Hi, Jan, The proposed features meet our requirements, further more, we would have something like option <ai_auto_reboot>, if possible, can we add one more option in AI manifest, i.e. <ai_enable_ssh>, so when reboot after install finished, ssh is also enable. That would be convenient for post install verification.
Rgs Jeffrey jan damborsky ??: > There is a request from QE team as well as from users to allow > process of installation to be observed remotely. For LiveCD case, > this requirement was satisfied by fix for bug 6431 which introduced > support for new configuration parameter 'livessh'. Is set to 'enable', > ssh access is enabled to LiveCD environment. > The same feature should be available for Automated Installation. > > Also, there is a need to be able to run the Automated Installer > in debug verbose mode in cases when more information about > installation process is desired - for instance when there is > a failure to be evaluated or during testing. > > Based on this, there would be two parts to address those > requirements: > > [1] enable ssh for AI > --------------------- > The suggestion is to take advantage of 'livessh' option introduced > for LiveCD. If set to 'enable', AI would allow remote access > using ssh by running following commands during boot: > > * generate ssh keys > # /lib/svc/method/sshd -c > > * enable ssh SMF service > # /usr/sbin/svcadm enable network/ssh > > [2] Enable debug verbose mode for Automated Installer > ----------------------------------------------------- > New option 'install_debug' would be introduced. If set > to 'enable', AI engine would be run in verbose debug > mode providing more information about installation process > (captured in standard /tmp/install_log file). > > For x86, option would be defined in GRUB menu.lst file, > it would be looked up in install.conf file for Sparc. > > Known Issues: > ------------- > Different mechanisms are used for x86 & Sparc - both have > some disadvantages - please see bug 7789 for more details. > Once bug 7789 is fixed, unified approach will be used > for passing parameters for both architectures. > > Please let me know, if you think the proposal should be modified > or different approach should be taken. > > Thank you very much, > Jan > > _______________________________________________ > caiman-discuss mailing list > caiman-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/caiman-discuss -- Great Thanks & Best Regards --- Jeffrey Huang SWAN: x51683, Ext: 0086-10-62673683 Solaris QE Group Email: jeffrey.huang at sun.com China Engineering & Research Institute, Sun Microsystems 7/F Chuang Xin Plaza, Tsinghua Science Park, Beijing China
