Re: Bigtop and fresh Ubuntu

[Roman Shaposhnik <ro...@shaposhnik.org>]

On Thu, Apr 28, 2016 at 11:01 AM, Konstantin Boudnik <co...@apache.org> wrote:
> John
>
> We don't set +x bit on the directories in s3, so the listing doesn't work.
> There's nothing special about these permissions - it has just happened this
> way. We either can fix the permissions (easy to do), or do as Evans suggested,
> which might be ok in the interim, but is a hassle in the long run.
>
> If there's no objections, I can just go ahead and try to fix the permissions
> first.

While I don't think there are any objections, I don't actually think S3 works
that way. Could you please try set +x and see it if it helps? It it doesn't we
can always do an indexing step and generate a whole bunch of index.html


Thanks,
Roman.

Re: Bigtop and fresh Ubuntu

[Roman Shaposhnik <ro...@shaposhnik.org>]

On Mon, May 2, 2016 at 12:31 PM, Konstantin Boudnik <co...@apache.org> wrote:
> Yeah, you're right - dunno what I was thinking about... S3 isn't a subject to
> the normal permission laws. So, if browsing is a real big deal for our users -
> we need to address it differently.
>
> The repo use case was always simple - to enable repository manager to install
> the packages. Web-browsing wasn't a part of it like ever. Which, honestly,
> makes all the sense for me.

Right. Well, we can also do a Groovy script than generates a bunch of
index.html.

Thanks,
Roman.

P.S. And I agree with you -- by "we" I really mean those who think it may
be important ;-)

Re: Bigtop and fresh Ubuntu

[Roman Shaposhnik <ro...@shaposhnik.org>]

On Mon, May 2, 2016 at 12:31 PM, Konstantin Boudnik <co...@apache.org> wrote:
> Yeah, you're right - dunno what I was thinking about... S3 isn't a subject to
> the normal permission laws. So, if browsing is a real big deal for our users -
> we need to address it differently.
>
> The repo use case was always simple - to enable repository manager to install
> the packages. Web-browsing wasn't a part of it like ever. Which, honestly,
> makes all the sense for me.

Right. Well, we can also do a Groovy script than generates a bunch of
index.html.

Thanks,
Roman.

P.S. And I agree with you -- by "we" I really mean those who think it may
be important ;-)

Re: Bigtop and fresh Ubuntu

[Konstantin Boudnik <co...@apache.org>]

Yeah, you're right - dunno what I was thinking about... S3 isn't a subject to
the normal permission laws. So, if browsing is a real big deal for our users -
we need to address it differently. 

The repo use case was always simple - to enable repository manager to install
the packages. Web-browsing wasn't a part of it like ever. Which, honestly,
makes all the sense for me.

Cos

On Sun, May 01, 2016 at 05:07PM, Roman Shaposhnik wrote:
> On Thu, Apr 28, 2016 at 11:01 AM, Konstantin Boudnik <co...@apache.org> wrote:
> > John
> >
> > We don't set +x bit on the directories in s3, so the listing doesn't work.
> > There's nothing special about these permissions - it has just happened this
> > way. We either can fix the permissions (easy to do), or do as Evans suggested,
> > which might be ok in the interim, but is a hassle in the long run.
> >
> > If there's no objections, I can just go ahead and try to fix the permissions
> > first.
> 
> While I don't think there are any objections, I don't actually think S3 works
> that way. Could you please try set +x and see it if it helps? It it doesn't we
> can always do an indexing step and generate a whole bunch of index.html

Re: Bigtop and fresh Ubuntu

[Konstantin Boudnik <co...@apache.org>]

Yeah, you're right - dunno what I was thinking about... S3 isn't a subject to
the normal permission laws. So, if browsing is a real big deal for our users -
we need to address it differently. 

The repo use case was always simple - to enable repository manager to install
the packages. Web-browsing wasn't a part of it like ever. Which, honestly,
makes all the sense for me.

Cos

On Sun, May 01, 2016 at 05:07PM, Roman Shaposhnik wrote:
> On Thu, Apr 28, 2016 at 11:01 AM, Konstantin Boudnik <co...@apache.org> wrote:
> > John
> >
> > We don't set +x bit on the directories in s3, so the listing doesn't work.
> > There's nothing special about these permissions - it has just happened this
> > way. We either can fix the permissions (easy to do), or do as Evans suggested,
> > which might be ok in the interim, but is a hassle in the long run.
> >
> > If there's no objections, I can just go ahead and try to fix the permissions
> > first.
> 
> While I don't think there are any objections, I don't actually think S3 works
> that way. Could you please try set +x and see it if it helps? It it doesn't we
> can always do an indexing step and generate a whole bunch of index.html