John,
A couple of replies in-line ...
On 01/04/11 10:43, John Fischer wrote:
Ethan and Sue,
Thanks, great feedback. I appreciate the help.
In addition to your comments I have updated the copyright dates
for all files concerned. See below.
I'll produce an updated webrev shortly. I am waiting on the ON
gatekeepers to be sure that I get the copyright changes correct.
Once that happens I'll provide a new webrev along with a differential
webrev.
Thanks,
John
On 01/ 3/11 08:24 PM, Ethan Quach wrote:
installadm.c
---------------
1408 - Can you elaborate on this comment? It seems like the code is
figuring out whether or not we're dealing with a new service or an
old one and setting path accordingly, not 'ensuring' that we're
dealing with a new one.
Replaced:
/*
* Ensure we're dealing with a new service setup
*/
With:
/*
* The new server setup is located at var/ai/. The new
* service setup adds the service name to the end of this
* path. First, create a path assuming that we are dealing
* with a new service setup. Then ensure that this path
* exists and use it if it does. Otherwise, default to
* an older service setup that uses the port instead of
* the service name.
*/
var/ai -> /var/ai
setup-service.sh
---------------------
185-189 - If this function is used for when creating a new service,
when would $VARAI/$port already exist?
Right. There is no reason. If it exists then that would be a
problem. I have added
an else clause to the if statement which returns -1 (i.e., error).
What I meant was, if we're creating an install service, why do we need
to check for and create /var/ai/<port> at all in the new code? With
this new webserver code, creation of services (even for older builds)
should set things up for the service in /var/ai/<svcname> no matter what
right?
-ethan
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
