On 10/13/10 11:40 AM, Sue Sohn wrote:
Hi Jean and Ginnie,
transfer_cpio.py
66 How was the default size of 1000 chosen? What are implications
for changing
this number?
Totally random. The idea was that we would compute the default values
somehow.
Karen can probably expand upon this.
In fact, this is an area in general that we were going to look at
soon. All/many
of the checkpoints need a default value.
So you anticipate that this will be decided before you push?
I don't know. I've asked Darren to add it to our CUD discussion list for
tomorrow.
transfer_ips.py
245 Rather than not executing any of this if dry_run is set, would
it make
sense to go through the motions and print log messages saying what
you would
be doing?
That's an interesting thought. But we can't. Since we don't have an
api_inst we
can't actually figure out what to do.
If it's at all possible, it would be nice to at least log what you'd
be doing at a high level, such as
"Updating the preferred publisher, <publisher>".
If you can get more detail for some of the other dry_run sections,
that would be good.
An effort will be made because it is important. For IPS we can't provide
details that require knowledge of what
a current IPS image might have. i.e. replace preferred publisher <foo>
with <bar> . We should be able to
do something like "Setting preferred publisher to <bar>"
transfer_svr4.py
55 What are the maintenance implications of this version number?
A nightmare. But it's a pkg constraint. They do this with the pkg
command too.
Can you explain where the version number comes from and how we are
supposed to know to update it? Is there a flag day when the version is
updated?
The version number is in
/usr/lib/python2.6/vendor-packages/pkg/client/api.py. We are required to
provide the correct number if we want to either create or get the IPS
instance handle. As far as I know there hasn't been a flag day in the
past. It would be nice to figure out the correct way to track this.
Jean
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
