[debbugs-tracker] bug#24406: closed (rm command: different behaviors seen with indirect blocks of a file)

[GNU bug Tracking System]

Your message dated Sat, 10 Sep 2016 12:07:59 -0700
with message-id <address@hidden>
and subject line Re: bug#24406: rm command: different behaviors seen with 
indirect blocks of a file
has caused the debbugs.gnu.org bug report #24406,
regarding rm command: different behaviors seen with indirect blocks of a file
to be marked as done.

(If you believe you have received this mail in error, please contact
address@hidden)


-- 
24406: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=24406
GNU Bug Tracking System
Contact address@hidden with problems

--- Begin Message --- Subject: rm command: different behaviors seen with indirect blocks of a file Date: Sat, 10 Sep 2016 16:51:49 +0000

Hi,

 

I am Mahesh, part of the team that is working on a tool to undelete the files on ext3 file system.

One of the ideas on which the tool is based is “the rm command marks the inode free but does not zero out the data blocks and the indirect blocks”

While this is validated to be true on my version of Linux, Ubuntu 16.04, some of my friends using fedora found out that the indirect blocks at the 13^th pointer of the inode is indeed being zeroed.

 

This difference of behavior is seemingly mysterious to me. Request you to point me in the direction as to why only some distributions of Linux seems to zero out the indirect block pointers associated with the files.

 

Thanks,

Mahesh.

--- End Message ---

> --- Begin Message ---
>
>
>
> Subject: 
>
> Re: bug#24406: rm command: different behaviors seen with indirect blocks of a file
>
>
>
>
> Date: 
>
> Sat, 10 Sep 2016 12:07:59 -0700
>
>
>
>
> User-agent: 
>
> Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.2.0
>
>
>
>
> Kothagere Siddalinga, Mahesh wrote:
> > Request you to point me in the direction as to why only some distributions of 
> > Linux seems to zero out the indirect block pointers associated with the files.
>
> I'm afraid that bug-coreutils is not a good mailing list to query about this. 
> You might doing your homework by reading the relevant file systems' source code.
>
>
> --- End Message ---