Status of reiserfs in Redhat 2.4.7-10 kernel ?

Status of reiserfs in Redhat 2.4.7-10 kernel ?

[ahorn]

Hi folks,

Can anyone point me towards a list of bugs that existed in the
implementation of reiserfs in this redhat kernel (the last kernel released
with a redhat 7.2 distribution CD)

I may obtain support and talk to redhat, but I just thought I'd check here
first and see if anyone knew. The customer is very conservative and
upgrading the kernel is not an option for other reasons. Therefore I need
to provide assurance that we can or cannot viably use reiserfs with that
kernel.

Thanks in advance for your help.

Cheers,

Al

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Philippe Gramoullé]

Hello,

What was true for 2.4.3 should be almost the same for 2.4.7-10.

See: http://www.geocrawler.com/archives/3/3455/2002/9/150/9512408/

IIRC, Reiserfs developers recommend at least 2.4.18/19

Out of curiosity, what are the "other reasons" that you can't upgrade the kernel for ?

Thanks,

Philippe

On Tue, 15 Apr 2003 16:10:51 -0700 (PDT)
ahorn@deorth.org wrote:

  | 
  | Hi folks,
  | 
  | Can anyone point me towards a list of bugs that existed in the
  | implementation of reiserfs in this redhat kernel (the last kernel released
  | with a redhat 7.2 distribution CD)
  | 
  | I may obtain support and talk to redhat, but I just thought I'd check here
  | first and see if anyone knew. The customer is very conservative and
  | upgrading the kernel is not an option for other reasons. Therefore I need
  | to provide assurance that we can or cannot viably use reiserfs with that
  | kernel.
  | 
  | Thanks in advance for your help.
  | 
  | Cheers,
  | 
  | Al
  | 
  | 
  | 
  | 

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[ahorn]

On Wed, 16 Apr 2003, Philippe [ISO-8859-15] Gramoullé wrote:

>Date: Wed, 16 Apr 2003 01:35:25 +0200
>From: "Philippe [ISO-8859-15] Gramoullé" <philippe.gramoulle@mmania.com>
>To: reiserfs-list@namesys.com
>Subject: Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?
>
>
>Hello,
>
>What was true for 2.4.3 should be almost the same for 2.4.7-10.
>
>See: http://www.geocrawler.com/archives/3/3455/2002/9/150/9512408/

Ouch.. :/

>
>IIRC, Reiserfs developers recommend at least 2.4.18/19
>
>Out of curiosity, what are the "other reasons" that you can't upgrade the kernel for ?

Compatibility with other essential parts of the whole system design.
It's unfortunatem, but that's what I have to work with.

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Oleg Drokin]

Hello!

On Tue, Apr 15, 2003 at 04:10:51PM -0700, ahorn@deorth.org wrote:

> Can anyone point me towards a list of bugs that existed in the
> implementation of reiserfs in this redhat kernel (the last kernel released
> with a redhat 7.2 distribution CD)

The list is way too long (and contains some pretty critical bugs). I afraid it will take
quite some time to compose the full list, so unless you are willing to pay for that,
I'd better work on something other.

Bye,
    Oleg

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Hans Reiser]

ahorn@deorth.org wrote:

>Hi folks,
>
>Can anyone point me towards a list of bugs that existed in the
>implementation of reiserfs in this redhat kernel (the last kernel released
>with a redhat 7.2 distribution CD)
>
>I may obtain support and talk to redhat, but I just thought I'd check here
>first and see if anyone knew. The customer is very conservative and
>upgrading the kernel is not an option for other reasons. Therefore I need
>to provide assurance that we can or cannot viably use reiserfs with that
>kernel.
>
>Thanks in advance for your help.
>
>Cheers,
>
>Al
>
>
>
>
>
>  
>
I am afraid that I can only say that Redhat kernels are not standard 
kernels, and non-standard kernels should not be preferred for use in 
production systems.  They are not as well tested (or as debugged) as 
Linus/Marcelo kernels.  Linus/Marcelo kernels are the official kernels, 
and Linus and Marcelo, not RedHat or any other distro, are the Linux 
kernel maintainers.  I personally try to avoid any use of a non-standard 
kernel in systems I care about because I know that no distro has the 
testing ability of those who test the Marcelo kernels, and Marcelo's 
level of skill and caution exceeds that of any other maintainer I am 
familiar with in detail, but I am more conservative than most.

-- 
Hans

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Anders Widman]

> Hi folks,

> Can anyone point me towards a list of bugs that existed in the
> implementation of reiserfs in this redhat kernel (the last kernel released
> with a redhat 7.2 distribution CD)

> I may obtain support and talk to redhat, but I just thought I'd check here
> first and see if anyone knew. The customer is very conservative and
> upgrading the kernel is not an option for other reasons. Therefore I need
> to provide assurance that we can or cannot viably use reiserfs with that
> kernel.

Could you not compile a module with later reiserfs code?

> Thanks in advance for your help.

> Cheers,

> Al

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[ahorn]

On Wed, 16 Apr 2003, Hans Reiser wrote:

>Date: Wed, 16 Apr 2003 11:33:24 +0400
>From: Hans Reiser <reiser@namesys.com>
>To: ahorn@deorth.org
>Cc: reiserfs-list@namesys.com
>Subject: Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?
>
>ahorn@deorth.org wrote:
>
>I am afraid that I can only say that Redhat kernels are not standard
>kernels, and non-standard kernels should not be preferred for use in
>production systems.  They are not as well tested (or as debugged) as
>Linus/Marcelo kernels.  Linus/Marcelo kernels are the official kernels,
>and Linus and Marcelo, not RedHat or any other distro, are the Linux
>kernel maintainers.  I personally try to avoid any use of a non-standard
>kernel in systems I care about because I know that no distro has the
>testing ability of those who test the Marcelo kernels, and Marcelo's
>level of skill and caution exceeds that of any other maintainer I am
>familiar with in detail, but I am more conservative than most.
>
>

Believe me, you are most definitely preaching to the choir in this case. I
appreciate you taking the time to compose this as I'll use it for further
ammunition to try and persuade the customer that patching and upgrading is
A Good Thing.

Cheers,

Al

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Chris Mason]

On Wed, 2003-04-16 at 03:33, Hans Reiser wrote:

> I am afraid that I can only say that Redhat kernels are not standard 
> kernels, and non-standard kernels should not be preferred for use in 
> production systems.  They are not as well tested (or as debugged) as 
> Linus/Marcelo kernels.  Linus/Marcelo kernels are the official kernels, 
> and Linus and Marcelo, not RedHat or any other distro, are the Linux 
> kernel maintainers.  I personally try to avoid any use of a non-standard 
> kernel in systems I care about because I know that no distro has the 
> testing ability of those who test the Marcelo kernels, and Marcelo's 
> level of skill and caution exceeds that of any other maintainer I am 
> familiar with in detail, but I am more conservative than most.

Grin, Hans and I have had this debate before, you might want to read
through the reiserfs list archives for all the details.  Just look for
the thread where Hubert Mantel most recently posted.

The short version is that I would pick a current vendor kernel over a
current vanilla kernel any day for heavy production use.    In terms of
performance and stability the vendor kernel will usually do better.  The
kinds of QA done on large hardware and heavy workloads is significantly
better before a vendor release than a vanilla release.

The problem is in defining 'current'.  I haven't followed the rhas
releases enough to know if they still qualify.

Anyway there's no need to start that debate again, neither Hans nor I
have changed our minds ;-)  

-chris

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Hans Reiser]

Chris Mason wrote:

>On Wed, 2003-04-16 at 03:33, Hans Reiser wrote:
>
>  
>
>>I am afraid that I can only say that Redhat kernels are not standard 
>>kernels, and non-standard kernels should not be preferred for use in 
>>production systems.  They are not as well tested (or as debugged) as 
>>Linus/Marcelo kernels.  Linus/Marcelo kernels are the official kernels, 
>>and Linus and Marcelo, not RedHat or any other distro, are the Linux 
>>kernel maintainers.  I personally try to avoid any use of a non-standard 
>>kernel in systems I care about because I know that no distro has the 
>>testing ability of those who test the Marcelo kernels, and Marcelo's 
>>level of skill and caution exceeds that of any other maintainer I am 
>>familiar with in detail, but I am more conservative than most.
>>    
>>
>
>Grin, Hans and I have had this debate before, you might want to read
>through the reiserfs list archives for all the details.  Just look for
>the thread where Hubert Mantel most recently posted.
>
>The short version is that I would pick a current vendor kernel over a
>current vanilla kernel any day for heavy production use.    In terms of
>performance and stability the vendor kernel will usually do better.  The
>kinds of QA done on large hardware and heavy workloads is significantly
>better before a vendor release than a vanilla release.
>
>The problem is in defining 'current'.  I haven't followed the rhas
>releases enough to know if they still qualify.
>
>Anyway there's no need to start that debate again, neither Hans nor I
>have changed our minds ;-)  
>
>-chris
>
>
>
>
>  
>
Test suites cannot compete with real users.  Marcelo gets lots of real 
users before he releases a final version, and the distros don't.  
However, I would also add that a conservative person would wait until a 
few weeks after Marcelo releases, and confirm that it really was stable 
according to those on the linux kernel mailing list.

-- 
Hans

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 600 bytes --]

On Wed, 16 Apr 2003 01:14:09 PDT, ahorn@deorth.org said:

> Believe me, you are most definitely preaching to the choir in this case. I
> appreciate you taking the time to compose this as I'll use it for further
> ammunition to try and persuade the customer that patching and upgrading is
> A Good Thing.

Point out to the customer that unless they patch or upgrade, they're sitting
there with a big bulls-eye called 'PTRACE' painted on their forehead.

Oh, and any of all the OTHER security/reliability issues that have been already
found and fixed in the last 2 years since that kernel came out... 

[-- Attachment #2: Type: application/pgp-signature, Size: 226 bytes --]

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Gerrit Hannaert]

Hans Reiser wrote:

> Test suites cannot compete with real users.  Marcelo gets lots of real
> users before he releases a final version, and the distros don't. 
> However, I would also add that a conservative person would wait until
> a few weeks after Marcelo releases, and confirm that it really was
> stable according to those on the linux kernel mailing list.

I think there is something to be said for test suites *and* for testing
by a good number of users, and new kernels should probably get a good
dose of both before running on any production system. You should
probably get as much testing as you can get, and excluding test suites
doesn't sound like a good idea to me - which may not be what Hans meant,
though ;-).

That said, I would really like the vendors' and 'vanilla' kernels to
converge a lot more than they do now... I don't like the idea that some
differences in the source trees have more 'historic' motivations than
being new features or bugfixes (just a hunch). Personally I've been
switching back and forth between vendor and patched 'vanilla' kernels
for the last few years...

- Gerrit

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Hans Reiser]

Gerrit Hannaert wrote:

>Hans Reiser wrote:
>
>  
>
>>Test suites cannot compete with real users.  Marcelo gets lots of real
>>users before he releases a final version, and the distros don't. 
>>However, I would also add that a conservative person would wait until
>>a few weeks after Marcelo releases, and confirm that it really was
>>stable according to those on the linux kernel mailing list.
>>    
>>
>
>I think there is something to be said for test suites *and* for testing
>by a good number of users, and new kernels should probably get a good
>dose of both before running on any production system. You should
>probably get as much testing as you can get, and excluding test suites
>doesn't sound like a good idea to me - which may not be what Hans meant,
>though ;-).
>
Test suites are what you use before burdening real testers with bugs 
that can be found other than the hard way.

>
>That said, I would really like the vendors' and 'vanilla' kernels to
>converge a lot more than they do now... I don't like the idea that some
>differences in the source trees have more 'historic' motivations than
>being new features or bugfixes (just a hunch). Personally I've been
>switching back and forth between vendor and patched 'vanilla' kernels
>for the last few years...
>
What really infuriates me most is when vendors try to inhibit people 
from switching around.  I am lobbying the GSA (US agency that sets 
procurement policies) to try to get them to refuse to buy support 
contracts that are distro specific.

>
>- Gerrit
>
>
>
>  
>


-- 
Hans

Re: Status of reiserfs in Redhat 2.4.7-10 kernel ?

[Chris Mason]

On Thu, 2003-04-17 at 11:57, Hans Reiser wrote:
> >
> >I think there is something to be said for test suites *and* for testing
> >by a good number of users, and new kernels should probably get a good
> >dose of both before running on any production system. You should
> >probably get as much testing as you can get, and excluding test suites
> >doesn't sound like a good idea to me - which may not be what Hans meant,
> >though ;-).
> >
> Test suites are what you use before burdening real testers with bugs 
> that can be found other than the hard way.
> 

Test suites are only a small part of the testing done by vendors.  Many
of the bug reports come from partners and others trying the beta kernels
in real world situations.  These run from certifications by partners to
benchmarks, to people who are just trying to get every penny out of
their hardware.  

> What really infuriates me most is when vendors try to inhibit people 
> from switching around.  I am lobbying the GSA (US agency that sets 
> procurement policies) to try to get them to refuse to buy support 
> contracts that are distro specific.

Grin, we've been here before too.  It's easy to preach this kind of
thing when you don't run your own distro.   But it's also easy to buy a
non-distro specific support contract from various places, it's all a
matter of price in the end.

-chris