Jan Damborsky wrote: > Hi Joe, > > thank you very much for code review. > Please see my comments in line. > > Jan > > > Joseph J. VLcek wrote: >> Jan Damborsky wrote: >>> Hi Joe, >>> >>> could I please ask you to review the fix for following bug ? >>> >>> 5271 Serial console setting should be carried through to installed >>> system >>> >>> I am soliciting your feedback, since changes were done >>> in ICT code - the area you are familiar with :-) >>> >>> * Webrev: >>> http://cr.opensolaris.org/~dambi/bug-5271 >>> >>> Thank you very much, >>> Jan >>> >>> Modules affected and tested: >>> * ICT - ict.py >>> >>> tests carried out: >>> * created x86 AI image based on build 119 >>> >>> testing the fix >>> * AI installation done on x4100 via serial console >>> - 'console=ttya' added to GRUB menu >>> - on installed system: >>> - GRUB menu didn't contain GRUB splash image >>> - happy face boot was disabled >>> - /boot/solaris/bootenv.rc file contained correct >>> propagate 'console' property set to 'ttya' >>> >>> regression test >>> * AI installation done on W2110z using local graphic monitor+keyboard >>> - on installed system: >>> - GRUB menu contained GRUB splash image >>> - happy face boot was enabled >>> - /boot/solaris/bootenv.rc file contained correct >>> 'console' property set to 'text' >>> >> >> Issue: >> --------- >> >> I don't think the 'timeout 30' is needed in the GRUB menu if the >> splashimage is not added. >> >> Have you investigated that it is? > > To be honest, I originally didn't try to remove 'timeout' assuming it is > not > related to GRUB splash image itself. My understanding was that 'timeout' > makes > sure that GRUB doesn't sit in menu forever, but instead selects 'default' > entry once 'timeout' expires. > I have just verified that if 'timeout 30' is removed in scenario when > menu.lst doesn't contain graphic stuff (GRUB splash image and happy face > boot), > GRUB doesn't proceed with default entry automatically, but instead waits > for user choice forever. > I think that we would like to have boot process behaving as consistently > as possible in both scenarios - in that case we would like to have timeout > set in the second case as well. Please let me know what you think. > Thank you very much. >
OK Thanks for looking into it! I'm now good with it the way you have it. Joe
