I had always supposed that HSFS filesystems and ISO9660 file systems (both
including the customary Rock Ridge extensions), were the same thing, but
seemingly not so.
 
I run a Sun Sparc machine using Solaris-10 on which I have a file

-rw-rw-r-- 1 chl chl   1445068800 Oct 20  2006 sol-10-u3-companion-ga.iso

which is the Sun Companion DVD as downloaded from Sun. I can attach it to
/dev/lofi/1 by using lofiadm (that's the Sun loop-back file system), and
then do

   mount -F hsfs /dev/lofi/1 /mountpoint

and it all works fine - I can see all the goodies on the Companion Disc
without problem. Hsfs is the file-system provided by Sun for CD-roms and
the like, and is supposedly compatible/identical to is06990).
 
Now I have burned it onto a DVD using cdrecord on a Linux system (an NSLU2
aka 'slug' running slugosbE as it happens). Moreover I have read it back
using readcd, and the read-back version is bit-for-bit identical to the
original, and can as easily be mounted via /dev/lof/* as before. So the
burning process seems to have worked faultlessly.
 
BUT, when I try to mout it on the Linux system, using

   mount -t iso9660 /dev/scd0 /mnt

it fails to mount, mumbling about "wrong fs type, bad option, bad
superblock on /dev/scd0, missing codepage or other error" (which appears
to be the generic error message for all mount failures), and the syslog
shows
 
May 22 15:19:14 (none) user.info kernel: attempt to access beyond end of device
May 22 15:19:14 (none) user.info kernel: sr0: rw=0, want=68, limit=4
May 22 15:19:14 (none) user.warn kernel: isofs_fill_super: bread failed, dev=sr0, iso_blknum=16, block=16
 
(sr0 is a soft link to scd0).
 
So How Come?
and Any Suggestions?
 
BTW, I also tried it as a udf file system with similar results.
 
Also, where can I find a list of FSTYPEs supported by Solaris; it seems to
recognise that 'udfs' is a valid FSTYPE (whereas 'foo' and 'iso9660' are
not), but won't mount it as such? And 'fstyp' tells me that it is 'hsfs',
as expected.
 
-- 
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131            Web: http://www.cs.man.ac.uk/~chl
Email: chl@clerew.man.ac.uk      Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9      Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5

Charles Lindsey wrote:

> it fails to mount, mumbling about "wrong fs type, bad option, bad
> superblock on /dev/scd0, missing codepage or other error" (which appears
> to be the generic error message for all mount failures), and the syslog
> shows
>  
> May 22 15:19:14 (none) user.info kernel: attempt to access beyond end of device
> May 22 15:19:14 (none) user.info kernel: sr0: rw=0, want=68, limit=4
> May 22 15:19:14 (none) user.warn kernel: isofs_fill_super: bread failed, dev=sr0, iso_blknum=16, block=16
>  
> (sr0 is a soft link to scd0).
>  
> So How Come?
> and Any Suggestions?

Look in /etc/filesystems, and check that you actually have HFS
installed. Also check that the hfs module is loaded.

-- 
I am Robert Billing, Christian, author, inventor, traveller, cook and
animal lover. "It burned me from within. It quickened; I was with book
as a woman is with child."

Quality e-books for portable readers: http://www.alex-library.com

In comp.unix.solaris Charles Lindsey <chl@clerew.man.ac.uk> wrote:
> Also, where can I find a list of FSTYPEs supported by Solaris;

/usr/lib/fs ?

> it seems to recognise that 'udfs' is a valid FSTYPE (whereas 'foo' and
> 'iso9660' are not), but won't mount it as such? And 'fstyp' tells me
> that it is 'hsfs', as expected.

'it' being fstyp or something else?  

Many fs-related programs (like fstyp, fsck, mount, etc.) are just
wrappers that call fs-specific versions of stuff in /usr/lib/fs. 

Try 'truss -f -t exec fstyp <device>'.  You'll see that it's calling
fs-specific versions of fstyp.  If you add support for a filesystem,
then it'll install support programs in there.

-- 
Darren

Charles Lindsey wrote:
> I had always supposed that HSFS filesystems and ISO9660 file systems (both
> including the customary Rock Ridge extensions), were the same thing, but
> seemingly not so.

Well - the kernel config help says:

config ISO9660_FS
      tristate "ISO 9660 CDROM file system support"
      help
        This is the standard file system used on CD-ROMs.  It was reviously
        known as "High Sierra File System" and is called "hsfs" on other
        Unix systems.

> I run a Sun Sparc machine using Solaris-10 on which I have a file
> 
> -rw-rw-r-- 1 chl chl   1445068800 Oct 20  2006 sol-10-u3-companion-ga.iso

....but that's no guarantee that Sun think so.

e.g., from http://osdir.com/ml/os.solaris.solarisx86/2005-01/msg00183.html

 > Solaris curently does not implement many iso-9660
 > features. Sun did declare "hsfs" to be in an "end-of-life"
 > sustained state (similar to pcfs).

Which makes one wonder what Sun expects to find on a CD.


-- 
              Just because I've written it doesn't mean that
                   either you or I have to believe it.

In <KK1yFI.C44@clerew.man.ac.uk> "Charles Lindsey" <chl@clerew.man.ac.uk> writes:

>I run a Sun Sparc machine using Solaris-10 on which I have a file

>-rw-rw-r-- 1 chl chl   1445068800 Oct 20  2006 sol-10-u3-companion-ga.iso

>Now I have burned it onto a DVD using cdrecord on a Linux system (an NSLU2
>aka 'slug' running slugosbE as it happens). Moreover I have read it back
>using readcd, and the read-back version is bit-for-bit identical to the
>original, and can as easily be mounted via /dev/lof/* as before. So the
>burning process seems to have worked faultlessly.
> 
>BUT, when I try to mout it on the Linux system, using

>   mount -t iso9660 /dev/scd0 /mnt

>it fails to mount, mumbling about "wrong fs type, bad option, bad
>superblock on /dev/scd0, missing codepage or other error" (which appears
>to be the generic error message for all mount failures), and the syslog
>shows

Since writing that, I have discovered how to use the loop-back fs on
Linux, and I was indeed able to mount that ISO image direct from hard disc.
So it is not an HSFS vs ISO9660 problem after all, but a purely Linux
problem. Hence Followups now restricted to just the Linux group in the
original bunch.

And then yesterday, it suddenly began to work! I.e. Linux can now mount it
directly from the DVD. So what had changed? The only thing I can think of
is that, after burning the DVD with cdrecord, I just left it in the drive
and tried to mount it. But Linux was not aware of its changed status. Just
pressing the eject button on the drive actually caused Linux to re-examine
the disc and decide that it was now a real-live-DVD, rather than an
unwritten blank, and so it was happy to mount it. Problem solved!

-- 
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131            Web: http://www.cs.man.ac.uk/~chl
Email: chl@clerew.man.ac.uk      Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9      Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5