Hi, On Wed, Jul 13, 2011 at 09:22:05AM -0700, Steven A. DuChene wrote: > Is this a dead mailing list or is anyone else reading these messages > at all??? > > If anyone else is on here please respond with any suggestions or > even semi-random thoughts.
give "si_monitortk" a try: 1) on you master node, make sure that /etc/init.d/systemimager-server-monitord was started 2) add "MONITOR_SERVER=192.168.1.1 MONITOR_CONSOLE=yes" to the APPEND line of your pxelinux.cfg (where 192.168.1.1 has to be replaced by the IP of your master server) 3) while the node gets imaged, run "si_monitortk" on the master an double click the node's line you'll get all output in a new window Once the imaging finished, the log file will also be saved in /root/si_install.log on the node. In case that grub did not install correctly, I'd suggest booting your node with "grub.exe" from grub4dos via PXE (via "KERNEL grub.exe"). That way, you might get the node initially up even without correct grub on the harddisk. > Otherwise I guess I have to assume systemimager is a dead project. the currently greatest version can be found in Bernard's repository http://download.systemimager.org/~bli/systemimager/ These beta-rpms work very well even with ext4 and modern hardware (except one small missing recent patch for HP's cciss controller) Best, thomas > If Bernard or Brian are reading this, do I have to turn anything on > specifically to get the system to write messages to log files in > /tmp ? I see mention in the code of si.log as well as si_monitor.log > but I am not sure if either are used by default. > -- > Steven DuChene > > > -----Original Message----- > >From: "Steven A. DuChene" <linux-clust...@mindspring.com> > >Sent: Jul 12, 2011 6:41 PM > >To: sisuite-users@lists.sourceforge.net > >Subject: Re: Not seeing boel messages on serial console > > > >I did try adding an NFS mount option to the autoinstall master script > >for my image and then trying to copy over the log file. I can see the > >remote system mounting the directory but I do not get any files. > > > >Still trying to figure out what is going on. > >-- > >Steven DuChene > > > > > >-----Original Message----- > >>From: "Steven A. DuChene" <linux-clust...@mindspring.com> > >>Sent: Jul 12, 2011 5:58 PM > >>To: sisuite-users@lists.sourceforge.net > >>Subject: Re: Not seeing boel messages on serial console > >> > >>In looking in the functions file in /etc/init.d of the initrd I see > >>it is also logging to /tmp/si.log > >> > >>Can anyone suggest a way that I can somehow have that file transferred > >>back to my install server just before the node reboots? > >>-- > >>Steven DuChene > >> > >> > >>-----Original Message----- > >>>From: "Steven A. DuChene" <linux-clust...@mindspring.com> > >>>Sent: Jul 12, 2011 5:45 PM > >>>To: sisuite-users@lists.sourceforge.net > >>>Subject: No seeing boel messages on serial console > >>> > >>>I am trying to boot a bunch of nodes in a special microserver blade chassis > >>>where the nodes do not have a traditional vga or graphics console. The only > >>>console they have were I can see the progress or lack there-of is a serial > >>>console. Also the only installation method is via pxe boot. I have passed > >>>the "console=ttyS0,9600n8" to the kernel in the pxelinux.cfg boot config > >>>file > >>>and as a result I see the kernel boot messages on the serial console. Once > >>>the kernel is done booting however I do not see any further messages except > >>>for any rsync bits which are normally quiet other than three or four small > >>>rsync status messages at the end. If I use the "--listing" option when I > >>>pull the image from the golden client that is running RHEL6.1 I do indeed > >>>see the whole list of files being rsync'ed fly by but this really slows the > >>>install down to a crawl. I do not see ANY of the customary boel progress > >>>messages at all. I see where initramfs is detected and init called. Then > >>>it talks about how init is started with BusyBox v1.9.1 (2011-04-21 > >>>16:31:21 PDT), > >>>then the disk controller modules are loaded. > >>> > >>>The last messages I see are: > >>> > >>>NET: Registered protocal family 10 > >>>lo: Disabled Privacy Extensions > >>> > >>>and then just before the node reboots after rsyncing over all the files are > >>>three or four rsnc connect messages and then the node reboots. > >>> > >>> > >>>My basic problem is after the node reboots from the install, it does not > >>>boot > >>>up to the OS. It just sits there after accessing the default file that > >>>tells > >>>it to boot from the local disk and goes no further. My suspicion is that > >>>some > >>>part of the grub execution or something else at the end of the install is > >>>not proceeding correctly. But since I cannot see any of the boel or other > >>>progress log messages I am not sure what is failing. > >>> > >>>BTW, I also have a basic RHEL6.1 kickstart environment setup and the same > >>>nodes > >>>will pxe boot and install just fine from the redhat kickstart environment > >>>that > >>>also uses the "console=ttyS0,9600n8" in the kernel append line. > >>> > >>>Any thoughts? BTW, I am booting with a RHEL6.1 UYOK kernel and initrd that > >>>are > >>>generated on the client during the perpareclient step. > >>>-- > >>>Steven DuChene ------------------------------------------------------------------------------ AppSumo Presents a FREE Video for the SourceForge Community by Eric Ries, the creator of the Lean Startup Methodology on "Lean Startup Secrets Revealed." This video shows you how to validate your ideas, optimize your ideas and identify your business strategy. http://p.sf.net/sfu/appsumosfdev2dev _______________________________________________ sisuite-users mailing list sisuite-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sisuite-users
