Glibc-3.2.1 make check error Pure LFS

Bill's LFS Login lfsbill at wlmcs.com
Mon Apr 28 08:27:23 PDT 2003


On Mon, 28 Apr 2003, [iso-8859-1] Michael Villeza wrote:

>  > As Ryan suggested in his post, I think there is a
> > strong possibility
> > that the TIMEOUT needs to be increased. The failing
> > module contains
> >
> >    #define TIMEOUT 10
> >
> > which allows 10 seconds for the test to run. The
> > test writes some
> > testdata, reads it, runs 10000 random seeks 6 times.
> > The speed of
> > the cpu would suggest a slow HD also and the limited
> > RAM may be causing
> > a fair amount of swap activity. So I think if you
> > change
> > glibc-2.3.2/stdio-common/tst-rndseek.c, line 13
> > to... 30?, maybe it will
> > go.
> >
> I followed your suggestion and got another error.
> Something like large file or unable to write large
> file. I tried to re-run make check and to my surprise,
> there were no errors. Pls tell me if this is ok so I
> can continue Pure LFS without any worries.

First, I *think* you can continue. But I would hope Greg or Ryan or
someone more knowledgeable than I would confirm. My suspicion, without
doing any investigation, is that there is also a TIMEOUT for the large
file test and the setting you used (30?) was still not large enough fo
the large file test to complete successfully.

If you can investigate and confirm if this is so, maybe increasing the
timeout further will solve it. But you problably to get into your logs
and see exactly what the message is. Often, "something like..." is not
sufficient for even the expert folks to give a solid answer.

Did you run 'make -k check'? If so, it will also have done the tests
after the large file test and I would then say you are "good to go".
If not, I suggest you run again (maybe investigate/adjust the TIMEOUT
higher still) and then use the 'make -k check' so that tests continue
even if one fails.

Then you can determine if continuation is warranted.

-- 
Bill Maltby
lfsbill at wlmcs.com

-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message



More information about the lfs-dev mailing list