[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-xorriso] bug with handling symbolic links?

From: Thomas Schmitt
Subject: Re: [Bug-xorriso] bug with handling symbolic links?
Date: Sun, 22 Feb 2015 19:28:50 +0100


> I have applied your patch and successfully created three DVDs

Thanks for testing. I will bother you again,
when i believe to have a final solution.

> xorriso : UPDATE : Found matching MD5 session tag: start=32 size=2179631

Be aware that -for_backup resp. -xattr "on" -acl "on"
still imposes the risk of CE size being too large for
the Linux fs/isofs restriction.
E.g. if setfattr(1) and/or setfacl(1) gave a file on
your disk nearly 2 KB or more of information.

If you want MD5s recorded but avoid this risk,
use -md5 "on" instead of -for_backup.
The AAIP data associated with MD5 are small enough
to fit into the first 2 KB, if Rock Ridge name and
symbolic link target path stay within X/Open limits.

Further be aware that -check_media does not verify the
sanity of the ISO filesystem but only of the string
of blocks which was written during a session. It tests
the fidelity of the medium.
This is of course very important when errors appear
and the medium shall be excluded as culprit.

More filesystem oriented is testing of each single data file

  -check_md5_r SORRY / -- 

It tests whether xorriso (resp. libisofs) can retrieve the
byte strings that were read and written as input file content
when the ISO filesystem was generated.
(Does not check symbolic links, though).

> I have also run isovfy and recursive diffs on all of them, without any of
> the previously indicated errors.

Recursive diff is a very realistic test.

The xorriso resp. libisoburn packages contain a program
test/compare_file. It is designed to compare content
and attributes under control of a find(1) run.
See in the README files of the packages, the paragraph:
"Result comparison with self produced ISO images".

Have a nice day :)


reply via email to

[Prev in Thread] Current Thread [Next in Thread]