Re: [dev] Interesting Web Browser Decoupling Concept

[sylvain . bertrand]

On Fri, Jun 16, 2017 at 11:51:18AM -0400, Calvin Morrison wrote:
> http://man.openbsd.org/crypto.3

Indeed. Still actually an internal library of openssl, not a library split from
the network code with a life of its own, as it should be.

Maybe libreSSL will do things right like gnutls on this matter.

Was told NaCl from bernstein is gaining momentum and why not a libreSSL build
based on nettle crypto services?

wait and see.

-- 
Sylvain

Re: [dev] Interesting Web Browser Decoupling Concept

[sylvain . bertrand]

On Fri, Jun 16, 2017 at 06:47:22PM +0200, Kamil Cholewiński wrote:
> On Fri, 16 Jun 2017, sylvain.bertr...@gmail.com wrote:
> > When people are using the word "openbsd", it does designate the kernel 
> > project.
> > Not the userland projects.
> 
> Completely not true. Please check your facts.

Checked! It has been true for the 2 decades with all the people I did interact
on the matter.

Even worse: has been the same for all *BSD kernels...

-- 
Sylvain

Re: [dev] [sbase] rm missing error message?

[Mattias Andrée]

On Fri, 16 Jun 2017 21:00:40 +0200
Quentin Rameau  wrote:

> Hi,
> 
> > Sounds like it says you must not write those error message if -f is
> > used. Kind of a strange requirement as `2>/dev/null` would do that.
> >  
> > > I don't fully understand the wording in POSIX on the page[0].
> > > 
> > > [0]
> > > http://pubs.opengroup.org/onlinepubs/9699919799/utilities/rm.html   
> 
> “It is less clear that error messages regarding files that cannot be
> unlinked (removed) should be suppressed. Although this is historical
> practice, this volume of POSIX.1-2008 does not permit the -f option to
> suppress such messages.”
> 

Didn't read that part.


pgpQ5vLmOB2u7.pgp
Description: OpenPGP digital signature

Re: [dev] [sbase] rm missing error message?

[Quentin Rameau]

Hi,

> Sounds like it says you must not write those error message if -f is
> used. Kind of a strange requirement as `2>/dev/null` would do that.
>
> > I don't fully understand the wording in POSIX on the page[0].
> > 
> > [0]
> > http://pubs.opengroup.org/onlinepubs/9699919799/utilities/rm.html 

“It is less clear that error messages regarding files that cannot be
unlinked (removed) should be suppressed. Although this is historical
practice, this volume of POSIX.1-2008 does not permit the -f option to
suppress such messages.”

Re: [dev] [sbase] rm missing error message?

[Mattias Andrée]

Sounds like it says you must not write those error message if -f is used.
Kind of a strange requirement as `2>/dev/null` would do that.

On Fri, 16 Jun 2017 20:14:58 +0200
Hiltjo Posthuma  wrote:

> On Fri, Jun 16, 2017 at 02:08:24PM -0300, Marc Collin wrote:
> > Hello all.
> > 
> > I found a case where sbase rm command fails but doesn't output any
> > error message, making it look like it succeeded.
> > 
> > mkdir ./test
> > mkdir ./test/test
> > sudo chown root:root ./test
> > sudo chown root:root ./test/test
> > rm -rf ./test
> > 
> > rm won't output anything and exit (apparently) cleanly.
> > But the ./test directory won't be deleted.
> > Shouldn't a meaningful message be printed to warn about the failure?
> > 
> > Regards.
> >   
> 
> Hi,
> 
> Regarding the status code: you specify -f so the exit status is not
> modified[0].
> 
> I'm not sure if it's required to print a warning message with the -f option.
> My interpretation is it's not neccesary. However on OpenBSD it says:
> 
>   $ rm -rf test/
>   rm: test/test: Permission denied
>   rm: test: Operation not permitted
> 
> I don't fully understand the wording in POSIX on the page[0].
> 
> [0] http://pubs.opengroup.org/onlinepubs/9699919799/utilities/rm.html
> 



pgporVpVwjpde.pgp
Description: OpenPGP digital signature

Re: [dev] [sbase] rm missing error message?

[Hiltjo Posthuma]

On Fri, Jun 16, 2017 at 02:08:24PM -0300, Marc Collin wrote:
> Hello all.
> 
> I found a case where sbase rm command fails but doesn't output any
> error message, making it look like it succeeded.
> 
> mkdir ./test
> mkdir ./test/test
> sudo chown root:root ./test
> sudo chown root:root ./test/test
> rm -rf ./test
> 
> rm won't output anything and exit (apparently) cleanly.
> But the ./test directory won't be deleted.
> Shouldn't a meaningful message be printed to warn about the failure?
> 
> Regards.
> 

Hi,

Regarding the status code: you specify -f so the exit status is not
modified[0].

I'm not sure if it's required to print a warning message with the -f option.
My interpretation is it's not neccesary. However on OpenBSD it says:

$ rm -rf test/
rm: test/test: Permission denied
rm: test: Operation not permitted

I don't fully understand the wording in POSIX on the page[0].

[0] http://pubs.opengroup.org/onlinepubs/9699919799/utilities/rm.html

-- 
Kind regards,
Hiltjo

[dev] [sbase] rm missing error message?

[Marc Collin]

Hello all.

I found a case where sbase rm command fails but doesn't output any
error message, making it look like it succeeded.

mkdir ./test
mkdir ./test/test
sudo chown root:root ./test
sudo chown root:root ./test/test
rm -rf ./test

rm won't output anything and exit (apparently) cleanly.
But the ./test directory won't be deleted.
Shouldn't a meaningful message be printed to warn about the failure?

Regards.

Re: [dev] Interesting Web Browser Decoupling Concept

[Kamil Cholewiński]

On Fri, 16 Jun 2017, sylvain.bertr...@gmail.com wrote:
> When people are using the word "openbsd", it does designate the kernel 
> project.
> Not the userland projects.

Completely not true. Please check your facts.

Re: [dev] Interesting Web Browser Decoupling Concept

[Calvin Morrison]

On 16 June 2017 at 11:49,   wrote:
> On Fri, Jun 16, 2017 at 02:06:30PM +0200, Laslo Hunhold wrote:
>> On Fri, 16 Jun 2017 09:53:07 +
>> sylvain.bertr...@gmail.com wrote:
>>
>> Dear Sylvain,
>>
>> > openbsd is as shitty as linux and their security thingy is just
>> > bullshit.
>>
>> are you serious? LibreSSL is proof enough that OpenBSD's approach is
>> probably the sanest for a general solution.
>> Not to go too OT, but the one-system-approach we find with OpenBSD is
>> probably the best in the long term. I won't dare to imagine how many
>> man-years have been wasted to keep the different Linux distributions up
>> and running instead of concentrating the efforts into one solution.
>
> When people are using the word "openbsd", it does designate the kernel 
> project.
> Not the userland projects.
>
> Regarding the libreSSL project, heard it's less worse than openSSL. Usually my
> network crypto enabled software uses gnutls, not *SSL code. gnutls uses
> "nettle" as its crypto engine. Wonder when libreSSL will split their crypto
> engine from their network engine: if it's already done, what the name of their
> crypto engine?
>
> --
> Sylvain
>

http://man.openbsd.org/crypto.3

Re: [dev] Interesting Web Browser Decoupling Concept

[sylvain . bertrand]

On Fri, Jun 16, 2017 at 02:06:30PM +0200, Laslo Hunhold wrote:
> On Fri, 16 Jun 2017 09:53:07 +
> sylvain.bertr...@gmail.com wrote:
> 
> Dear Sylvain,
> 
> > openbsd is as shitty as linux and their security thingy is just
> > bullshit.
> 
> are you serious? LibreSSL is proof enough that OpenBSD's approach is
> probably the sanest for a general solution.
> Not to go too OT, but the one-system-approach we find with OpenBSD is
> probably the best in the long term. I won't dare to imagine how many
> man-years have been wasted to keep the different Linux distributions up
> and running instead of concentrating the efforts into one solution.

When people are using the word "openbsd", it does designate the kernel project.
Not the userland projects.

Regarding the libreSSL project, heard it's less worse than openSSL. Usually my
network crypto enabled software uses gnutls, not *SSL code. gnutls uses
"nettle" as its crypto engine. Wonder when libreSSL will split their crypto
engine from their network engine: if it's already done, what the name of their
crypto engine?

-- 
Sylvain

Re: [dev] Interesting Web Browser Decoupling Concept

[Laslo Hunhold]

On Fri, 16 Jun 2017 09:53:07 +
sylvain.bertr...@gmail.com wrote:

Dear Sylvain,

> openbsd is as shitty as linux and their security thingy is just
> bullshit.

are you serious? LibreSSL is proof enough that OpenBSD's approach is
probably the sanest for a general solution.
Not to go too OT, but the one-system-approach we find with OpenBSD is
probably the best in the long term. I won't dare to imagine how many
man-years have been wasted to keep the different Linux distributions up
and running instead of concentrating the efforts into one solution.

With best regards

Laslo Hunhold

-- 
Laslo Hunhold 

Re: [dev] Interesting Web Browser Decoupling Concept

[sylvain . bertrand]

On Thu, Jun 15, 2017 at 04:46:55PM +0200, Kamil Cholewiński wrote:
> On Thu, 15 Jun 2017, Dominykas Mostauskis  
> wrote:
> software. Go install base OpenBSD on a potato, it has everything you
> choices for hacking on C, Perl or shell. It has a toolchain, and loads

openbsd is as shitty as linux and their security thingy is just bullshit.

perl? really? lol.

-- 
Sylvain