[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#103823: bug #103823 on boot-floppies



tags 103823 + potato
thanks

On Fri, Jan 11, 2002 at 05:37:28PM +0100, GAUFILLET Pierre wrote:
> Sorry to send you directly this mail, but I do not know clearly how the
> debian bug mailing list is working...

I will forward this mail to the bug. To send mail to the bug, which
also appears on debian-boot for boot-floppies bugs, just send to
nnnnn@bugs.debian.org which in this case is 103823@bugs.debian.org.

Unfortunately I'm not familiar enough with this code to be able to
help. However, I see you are using the potato boot-floppies. You may
have better luck by downloading the current woody boot-floppies.

http://http.us.debian.org/debian/dists/woody/main/

>  
> So, I met the same problem on my configuration (BP6 SMP, kernel 2.2.19, NIC
> 8139C, the 8139too driver is integrated into the kernel, boot-floppies
> 2.2.23). During the network configuration, dbootstrap hangs.
> After a few tests, I found that, as I saw it in the previous mails,
> dbootstrap freezes was during network activation (call to ifconfig ...), and
> gdb indicates that the faulty instruction was :
>  
>  while (fgets(line,MAXLINE,output) != NULL) in the function execlog in
> util.c, with output=popen("/sbin/ifconfig...","r"); just before
>  
> More precisely, the program stops in fgets, in a "read" system call on the
> pipe "output".
>  
> I do not understand the reason of this, but I tried some other things that
> can perhaps help you :
>  - I add a breakpoint on the fgets line and run dbootstrap until here. Then,
> with gdb I try "print feof(output)", awating -1 because ifconfig does not
> issue anything with this command line, but it returns 0 ! 
>  - I add fungetc(fgetc(output),output) in dbootstrap, just before the while
> and it freezes on the fgetc call this time.
>  - I comment out the log instructions (the whole while loop), and then
> dbootstrap runs correctly (excepted for the log of course).
>  
> Well, my idea now is that the problem is not in the 8139too driver, but
> rather in the pipe management of the kernel or the libc. Before I read your
> mails, I was thinking about something regarding SMP (bad protection on pipe
> accesses for example), but it seems the problem has been identified even
> without SMP, right ?
>  
> I understood you have not any 8139 NIC so you can not try anything about
> that. Well, if I can help or try something, tell me.
>  
> Regards,
>  
> Pierre
>  
> PS : If you want to answer me, please use pierre.gaufillet@magic.fr
> <mailto:pierre.gaufillet@magic.fr> 
>  
> 

-- 
*------v--------- Installing Debian GNU/Linux 3.0 --------v------*
|      <http://www.debian.org/releases/woody/installmanual>      |
|   debian-imac (potato): <http://debian-imac.sourceforge.net>   |
|            Chris Tillman        tillman@azstarnet.com          |
|                   May the Source be with you                   |
*----------------------------------------------------------------*



Reply to: