On 01/ 8/10 03:41 PM, [email protected] wrote:
On Fri, Jan 08, 2010 at 02:45:50PM -0600, Shawn Walker wrote:
I keep thinking that the transport should be shutdown in some other
case too.
Did you have any specific case in mind? When the client is done with
the download, we're almost certainly done unless an informational
operation is requested later by the GUI.
Just the info/list cases.
The one other case is where we prepare and evaluate the plan, but
pre-execute is never called because the user cancels before the plan is
executed. I could only see that possibly being an issue in the GUI for
the moment though since the CLI will exit if the plan isn't executed.
How long will the connections stay open/idle? I'm primarily
thinking of the pkg list / pkg info cases in the GUI.
Keep in mind that with pipelining, we only keep one connection open per
host as it is. Libcurl gets to decide how long it caches connections.
That's not tunable by the libary's interface. For informational
operations, chances are that you'll open only one (or re-use an existing
connection).
Remember that the problem we saw in the test suite was the consequence
of generating lots of Transport objects, each with one connection, that
weren't garbage collected during the test run. This is a bit of a
corner case, IMO.
Yeah; I'm happy with this changeset as is then. Ship it.
--
Shawn Walker
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
