As I said before, I am fine with your code changes. I am pretty sure that introducing some weird checking into the DC code is probably not the way to go. However, we need to do something. The DC page on opensolaris.org says
In order to use the Distribution Constructor, you must have installed Solaris Nevada build 71 or a later build on your system. and that is clearly not true anymore. The SXDE 1/08 won't work (based on 79b) for several reasons, and preview 2 will probably fail due to the SUNWclofi issue. I think the main page on opensolaris.org needs to be updated, and there should be a bug against the man page for lofiadm. Channing Jean McCormack wrote: > Yes. That is true. Has nothing to do with the man pages though. > So we have the following choices: > > 1) hard code the checking per first review but that's a maintenance > burden if/when we add > new compression algorithms. > > 2) Check using lofiadm on the system which might not be the same as > what you will > build the CD with. Could give erroneous errors. Are there any > stipulations as to what > needs to be installed on the developer's machine? The only time this > becomes an issue > is when we add a new compression algorithm. However, that's now. > > 3) Could we move the checking until later in the build so we have > access to the lofiadm > that will be in the proto area? I think this would be a bad user > experience and be quite > inconsistent with the rest of our parameter checking. > > Jean > > > > Channing Lovely wrote: >> Ok, here is the scenario I am concerned with... >> >> our intrepid open source developer decides to start with snv_86 that >> he downloaded from Sun's website. >> >> He goes to opensolaris.org, downloads and installs the hg package. >> >> He goes to opensolaris.org, and hg clone's a distro constructor >> workspace. He uses one of the sample .conf files that shows >> >> COMPRESSION_TYPE=lzma >> >> and then runs >> >> ./build_dist.bash test1.conf (for example) >> >> and now things go bad, because he doesn't have SUNWclofi on his >> machine, so lofiadm complains and I think we are done. >> >> Now, if our intrepid developer had started using the RC0 cd, he would >> be fine because SUNWclofi is there. SUNWclofi isn't on preview2, I >> don't think, or on a "stock" snv install either. >> >> I think there is a hidden hard dependency on SUNWclofi being on the >> build machine or distro constructor may die. >> >> I could be wrong. If I am, please educate me. >> >> Channing >> >> Jean McCormack wrote: >>> Channing Lovely wrote: >>>> Ok, so you cleaned up the verbiage saying that the intrepid user >>>> must use a compression algorithm supported by lofiadm. How is our >>>> intrepid user supposed to know that? The lofiadm man page only >>>> talks about gzip. So, at this point we are distributing DC code >>>> that is using an option that is not legal per man pages, and if our >>>> public audience follows the instructions on "getting started with >>>> distro constructor" and try to build themselves, it will die. There >>>> is some breakage here, not sure where it needs to be addressed. >>> Why will it die? gzip is fine. lzma is fine even if the man pages >>> don't say so. >>> >>> Jean >>>> >>>> The code changes are fine, with the caveat noted above. >>>> >>>> Channing >>>> >>>> Jean McCormack wrote: >>>>> Dave Miner wrote: >>>>> >>>>>> Jean McCormack wrote: >>>>>> >>>>>>> Addresses: >>>>>>> http://defect.opensolaris.org/bz/show_bug.cgi?id=1125 >>>>>>> >>>>>>> Webrev: >>>>>>> http://cr.opensolaris.org/~jeanm/distro_constructor/ >>>>>>> >>>>>>> >>>>>> A couple of things: >>>>>> >>>>>> - It would be nice if this also allowed gzip-9, since lofiadm >>>>>> allows that to be used as well >>>>>> >>>>>> - Relatedly, I'd rather we didn't duplicate lofiadm's error >>>>>> checking of its command-line parameters, since this creates >>>>>> another place that needs changing if and when lofiadm acquires >>>>>> additional compression. You could pass this through to a dummy >>>>>> lofiadm invocation to let it validate, perhaps. >>>>>> >>>>>> Dave >>>>>> >>>>> I've updated the webrev. >>>>> >>>>> Jean >>>>> _______________________________________________ >>>>> caiman-discuss mailing list >>>>> caiman-discuss at opensolaris.org >>>>> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss >>>>> >>>> >>> >> >
