Brock Pytlik wrote:
> Dave Miner wrote:
>> [EMAIL PROTECTED] wrote:
>>> Hi Brock,
>>>
>>>> For pkg install:
>>>> 1) If a fmri with the version including the timestamp is given, no 
>>>> catalog refresh will be done.
>>>>
>>>> 2) Else: If the fmri specifies an authority or the package was 
>>>> previously installed from a specific authority, then that 
>>>> authority's catalog will be refreshed, but no others will. If that 
>>>> authority is unavailable, installation will fail.
>>>>
>>>> 3) Else: The catalogs for the preferred authority and the authority 
>>>> the package was previously installed from will be refreshed. If 
>>>> neither authority is available, the install will fail. If at least 
>>>> one of the authorities is available, then installation will proceed.
>>> It might be worthwhile to start with a simpler approach first, and then
>>> make this more complex, as necessary.  I might save the time of the last
>>> refresh in the catalog attributes and then refresh all catalogs if we
>>> haven't refreshed in N minutes.
>>>
>>> For install operations, it might also make sense to add a command-line
>>> flag that will perform the install without refreshing the catalogs.  If
>>> we already have the manifest for a package, we might have its content in
>>> the cache.  If that's the case, we can perform the install without ever
>>> having to hit the network.  I'm assuming that this option won't get used
>>> very much, though.
>>>
>> In particular, I'd like to ensure that bulk operations such as those 
>> done by an installer or distro constructor aren't going to be 
>> unnecessarily slowed by automatic refresh.
>>
>> Dave
> Ok, I'll add an option flag to all this behavior to be turned off. Out 
> of curiosity, is there a reason those might choose to do "pkg install X" 
> "pkg install Y" "pkg install Z" etc... instead of "pkg install X Y Z"?
> 

Most of the distribution construction is done using the slim_install 
cluster, but there are certain packages which must be installed 
separately and in a defined order (SUNWcs, SUNWcsd) because of the way 
actions are ordered.

In the case of, say, automated installation using a user-defined package 
list, we might also opt for the individual invocations for the same 
reason, since we can't tell a priori whether the listed packages would 
be affected by similar problems.

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

Reply via email to