On 08/27/10 05:08 PM, Darren Reed wrote:
Shawn Walker wrote:
...
It says *network* access; the pkgsend man page quite clearly states
that you can access a repository directly using file:// URIs.

But the path to the repository is the same as path to the depot given
with "-d" to pkg.depotd, right?

Given that it then seems to me that the depot and repository are one and
the same?

But, they are not. Yes, the path is the same because it is a path to a repository. pkg.depotd provides *network* access to a repository.

...
Why can't I configure both at the same time, through the same
configuration file?

Because they can be used independently.

How do I use a depot independently from a repository?

I can understand that I don't need a depot server in order to access a
repository using file://, but I don't understand what I can do with
pkg.depotd that isn't related to a repository - at least the man page
doesn't tell me that I can.

Yes, pkg.depotd itself is useless without a repository, but a repository can be used without pkg.depotd.

They're also independent in the sense that they're often managed by different groups of people. The group of people that produce and create package repositories (at least for us) is likely to be very different than the group that manages depot servers.

As a result, configuration and administration is separate.

If my depot only ever has one repository, surely this would seem
sensible?

No, see above.

Furthermore, why should it be possible for the default publisher in a
repository be allowed to be different from that of a depot?

A depot can't have a default publisher.

Then why does the pkg.depotd man page list "pkg/default_publisher" as a
configuration knob?
Is that a bug?

Yes; I'll file a bug to get it removed. There's been a lot of change in this area lately and that appears to have been missed.


Can there be multiple repositories in a single depot?

A depot can't have things "in it"; a depot just serves repositories.
If your question is whether a depot can serve multiple repositories,
that's something still under consideration.

That's counter intuitive to the normal use of the word "depot", when you
think of what a "bus depot" is. I suppose you might say that a "bus
depot" serves buses but there are most definitely buses in it at various
times of the day. A bus depot could be a home to different types of
buses, or even non-bus vehicles, but ... I don't know if you'd say there
were multiple repositories there.

Sorry, but I don't see it that way.

...
FWIW, I think the idea of there being a 1 to 1 relationship between a
depot server and a repository makes a lot of sense. To do otherwise
would seem to require changes to the pkg FMRI.

Not necessarily; one could differentiate simply by port number, or URI used to access the repository.

...
No, it doesn't correspond to the default settings for pkg/server. The
default setting for pkg/server is actually port 80; not 10000.

The 10000 default is a leftover from early developer use at the
beginning of the project. I'd rather remove the default logic entirely
if it's going to cause this much confusion.

The first examples in the pkgsend man page starts with "eval `pkgsend
open [email protected]`". How is a user expected to use that example as
something that will work?

As I said before, feel free to file RFEs at defect.opensolaris.org.

With that said, there are a number of cases where the example could or would work.

Some things are leftovers from early phases of the project; this is likely one of them.

...
Yes, it does give an example of an FMRI that contains only the
publisher's prefix plus the name of the package. That *is* a "full
FMRI" (whatever that means).

Well pkg(5) mentions "full FMRI", so I assume it has to mean something?

As in, "whatever you meant by full FMRI".

...
Please file an RFE with suggestions as to how you would like to see
man page text for pkg(5) expanded.

Thanks,
-Shawn
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss

Reply via email to