callback=0,
    user_data=0xab4b000) at gmain.c:4173
#19 0xb71e579a in IA__g_main_context_dispatch (context=0xa0655a8)
    at gmain.c:2068
#20 0xb71e8eb8 in g_main_context_iterate (context=0xa0655a8, block=1,
    dispatch=1, self=0xa0425e0) at gmain.c:2701
#21 0xb71e93cb in IA__g_main_loop_run (loop=0xa5ec778) at gmain.c:2924
#22 0xb7a4bb4f in IA__gtk_main () at gtkmain.c:1172
#23 0xb5d89896 in nsAppShell::Run (this=0xa14e788) at nsAppShell.cpp:139
#24 0xb5c5178f in nsAppStartup::Run (this=0xa14e748) at nsAppStartup.cpp:151
#25 0x08051080 in XRE_main (argc=0, argv=0x805bfa4, aAppData=0x8061c00)
    at nsAppRunner.cpp:2883
#26 0x0804ac0f in main (argc=Cannot access memory at address 0x3e05
) at nsBrowserApp.cpp:61

--
AL I:40: Do what thou wilt shall be the whole of the Law.

Hi,

I can reproduce the problem. D-Bus is running for system and user.
Just click on file and on save, select a directory, press OK and
firefox will crash:
# firefox
process 13421: arguments to
dbus_connection_send_with_reply_and_block() were incorrect, assertion
"(error) == NULL || !dbus_error_is_set ((error))" failed in file
dbus-connection.c line 3301.
This is normally a bug in some application using the D-Bus library.
  D-Bus not built with -rdynamic so unable to print a backtrace
/usr/lib/firefox-2.0.0.16/run-mozilla.sh: line 131: 13421 Abgebrochen
           /usr/bin/soundwrapper "$prog" ${1+"$@"}


--
AL I:40: Do what thou wilt shall be the whole of the Law.