README-RH318-ClassroomSetup.txt This text file explains how to set up the classroom before teaching RH318. Before class, all instructors will need to download and set up the following content: a) Download/set up an RHCE 6.3 USB stick and RHCI CD b) Download the following RH318 specific files from trainingforums unless otherwise noted: jbappplatform-6-x86_64-server-6-rpm-.tar README-RH318-ClassroomSetup-*.txt (this file) rh318-config-3.1*.rpm rh318-errata-.tar RH318-RHEV3.1-*-slides.pdf rhel-server-supplementary-6.3-x86_64-dvd.iso (from RHN) rhel-x86_64-rhev-agent-6-server-.tar rhel-x86_64-rhev-mgmt-agent-6-.tar rhel-x86_64-server-6-rhevh-.tar rhel-x86_64-server-6-rhevm-3.1-.tar c) Download or obtain Microsoft content: Additional classroom setup content from Microsoft (Windows 7 Professional with SP1 ISOs and license keys) is necessary to run this class. Instructors working for Red Hat can obtain the required materials from trainingforums. External training partners will need to obtain and/or download appropriate Microsoft materials themselves as detailed below; Red Hat will not distribute this material. Red Hat instructors will need to download the following files from trainingforums and refer to the README-W7 file for specific instructions: en_windows_7_professional_n_with_sp1_x64_dvd_u_677207.iso keys.w7 External training partners will need to obtain the ISOs from Microsoft and will need to create a text file containing appropriate license keys named keys.w7. d) Place the RH318 specific files and Microsoft content on a USB stick labelled GLSINST in a top level directory called "rh318". Note: The above RH318 and Microsoft specific material runs a little over 5 GB and can be placed on its own media or the same media as your RHCE 6.3 USB stick provided you have a larger stick. If you use separate media, be sure that the second stick is also labeled GLSINST. Monday morning: PLEASE NOTE: The first three steps here configuring your central instructor system can take close to an hour. If you are unable to gain access to the facility 1.5 hours before class begins, expect to have the students enter the PXE boot statements on their pair of machines as part of the introduction, or perform the final step yourself during the introduction or during a short break following the introductions. 1) Install instructor machine per normal RHCE 6.3 classroom setup stopping once the step /root/bin/gls-verify-files reports normal continuing with the steps outlined here. 2) Install rh318 materials: - Mount /media/GLSINST that contains rh318 directory - Install package # yum install -y /media/GLSINST/rh318/rh318-config*.rpm - Expand/copy remaining material from USB # expand-rh318 - As instructed by expand-rh318, reboot instructor system 3) Finalize instructor system: - Secure instructor system (change instructor and root passwords) # passwd instructor # passwd - Optional: Configure the firewall for NAT # /root/bin/iptables-nat-masquerade 4) Install student workstations and instructor demo workstation For instructor: - On instructor system, install rhevm.pod0 VM # gls-setup-rhevm-pod0 - PXE boot the two additional instructor machines boot: h1p0 boot: h2p0 For students: - PXE boot every other machine (the first of a student pair) boot: desktopX (where X is the pod number 1-20) - PXE boot every other machine (the second of a student pair) boot: h1pX (where X is the pod number 1-20) PLEASE NOTE: The installation of desktopX will automatically install a virtual machine called rhevm (rhevm.podX.example.com) after it reboots. It would be in your best interest to warn students to stay off the machines to allow the installation *full use* of the system to speed its installation. 5) Destroy only the h1pX machines boot: destroy The machine will be reinstalled by the students and instructor as part of a demo and lab exercise. Leave the h2p0 machine installed as you will use it later in the class. NOTE: Technically the installation of h1pX was done to allow the instructor to verify that the system is currently installable with RHEL 6.3 in case some systems need to be "moved around" before class starts. YOU ARE NOW COMPLETE: The classroom should now be completely ready for the labs. TROUBLESHOOTING: - Projector resolution too low Projectors offering 1024x768 resolution appear to be the norm for many classrooms. The RHEV Administrative Portal will attempt to squeeze content in to that resolution - though admittedly, not well. You may need to adjust the size of individual panes to allow visualization of some of the "Tabs", i.e. narrow the left pane to widen the main pane. Keep an eye out for workstations offering that resolution also. - Resetting the rhevm VMs If you find yourself wanting to revert the rhevm VMs to their original state: On student machines: # lab-318-resetvm rhevm On instructor machine: # /var/ftp/pub/gls/ulsbin/lab-318-resetvm rhevm - Destroying the rhevm VMs If you find yourself wanting to entirely destroy the rhevm virtual machine on a particular desktop (or instructor): # virsh destroy rhevm # virsh undefine rhevm # rm /var/lib/libvirt/images/rhevm.* - Manually rebuilding the rhevm VMs You can run the scripts by hand to build the rhevm VMs after destroying them as above: On student machines: # install-318-vm rhevm Once the VM has finished installing and is shutdown, verify it: # virsh list --all Create snapshot/overlay: # lab-318-resetvm rhevm On instructor machine: # gls-setup-rhevm-pod0 - Powering down/powering up all machines Power down: * In RHEV-M, put the host system(s) in Maintenance * Once confirmed, shut down the host system(s) * Shut down the desktop systems (which will hibernate the rhevm system(s) * Power down the instructor machine Power up: * Power on the instructor machine * Power on the host system(s) * Power on the desktop system(s) * Fix the system clock on the rhevm machine(s) (see below) * Activate the host(s) and wait for the data center(s) to become active - After shutting down and powering up the rhevm system(s), the clock may become out of sync Symptoms: Cannot use rhevm-shell, cannot use the API, cannot login to RHEV-M with a message saying: "Authentication Failed. oVirt Engine clock is not synchronized with directory services (must be within 5 minutes difference). Please verify the clocks are synchronized and restart the service." Suspending/hibernating a virtual machine may cause the clock to become out of sync. On rhevm.podX, check the system clock and the hardware clock: # date # hwclock Often times, the system clock will be out of sync, but the hardware clock will be synchronized. If that is the case, use hwclock to sync the system clock using the hardware clock, then restart ntpd: # hwclock --hctosys # service ntpd restart If both the system clock and the hardware clock are out of sync, use ntpdate to update the clock: # service ntpd stop # ntpdate instructor # service ntpd start Once the system clock has been brought back into synchronization, restart the ovirt-engine service: # service ovirt-engine restart