reiserfs4

reiserfs4

[Vladimir Lazarenko]

Greets,

Is there going to be some kind of a converter for new reiserfs version?
I'm running 2.4.22-rc1 now, with its current reiserfs implementation, and I 
heard many good things about reiserfs v4, would it be possible to convert 
filesystems without data loss? 

It's going to be a major pain if I'll have to back things up and reinitialize 
partitions...

Thanks in advance.

-- 
Best regards,
Vladimir Lazarenko

Re: reiserfs4

[Andreas Dilger]

On Aug 07, 2003  03:05 +0200, Vladimir Lazarenko wrote:
> Is there going to be some kind of a converter for new reiserfs version?
> I'm running 2.4.22-rc1 now, with its current reiserfs implementation, and I 
> heard many good things about reiserfs v4, would it be possible to convert 
> filesystems without data loss? 
> 
> It's going to be a major pain if I'll have to back things up and reinitialize 
> partitions...

Why do people ever want a "converter"?

If you are converting your current filesystem to an _experimental_
filesystem, wouldn't you want to have a backup in case the new filesystem
had a bug in it?

Considering that such a conversion tool would be used only very rarely,
wouldn't you want to make a backup in case the conversion tool was broken?

The safest conversion is to make a backup with tar or similar, and then
restore it after a formatting the new filesystem.

Cheers, Andreas
--
Andreas Dilger
http://sourceforge.net/projects/ext2resize/
http://www-mddsp.enel.ucalgary.ca/People/adilger/

Re: reiserfs4

[Ivan Gyurdiev]

> Why do people ever want a "converter"?

That's been discussed before.
Because people don't have the resources (hard disk space, tape drives, 
money)  to backup their data, and might still be interested in testing a 
new filesystem. They might be willing to take a risk with the new fs 
and converter. Amazing as it may sound, people do that. I am such a 
tester, and I'd find a converter to be a useful tool. But since the 
previous discussion on the subject concluded it'd be really hard to 
impossible to write one, I guess I'll have to settle for new hard drive(s).

"It's going to be a major pain if I'll have to back things up and 
reinitialize partitions..."
<-- This is why

Re: reiserfs4

[Oleg Drokin]

Hello!

On Thu, Aug 07, 2003 at 01:12:44AM -0400, Ivan Gyurdiev wrote:

> >Why do people ever want a "converter"?
> That's been discussed before.
> Because people don't have the resources (hard disk space, tape drives, 
> money)  to backup their data, and might still be interested in testing a 
> new filesystem. They might be willing to take a risk with the new fs 
> and converter. Amazing as it may sound, people do that. I am such a 
> tester, and I'd find a converter to be a useful tool. But since the 
> previous discussion on the subject concluded it'd be really hard to 
> impossible to write one, I guess I'll have to settle for new hard drive(s).

This is no longer true.
There is sort of "universal" fs convertor for linux that can convert almost
any fs to almost any other fs.
The only requirement seems to be that both fs types should have read/write support in Linux.
http://tzukanov.narod.ru/convertfs/

Bye,
    Oleg

Re: reiserfs4

[Michael Buesch]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thursday 07 August 2003 07:02, Andreas Dilger wrote:
> Why do people ever want a "converter"?

It's because most people don't know of the complexity
of a file-system and think that's a trivial thing.
And most people are lazy and blindly trust any software :)

- -- 
Regards Michael Buesch  [ http://www.8ung.at/tuxsoft ]
Penguin on this machine:  Linux 2.6.0-test2 - i386

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE/MiNVoxoigfggmSgRAtJAAJ0RTjMfzh8nRB/NGFDdLjMbi0WnXACeIAsa
JPjzxlprtgR1HQt+eFTLaYo=
=N8sT
-----END PGP SIGNATURE-----

Re: reiserfs4

[Tomas Szepe]

> [green@namesys.com]
> 
> > >Why do people ever want a "converter"?
> > That's been discussed before.
> > Because people don't have the resources (hard disk space, tape drives, 
> > money)  to backup their data, and might still be interested in testing a 
> > new filesystem. They might be willing to take a risk with the new fs 
> > and converter. Amazing as it may sound, people do that. I am such a 
> > tester, and I'd find a converter to be a useful tool. But since the 
> > previous discussion on the subject concluded it'd be really hard to 
> > impossible to write one, I guess I'll have to settle for new hard drive(s).
> 
> This is no longer true.
> There is sort of "universal" fs convertor for linux that can convert almost
> any fs to almost any other fs.
> The only requirement seems to be that both fs types should have read/write support in Linux.
> http://tzukanov.narod.ru/convertfs/

I'm afraid I cannot recommend using this tool.

A test conversion from reiserfs to ext3 (inside a vmware machine)
screwed up the data real horrorshow: directory structure seems
ok but file contents are apparently shifted.

-- 
Tomas Szepe <szepe@pinerecords.com>

Re: reiserfs4

[Hans Reiser]

Vladimir Lazarenko wrote:

>Greets,
>
>Is there going to be some kind of a converter for new reiserfs version?
>I'm running 2.4.22-rc1 now, with its current reiserfs implementation, and I 
>heard many good things about reiserfs v4, would it be possible to convert 
>filesystems without data loss? 
>
>It's going to be a major pain if I'll have to back things up and reinitialize 
>partitions...
>
>Thanks in advance.
>
>  
>
there is a third party working on something called convertfs which looks 
cool to me.  google for it....

-- 
Hans

Re: reiserfs4

[Matthias Schniedermeyer]

On Thu, Aug 07, 2003 at 03:21:11PM +0200, Tomas Szepe wrote:
> > [green@namesys.com]
> > 
> > > >Why do people ever want a "converter"?
> > > That's been discussed before.
> > > Because people don't have the resources (hard disk space, tape drives, 
> > > money)  to backup their data, and might still be interested in testing a 
> > > new filesystem. They might be willing to take a risk with the new fs 
> > > and converter. Amazing as it may sound, people do that. I am such a 
> > > tester, and I'd find a converter to be a useful tool. But since the 
> > > previous discussion on the subject concluded it'd be really hard to 
> > > impossible to write one, I guess I'll have to settle for new hard drive(s).
> > 
> > This is no longer true.
> > There is sort of "universal" fs convertor for linux that can convert almost
> > any fs to almost any other fs.
> > The only requirement seems to be that both fs types should have read/write support in Linux.
> > http://tzukanov.narod.ru/convertfs/
> 
> I'm afraid I cannot recommend using this tool.
> 
> A test conversion from reiserfs to ext3 (inside a vmware machine)
> screwed up the data real horrorshow: directory structure seems
> ok but file contents are apparently shifted.

That answers the question that poped up in my mind.

"How does the tool know where the blocks are, and in which order it can
'move' then without corrupting the data.(*)

Seems it doesn't know it.

But it is possibel(*2) to do what the programm wants to do, you only
have to find out the order in which you have to copy the blocks to
prevent garbage. That's all the magic.



*: I mean the point where it copies the data from the sparse-file to the
block-device.

*2: All you (theoretically) need is 1 free block, or you go over
temp-space(e.g. memory) somewhere else.


Bis denn

-- 
Real Programmers consider "what you see is what you get" to be just as 
bad a concept in Text Editors as it is in women. No, the Real Programmer
wants a "you asked for it, you got it" text editor -- complicated, 
cryptic, powerful, unforgiving, dangerous.

Re: reiserfs4

[Oleg Drokin]

Hello!

On Thu, Aug 07, 2003 at 04:23:12PM +0200, Matthias Schniedermeyer wrote:
> > > There is sort of "universal" fs convertor for linux that can convert almost
> > > any fs to almost any other fs.
> > > The only requirement seems to be that both fs types should have read/write support in Linux.
> > > http://tzukanov.narod.ru/convertfs/
> > I'm afraid I cannot recommend using this tool.
> > A test conversion from reiserfs to ext3 (inside a vmware machine)
> > screwed up the data real horrorshow: directory structure seems
> > ok but file contents are apparently shifted.
> That answers the question that poped up in my mind.
> "How does the tool know where the blocks are, and in which order it can
> 'move' then without corrupting the data.(*)

Well, there is FIBMAP ioctl that does this.

> Seems it doesn't know it.

It does. And our tests were more succesful, I believe.

> But it is possibel(*2) to do what the programm wants to do, you only
> have to find out the order in which you have to copy the blocks to
> prevent garbage. That's all the magic.

Sure.

Bye,
    Oleg

Re: reiserfs4

[Nikita Danilov]

Matthias Schniedermeyer writes:
 > On Thu, Aug 07, 2003 at 03:21:11PM +0200, Tomas Szepe wrote:
 > > > [green@namesys.com]
 > > > 

[...]

 > 
 > That answers the question that poped up in my mind.
 > 
 > "How does the tool know where the blocks are, and in which order it can
 > 'move' then without corrupting the data.(*)

There is FIBMAP ioctl for this.

 > 
 > Seems it doesn't know it.
 > 
 > But it is possibel(*2) to do what the programm wants to do, you only
 > have to find out the order in which you have to copy the blocks to
 > prevent garbage. That's all the magic.
 > 

Nikita.

Re: reiserfs4

[mdew]

On Fri, 2003-08-08 at 01:21, Tomas Szepe wrote:

> > This is no longer true.
> > There is sort of "universal" fs convertor for linux that can convert almost
> > any fs to almost any other fs.
> > The only requirement seems to be that both fs types should have read/write support in Linux.
> > http://tzukanov.narod.ru/convertfs/
> 
> I'm afraid I cannot recommend using this tool.
> 
> A test conversion from reiserfs to ext3 (inside a vmware machine)
> screwed up the data real horrorshow: directory structure seems
> ok but file contents are apparently shifted.

I'm looking at converting (sometime soon) a JFS system to XFS using
convertfs, I'm hoping this "converting" process will come out bug-free. 
Other than backing up all the data, and re-formating to XFS, would any
one have suggestings?

convertfs /dev/hde1 jfs xfs

-- 
mdew <mdew@mdew.dyndns.org>

Re: reiserfs4

[Ivan Gyurdiev]

> This is no longer true.
> There is sort of "universal" fs convertor for linux that can convert almost
> any fs to almost any other fs.
> The only requirement seems to be that both fs types should have read/write support in Linux.
> http://tzukanov.narod.ru/convertfs/

Tried the tool.
Didn't work for me, and I was told by the author that it requires 50% 
space for operation..which defeats the purpose for which I was hoping to 
use it (lack of space). If it works with 50% free, I suppose it will 
make people's lives easier, though the same could be achieved with a few 
resizes and parted. I was more interested in something that requires 
very little space to operate and is able to convert the fs.

Re: reiserfs4

[Matthias Schniedermeyer]

On Thu, Aug 07, 2003 at 06:25:44PM +0400, Oleg Drokin wrote:
> Hello!
> 
> On Thu, Aug 07, 2003 at 04:23:12PM +0200, Matthias Schniedermeyer wrote:
> > > > There is sort of "universal" fs convertor for linux that can convert almost
> > > > any fs to almost any other fs.
> > > > The only requirement seems to be that both fs types should have read/write support in Linux.
> > > > http://tzukanov.narod.ru/convertfs/
> > > I'm afraid I cannot recommend using this tool.
> > > A test conversion from reiserfs to ext3 (inside a vmware machine)
> > > screwed up the data real horrorshow: directory structure seems
> > > ok but file contents are apparently shifted.
> > That answers the question that poped up in my mind.
> > "How does the tool know where the blocks are, and in which order it can
> > 'move' then without corrupting the data.(*)
> 
> Well, there is FIBMAP ioctl that does this.
> 
> > Seems it doesn't know it.
> 
> It does. And our tests were more succesful, I believe.
> 
> > But it is possibel(*2) to do what the programm wants to do, you only
> > have to find out the order in which you have to copy the blocks to
> > prevent garbage. That's all the magic.
> 
> Sure.

Ups. Seems i am wrong, i "grep"ed for FIBMAP and the tool uses it. So i
guess the tool should be able to do everything correctly.

I take everything back and claim the opposite. :-)
(german idiom. "Ich nehm alles zurueck und behaupte das Gegenteil")




Bis denn

-- 
Real Programmers consider "what you see is what you get" to be just as 
bad a concept in Text Editors as it is in women. No, the Real Programmer
wants a "you asked for it, you got it" text editor -- complicated, 
cryptic, powerful, unforgiving, dangerous.

Re: reiserfs4

[Hans Reiser]

Tomas Szepe wrote:

>>[green@namesys.com]
>>
>>    
>>
>>>>Why do people ever want a "converter"?
>>>>        
>>>>
>>>That's been discussed before.
>>>Because people don't have the resources (hard disk space, tape drives, 
>>>money)  to backup their data, and might still be interested in testing a 
>>>new filesystem. They might be willing to take a risk with the new fs 
>>>and converter. Amazing as it may sound, people do that. I am such a 
>>>tester, and I'd find a converter to be a useful tool. But since the 
>>>previous discussion on the subject concluded it'd be really hard to 
>>>impossible to write one, I guess I'll have to settle for new hard drive(s).
>>>      
>>>
>>This is no longer true.
>>There is sort of "universal" fs convertor for linux that can convert almost
>>any fs to almost any other fs.
>>The only requirement seems to be that both fs types should have read/write support in Linux.
>>http://tzukanov.narod.ru/convertfs/
>>    
>>
>
>I'm afraid I cannot recommend using this tool.
>
>A test conversion from reiserfs to ext3 (inside a vmware machine)
>screwed up the data real horrorshow: directory structure seems
>ok but file contents are apparently shifted.
>
>  
>
Are you sure that vmware does not affect the result?

-- 
Hans

Re: reiserfs4

[Tomas Szepe]

> [reiser@namesys.com]
> 
> Tomas Szepe wrote:
> 
> >I'm afraid I cannot recommend using this tool.
> >
> >A test conversion from reiserfs to ext3 (inside a vmware machine)
> >screwed up the data real horrorshow: directory structure seems
> >ok but file contents are apparently shifted.
> >
> Are you sure that vmware does not affect the result?

No, I can't be sure.  (Nevertheless I'm inclined to believe
this is rather a FIBMAP related kernel bug that has been
introduced after the current version of the convertfs toolset
was released in March 2002.)

Unfortunately at the moment I don't have a scratch partition
on any of my disks so I can't re-test on a live system.

-- 
Tomas Szepe <szepe@pinerecords.com>