Dave,
Please find responses inline. Took your suggestions unless otherwise noted.
current webrev: http://cr.opensolaris.org/~wmsch/profile4/
differences with last webrev http://cr.opensolaris.org/~wmsch/profile3-diff/
On 02/21/11 05:46 PM, Dave Miner wrote:
On 02/16/11 10:43 AM, William Schumann wrote:
To reviewers, thank you.
I have integrated some significant changes in two recent changesets:
- webserver design changes
- 7015429 Convert installadm.c to python
For the webserver changes:
I moved code formerly in my now defunct locate_profile.py to cgi_get_manifest.py
Modified ai_get_manifest.py to get:
- manifest
- any profiles
- any logging messages from the CGI that might be informative to the AI client
user on the console
Anyone familiar with webserver design changes, please check the above modules
for any incorrect assumptions.
I've not received any feedback on pyunit tests.
current webrev: http://cr.opensolaris.org/~wmsch/profile3/
differences with last webrev http://cr.opensolaris.org/~wmsch/profile2-diff/
last webrev: http://cr.opensolaris.org/~wmsch/profile2/
ai-webserver/Makefile, 87: had we previously discussed adding hostname here?
hostname has been in the design document since the early revisions. It was
missing from the Makefile due to an oversight.
I'm concerned about diverging possible criteria between manifests and profiles.
CR 7006611 addresses adding hostname for manifests, so this much has been decided. I'd suggest addressing manifests separately,
since I don't see any bugs here and the CR should be a fairly simple fix.
cgi_get_manifest.py
327: under what conditions do we run into a situation with criteria missing? Does it happen when we're serving older clients?
Just want to understand if we're going to be generating a lot of noise here or not.
At the moment, it shouldn't happen at all. I can see it happening in the future if new criteria is added - the old client sends the
new server different criteria. However, in this instance, the protocol version added by John would indicate the old protocol
version and the CGI would log accordingly.
434: Why return this message to the client? There's nothing it can do about it and really doesn't affect the validity of the
profile from the client's point of view.
Perhaps it is sufficient for the message to be sent only for 'validate' and 'create-profile' subcommands. Removing this logging
entirely from client/server.
export_profile.py, 120: s/delete/export/
ai_get_manifest.py, 49: now that we have build 159, /system/volatile would be preferred. I presume we're going to let the
upcoming AI->CUD relocate the manifest.xml our of /tmp?
That sounds like a more efficient approach. I'm working with the CUD engineers
so that this will be done correctly.
Thank you,
William
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss