2010/12/7 Marcus Meissner :

On Tue, Dec 07, 2010 at 11:08:33AM +0100, Richard Guenther wrote:

...

On Tue, 7 Dec 2010, Cristian Morales Vega wrote:

...

I'm going to update the libogg package and do some cleanup in the process. My problem is with this:

# bug437293 %ifarch ppc64 Obsoletes:      libogg-64bit %endif #

Is it needed anymore? Probably bnc#437293 isn't security related? Why is it access restricted?

Looks like the SLES10 -> SLES11 upgrade path.

Yes, where we moved SLES 10 to SLES 11 from ppc32 to ppc64 as base.

OK. It's my understanding that SLE 10 is still supported, true? (SLE 9 not anymore?) -- To unsubscribe, e-mail: opensuse-packaging+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-packaging+help@opensuse.org

2010/12/7 Marcus Meissner :

On Tue, Dec 07, 2010 at 11:12:04AM +0100, Cristian Morales Vega wrote:

...

2010/12/7 Marcus Meissner :

...

On Tue, Dec 07, 2010 at 11:08:33AM +0100, Richard Guenther wrote:

...

On Tue, 7 Dec 2010, Cristian Morales Vega wrote:

...

I'm going to update the libogg package and do some cleanup in the process. My problem is with this:

# bug437293 %ifarch ppc64 Obsoletes:      libogg-64bit %endif #

Is it needed anymore? Probably bnc#437293 isn't security related? Why is it access restricted?

Looks like the SLES10 -> SLES11 upgrade path.

Yes, where we moved SLES 10 to SLES 11 from ppc32 to ppc64 as base.

OK. It's my understanding that SLE 10 is still supported, true? (SLE 9 not anymore?)

SLES 9 is still supported (regular until mid of next year, perhaps with extension programs), SLES 10 too for some more years (3 at least).

Thanks. "extension programs" are very long? It's difficult to know when this kind things can be removed this way. Now I would also need to see bnc#93878 (-fsigned-char removed from default cflags). This one is less obvious, perhaps should be reported upstream... -- To unsubscribe, e-mail: opensuse-packaging+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-packaging+help@opensuse.org

2010/12/7 Marcus Meissner :

On Tue, Dec 07, 2010 at 11:23:51AM +0100, Cristian Morales Vega wrote:

...

Thanks. "extension programs" are very long? It's difficult to know when this kind things can be removed this way.

The above Obsoletes is not really relevant for openSUSE itself, but it does not hurt either.

I think you are free to remove it right now if you want to.

How do you maintain SLE packages? Do you have patches over openSUSE's spec files to add this kind of things?

...

Now I would also need to see bnc#93878 (-fsigned-char removed from default cflags). This one is less obvious, perhaps should be reported upstream...

I dont think this signed char vs unsigned char is an issue anymore.

OK. But I would like to clarify the situation with this kind of bug reports. If they aren't security related, how they got "protected" to start with? Even if they would be security related, shouldn't them be public after some time (e.g. patch release + 2 months)? -- To unsubscribe, e-mail: opensuse-packaging+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-packaging+help@opensuse.org

On Tue, Dec 07, 2010 at 11:47:32AM +0100, Cristian Morales Vega wrote:

2010/12/7 Marcus Meissner :

...

On Tue, Dec 07, 2010 at 11:23:51AM +0100, Cristian Morales Vega wrote:

...

Thanks. "extension programs" are very long? It's difficult to know when this kind things can be removed this way.

The above Obsoletes is not really relevant for openSUSE itself, but it does not hurt either.

I think you are free to remove it right now if you want to.

How do you maintain SLE packages? Do you have patches over openSUSE's spec files to add this kind of things?

No patches. So far we split off the SLE package sets as SLE entered Beta. How we handle this with openSUSE -> SLE 12 is still open as the new community involved development gives its challenges.

...

...

Now I would also need to see bnc#93878 (-fsigned-char removed from default cflags). This one is less obvious, perhaps should be reported upstream...

I dont think this signed char vs unsigned char is an issue anymore.

OK. But I would like to clarify the situation with this kind of bug reports. If they aren't security related, how they got "protected" to start with? Even if they would be security related, shouldn't them be public after some time (e.g. patch release + 2 months)?

They are internal, as the product (SLE11 GA) is internal. That the bugs are referenced externally is not so good of course here, but the bugs will usually stay internal. Ciao, Marcus -- To unsubscribe, e-mail: opensuse-packaging+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-packaging+help@opensuse.org

2010/12/7 Stefan Seyfried :

On Tue, 7 Dec 2010 10:31:55 +0100 Cristian Morales Vega wrote:

...

I'm going to update the libogg package and do some cleanup in the process. My problem is with this:

# bug437293 %ifarch ppc64 Obsoletes:      libogg-64bit %endif #

Is it needed anymore? Probably bnc#437293 isn't security related? Why is it access restricted?

If you cannot access the bug, simply remove the condition from the specfile and let the ones that are paid to do it figure it out later.

Why should *you* bother?

I am pretty sure it is simply a SLES bug and those are top secret, you know.

I finally keep it but removed the -fsigned-char patch (without more info it's easier it breaks something in a system that uses unsigned as default char than it is for it to fix something). No need to get actively hostile to SLE for, at best, win a few microseconds in the dependency resolution. But it isn't the first time I find myself in this situation and every time I get a little nearer to your posture. I would really appreciate a change in the policy. -- To unsubscribe, e-mail: opensuse-packaging+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-packaging+help@opensuse.org