Re: Re[2]: Problem with running tests during installing coreutils-5.94

[Philip Rowlands]

On Tue, 18 Apr 2006, Adam Waltman wrote:

> So the tips how should I use gdb were really heplful.
> Here is the output
> k13 coreutils-5.94 # gdb -x <( printf "run\nbt" ) --args
> src/shred --remove --zero adam adam1
[snip]

> Program terminated with signal SIGSEGV, Segmentation fault.
> The program no longer exists.
> No stack.
> (gdb)

Oh. That's unexpected.

When a program segfaults gdb should step in and let you examine its 
internal state. I'm not the gdb expert; that's how it's always worked 
for me, and "The program no longer exists." seems very strange.

Can you force the app to dump a corefile, then run gdb on that?
("ulimit -c unlimited")

> My problem is that every time I try it when I try to run this failing
> command for the second time it stops
> k13 coreutils-5.94 # ps aux | grep shred
> root      2100  0.0  1.1   6480  4504 pts/1    S    13:14   0:00 gdb -x
> /dev/fd/63 --args src/shred --remove --zero adam adam1
> root      2101  0.0  0.1   1796   468 pts/1    D+   13:14   0:00
> /var/tmp/portage/coreutils-5.94-r1/work/coreutils-5.94/src/shred --remove --
> zero adam adam1
> in state D but I do not know how to resume it, I also cannot kill it. I have
> to reboot my server every 5 minutes to give it another try :-)

Unkillable processes stuck in state "D" sound like possible hardware 
error. Perhaps there are obvious error messages in the kernel log which 
could confirm this.


Cheers,
Phil