public inbox for gentoo-dev@lists.gentoo.org
 help / color / mirror / Atom feed
* [gentoo-dev] depscan.sh segfault issue
@ 2001-11-11 23:25 jano
  2001-11-11 23:46 ` jano
  0 siblings, 1 reply; 5+ messages in thread
From: jano @ 2001-11-11 23:25 UTC (permalink / raw
  To: gentoo-dev

Hi,

my depscan.sh segfaulting issue is still here.  I upgraded baselayout to
next version (1.6.5 from 1.6.4), but couldn't figure out how to make it
work.
Per rec of some people on irc, I ran depscan.sh through strace.  Sorry, i
have no idea how to debug a seg fault, but the output of depscan through
strace is here, for anyone interested:

http://www.portablehole.net/~jano/fridge/strace.depscan.sh.bz2
(28k, decompresses to 3.5MB file)

system specs:
kernel 2.4.14 with preempt and xfs patches
p3 800x2, 512MB ram
via apollopro chipset





^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [gentoo-dev] depscan.sh segfault issue
  2001-11-11 23:25 [gentoo-dev] depscan.sh segfault issue jano
@ 2001-11-11 23:46 ` jano
  2001-11-12 12:17   ` Martin Schlemmer
  2001-11-12 20:57   ` Aron Griffis
  0 siblings, 2 replies; 5+ messages in thread
From: jano @ 2001-11-11 23:46 UTC (permalink / raw
  To: gentoo-dev

Ahhhh poop, feel like a dork replying to myself.  But I made some progress.

I've narrowed down the problem to the gpm init.d script.  For some reason,
depscan doesn't like it at all.  I messed around, and found an interesting,
albeit STRANGE, pattern:

if I named gpm to:
gpm-
gpm~
gpm1
gmp
gpf
mouse
mouse-script

and tried to add it using rc-update (makeing sure it was removed first),
depscan would segfault.  However, when I named gpm to
testing
test
blah
foobar
it worked great.

I also tried commenting out everything inside the script, moving stuff
around, blah blah blah, no go.

Uhhhm.... HUH???????????????????????  This is mucked up. :D

>
> Hi,
>
> my depscan.sh segfaulting issue is still here.  I upgraded baselayout
> to next version (1.6.5 from 1.6.4), but couldn't figure out how to make
> it work.
> Per rec of some people on irc, I ran depscan.sh through strace.  Sorry,
> i have no idea how to debug a seg fault, but the output of depscan
> through strace is here, for anyone interested:
>
> http://www.portablehole.net/~jano/fridge/strace.depscan.sh.bz2
> (28k, decompresses to 3.5MB file)
>
> system specs:
> kernel 2.4.14 with preempt and xfs patches
> p3 800x2, 512MB ram
> via apollopro chipset
>
>
>
> _______________________________________________
> gentoo-dev mailing list
> gentoo-dev@cvs.gentoo.org
> http://cvs.gentoo.org/mailman/listinfo/gentoo-dev






^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [gentoo-dev] depscan.sh segfault issue
  2001-11-11 23:46 ` jano
@ 2001-11-12 12:17   ` Martin Schlemmer
  2001-11-12 20:57   ` Aron Griffis
  1 sibling, 0 replies; 5+ messages in thread
From: Martin Schlemmer @ 2001-11-12 12:17 UTC (permalink / raw
  To: gentoo-dev

[-- Attachment #1: Type: text/plain, Size: 1701 bytes --]

On Mon, 2001-11-12 at 07:49, jano wrote:
> 
> Ahhhh poop, feel like a dork replying to myself.  But I made some progress.
> 
> I've narrowed down the problem to the gpm init.d script.  For some reason,
> depscan doesn't like it at all.  I messed around, and found an interesting,
> albeit STRANGE, pattern:
> 
> if I named gpm to:
> gpm-
> gpm~
> gpm1
> gmp
> gpf
> mouse
> mouse-script
> 
> and tried to add it using rc-update (makeing sure it was removed first),
> depscan would segfault.  However, when I named gpm to
> testing
> test
> blah
> foobar
> it worked great.
> 
> I also tried commenting out everything inside the script, moving stuff
> around, blah blah blah, no go.
> 
> Uhhhm.... HUH???????????????????????  This is mucked up. :D
> 

Yep.  I cant think it being depscan.sh itself, rather maybe a 
bug in bash, or something.  Try another version bash, or maybe
remerge bash with less optimizations.

> >
> > Hi,
> >
> > my depscan.sh segfaulting issue is still here.  I upgraded baselayout
> > to next version (1.6.5 from 1.6.4), but couldn't figure out how to make
> > it work.
> > Per rec of some people on irc, I ran depscan.sh through strace.  Sorry,
> > i have no idea how to debug a seg fault, but the output of depscan
> > through strace is here, for anyone interested:
> >
> > http://www.portablehole.net/~jano/fridge/strace.depscan.sh.bz2
> > (28k, decompresses to 3.5MB file)
> >
> > system specs:
> > kernel 2.4.14 with preempt and xfs patches
> > p3 800x2, 512MB ram
> > via apollopro chipset
> >
> >


Greetings,

-- 

Martin Schlemmer
Gentoo Linux Developer, Desktop Team Developer
Cape Town, South Africa


[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [gentoo-dev] depscan.sh segfault issue
  2001-11-11 23:46 ` jano
  2001-11-12 12:17   ` Martin Schlemmer
@ 2001-11-12 20:57   ` Aron Griffis
  2001-11-12 21:16     ` jano
  1 sibling, 1 reply; 5+ messages in thread
From: Aron Griffis @ 2001-11-12 20:57 UTC (permalink / raw
  To: gentoo-dev

jano wrote:	[Mon Nov 12 2001, 12:49:20AM EST]
> I've narrowed down the problem to the gpm init.d script.  For some reason,
> depscan doesn't like it at all.  I messed around, and found an interesting,
> albeit STRANGE, pattern:

Could you please change the top line of /etc/init.d/depscan.sh to

    #!/bin/bash -x

then run your test with the script named "gpm", and send me the output?
Don't send it to the list... it will be lengthy.  You can run rc-update
as follows (or with whatever arguments you're using).

    rc-update add gpm default &> rc-update.out

Thanks,
Aron

P.S. No guarantees... I'm particularly busy at the moment, but I'd like
to take a look.



^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [gentoo-dev] depscan.sh segfault issue
  2001-11-12 20:57   ` Aron Griffis
@ 2001-11-12 21:16     ` jano
  0 siblings, 0 replies; 5+ messages in thread
From: jano @ 2001-11-12 21:16 UTC (permalink / raw
  To: gentoo-dev

Aha, that did it.  Thanks for the bash -x.  Here's the problem.  The gpm
init script has:

need {
need localmount
}

at the top.  It should've been "depend {}", not "need{}".

Looking at files/gpm.rc6, seems this has been fixed, but fixed after I
installed it.  I wonder why it "worked" if I renamed the script to test?

Problem solved :)

> jano wrote:	[Mon Nov 12 2001, 12:49:20AM EST]
>> I've narrowed down the problem to the gpm init.d script.  For some
>> reason, depscan doesn't like it at all.  I messed around, and found an
>> interesting, albeit STRANGE, pattern:
>
> Could you please change the top line of /etc/init.d/depscan.sh to
>
>    #!/bin/bash -x
>
> then run your test with the script named "gpm", and send me the output?
> Don't send it to the list... it will be lengthy.  You can run rc-update
> as follows (or with whatever arguments you're using).
>
>    rc-update add gpm default &> rc-update.out
>
> Thanks,
> Aron
>
> P.S. No guarantees... I'm particularly busy at the moment, but I'd like
> to take a look.
>
> _______________________________________________
> gentoo-dev mailing list
> gentoo-dev@cvs.gentoo.org
> http://cvs.gentoo.org/mailman/listinfo/gentoo-dev






^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2001-11-13  4:15 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-11-11 23:25 [gentoo-dev] depscan.sh segfault issue jano
2001-11-11 23:46 ` jano
2001-11-12 12:17   ` Martin Schlemmer
2001-11-12 20:57   ` Aron Griffis
2001-11-12 21:16     ` jano

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox