Hi Evan, - I checked bug 7838, it doesn't contain any detail of your implementation. I was hoping to understand more why you implemented things the way you do there. So, I guess I will ask my question here. :-)
1) Why add the "entire" package to the AI manifests, and why make it the first package? I think it is sufficient to get the version number from SUNWcsd or SUNWcs. I believe those are always there in any image or installation. 2) At this time, the version number for the "Install running on", and the "System installed with:" are printed at 2 separate locations in the log file. I think it is much better for people to find the information if they are printed one following the other. 3) In log_bld_info(), you used temp file to capture the output of "pkg.....". I think it is much simpler to use "popen()" to execute the command and get the output and completely avoid the temp file. 4) The format of the output contains the name of the package you used for querying the output, should we parse the output to not include the name of the package? Thanks, --Karen Evan Layton wrote: > I need two reviewers for: > > 7838 Log installer build number and installed software build number in > install_log > http://defect.opensolaris.org/bz/show_bug.cgi?id=7838 > > and > > 6810 AI image missing D-Trace toolkit > http://defect.opensolaris.org/bz/show_bug.cgi?id=6810 > > Webrev: > > http://cr.opensolaris.org/~evanl/7838/ > > > I've tested this with both an AI and slim CD image and get the proper > output. > > In the log you'll see lines similar to: > <OM Apr 15 19:14:09> Install running on: > > FMRI: pkg:/entire at 0.5.11,5.11-0.111:20090331T092149Z > > and > > <OM Apr 15 19:40:42> System installed with: > > FMRI: pkg:/entire at 0.5.11,5.11-0.111:20090415T070733Z > > > Thanks, > -evan > _______________________________________________ > caiman-discuss mailing list > caiman-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
