Glibc Timeout errors - revisited

Erika Pacholleck pchllck at nexgo.de
Tue Jul 15 09:17:03 PDT 2003


[15.07.2003] James Robertson <-- :
> What I find
> interesting is that the author says that "/dev/shm is automagically
> created if you use devfs". I use devfs on my host, but do not have a
> /dev/shm directory or mount present.

Hoops, here it is present, I disabled the /dev/pts stuff, enabled
devfs and also the mount at boot (easier to append a nomount if I do
not want it).

> My host is linux kernel 2.4.20 and running glibc 2.3.1.  I do not have a
> copy of the config.cache file that the kernel was built with to see what
> the value of CONFIG_TMPFS is, but it does not seem to matter.

grep TMPFS /usr/src/linux-<ver>/.config will do.
And to my view it does matter, just because of your following error.

> If I try and mount either manually (after creating /dev/shm) or through
> the fstab entry, I get the following error:
> 
> mount: wrong fs type, bad option, bad superblock on tmpfs,
>        or too many mounted file systems.
> 
> the manual mount command was mount -t tmpfs tmpfs /dev/shm

A normal not mounted one (I do not even have an fstab entry for that)
created at boot time looks like this
drwxr-x-rx  1  root  root  /dev/shm/

I use the same manual command as you and it changes to
drwxrwxwrt  1  root  root  /dev/shm/

and /proc/mounts as well as a real mtab file shows it's mounted.
And then I put whatever I want into /dev/shm.

Guessing from your message it looks like not being in your kernel,
or better said unknown as filesystem which is recognizable/usable
externally by users.

> I wonder if tmpfs was turned on when the kernel was compiled? How does
> one find that out?

alternately to the first mentioned one, if that is a regularly built
kernel (and not a test kernel, just minimum, all compiled in, and then
quickly copied to boot) you could

grep TMPFS /usr/include/linux/autoconf.h

-- 
Erika ...---...: pacholleck at nexgo dot de
-- 
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