Differences

This shows you the differences between two versions of the page.

--- cluster:144 [2015/12/11 16:20]
hmeij [Warewulf Golden Image]
+++ cluster:144 [2018/07/26 18:52] (current)
hmeij07 [Deploying]
@@ Line 45: / Line 45: @@
 As opposed to the stateless, which grabs it's OS content from the master node, in the "golden image" approach we are going to retrieve the content of a selected node (node b0 in this example) to the Warewul master (node petaltail in this case).
-Set ''rsync'' to work between the two. Unmount any NFS file systems on the node. Adjust the exclusion selections in file (ie after /home is unmounted on node I do want the mountpoint...scratch space, etc).
+Set ''rsync'' to work between the two. Unmount any NFS file systems on the node. Adjust the exclusion selections in template file (ie after /home is unmounted on node I do want the mount point...scratch space, etc).
-  * /usr/lib/libexec/warewulf/wwmkchroot/golden-ststem.tmpl
 <code>
 # minder: all NFS file systems unmounted?
+# or add rsync excludes in
+# /usr/libexec/warewulf/wwmkchroot/golden-system.tmpl
 mkdir /var/chgroots/goldimages; cd /var/chgroots/goldimages
@@ Line 65: / Line 67: @@
 wwsh object modify -s bootloader=sda b6
-wwsh object modify -s diskformat=sda1,sda3 b6
+wwsh object modify -s diskformat=sda1,sda7 b6
 wwsh object modify -s filesystems= \
 "mountpoint=/boot:dev=sda1:type=ext4:size=500, \
 dev=sda3:type=swap:size=2048, \
-mountpoint=/:dev=sda7:type=ext4:size=70000" \
+mountpoint=/:dev=sda7:type=ext4:size=fill" \
 b6
+# see note below on filesystems....
 </code>
-More on the ''filesystems'' option later. In this I want to set up /, /boot and swap. Sizes in MB. Only one hard drive on this node (sda).
+More on the ''filesystems'' option later. In this example I want to set up /, /boot and swap. Sizes in MB. Only one hard drive on this node (sda).
-Next we need to get the node booted and trasnfer the VNFS image made from the node b0 contents. At this time look on your master node in /var/lib/mysql and make sure you have enugh disk space (these VNFS images will be around 1 GB I observed). Also, back up that database with ''mysqldump''.
+Next we need to get the node booted and trasnfer the VNFS image made from the node b0 contents. At this time look on your master node in /var/lib/mysql and make sure you have enough disk space (these VNFS images will be around 1 GB as observed). Also, back up that database with ''mysqldump''.
 <code>
@@ Line 85: / Line 89: @@
 # switch node to image VNFS
 wwsh provision set b6 --vnfs=b0.chroot
+# Rajil adds for GPU
+# Keep in mind that nouveau should be disabled,
+wwsh provision set c038 --kargs=\
+"acpi_pad.disable=1 console=ttyS0,115200 rd.driver.blacklist=nouveau nouveau.modeset=0"
 # just to be prudent
@@ Line 104: / Line 113: @@
 The console of the target node will now show the IP being assigned, the ''getvnfs'' process (be patient) and finally the root login. Check around the node for improvements, tweeks, then adjust and recreate the VNFS.
-After all that is done, disable provisioning so that the master ignores the PXE boot and the target node botos of local disk.
+After all that is done, disable provisioning so that the master ignores the PXE boot and the target node boots of local disk.
 <code>
@@ Line 112: / Line 121: @@
 </code>
+**filesystems**
+This is currently not working as expected. In my first attempts I'd specify sda1 (size=500), sda2 (size=2048, type=swap) and sda3 (size=fill) but what I end up with is a standard layout it looks like.  Any sizes are also ignored. So for now I just pick the ones I want (sda1, sda3, sda7).
+Note: this problem turns out to be hardware related, it is not appearing on newer hardware
+ --- //[[hmeij@wesleyan.edu|Henk]] 2017/09/13 08:11//
+This also happens after I remove any UUID references in /etc/fstab, clean up /etc/mtab and clean any and all files in /dev/disk/by-uuid and rebuild VNFS.
+<code>
+fdisk -l
+Disk /dev/sda: 80.0 GB, 80026361856 bytes
+heads, 63 sectors/track, 9729 cylinders
+ Units = cylinders of 16065 * 512 = 8225280 bytes
+ Sector size (logical/physical): 512 bytes / 512 bytes
+ I/O size (minimum/optimal): 512 bytes / 512 bytes
+ Disk identifier: 0x000ce092
+    Device Boot      Start         End      Blocks   Id  System
+ /dev/sda1   *           1          13      104391   83  Linux
+ /dev/sda2              14        1543    12289725   83  Linux
+ /dev/sda3            1544        1798     2048287+  82  Linux swap / Solaris
+ /dev/sda4            1799        9729    63705757+   5  Extended
+ /dev/sda5            1799        2053     2048256   83  Linux
+ /dev/sda6            2054        2184     1052226   83  Linux
+ /dev/sda7            2185        9729    60605181   83  Linux
+df -h
+ Filesystem      Size  Used Avail Use% Mounted on
+ /dev/sda7        57G  1.9G   53G   4% /
+ tmpfs            12G     0   12G   0% /dev/shm
+ /dev/sda1        95M   49M   42M  54% /boot
+</code>
+Warewulf 3.6.99 and CentOS 6.5
+==== Part 2 ====
+To avoid the problems detailed above I started over with a fresh node.  First I installed CentOS 6.7 vanilla on the hard disk with partitions sda1 (/boot, 500 MB), sd2 (2048 MB, swap) and sda3 (/, size=fill). Since I had a new kernel now we need to make a new bootstrap image and provision it.
+<code>
+  wwbootstrap --chroot=/var/chroots/goldimages/n0.chroot 2.6.32-573.12.1.el6.x86_64
+  wwsh provision set n22 --vnfs=n0.chroot --bootstrap=2.6.32-573.12.1.el6.x86_64
+</code>
+Before we create the n0.chroot VNFS image edit the file and comment out these lines, with a golden image approach we do not have a hybridization path but we want these file systems included.
+<code>
+# /etc/warewul/vnfs.conf
+# comment out for golden image
+#hybridize += /usr/X11R6
+#hybridize += /usr/lib/locale
+#hybridize += /usr/lib64/locale
+#hybridize += /usr/include
+</code>
+I also cleaned the following files inside the chroot environment:
+<code>
+# /etc/fstab, edited (clean also mtab and remove files /dev/disk/by-uuid/*
+tmpfs                   /dev/shm                tmpfs   defaults        0 0
+devpts                  /dev/pts                devpts  gid=5,mode=620  0 0
+sysfs                   /sys                    sysfs   defaults        0 0
+proc                    /proc                   proc    defaults        0 0
+.11.103.42:/home /home nfs tcp,soft,intr,bg
+.11.103.42:/sanscratch /sanscratch nfs tcp,soft,intr,bg
+# /boot/grub/device.map
+# This file was written by Warewulf bootstrap (capability setup-filesystems)
+(hd0) /dev/sda
+# /etc/grub.conf
+# This file was written by Warewulf bootstrap (capability setup-filesystems)
+serial --speed= --unit= --word= --parity=
+terminal_input console serial; terminal_output console serial
+default 0
+timeout 10
+root (hd0,0)
+title CentOS release 6.7 - 2.6.32-573.12.1.el6.x86_64
+    kernel /vmlinuz-2.6.32-573.12.1.el6.x86_64 ro root=/dev/sda3 rhgb quiet
+    initrd /initramfs-2.6.32-573.12.1.el6.x86_64.img
+</code>
+Now when first provisioning happens the right partitions are created and the node imaged. With bootload=EXIT or simply shutting down the warewulf master dhcpd process, the node is now booting from local disk.
+Yea.
+==== Deploying ====
+As part of my deployment I edited out any device information in file ''/etc/udev/rules.d/70-persistent-net.rules'', that information will be filled in on boot. Remade my VNFS. The reason for this is that in migrating from RHEL 5.5 to CentOS 6.7 I observed the order of the NICs to reverse themselves (but not always). Not sure why but will tell me what NIC eth0 will be.
+Then we build a template file with node specs in it like so:
+<code>
+# HP blades reverses HWADDR (don't ask)
+ # use nic port bottom (no changes)
+ # set both ipaddr/hwaddr then PXEboot, wait...ww times out but images
+ # post edit
+ #   /etc/udev/rules.d fix eth? entries per below, then startup scripts
+ #   eth0 down CC 10.10.x.x onboot=no
+ #   eth1 up   CD 192.168.x.x onboot=yes
+ # reboot (with bootload=EXIT)
+ n1  1C:C1:DE:1C:82:D5 192.168.102.10 1C:C1:DE:1C:82:D4 10.10.102.10 10.11.103.10
+ n2  1C:C1:DE:1C:81:01 192.168.102.11 1C:C1:DE:1C:81:00 10.10.102.11 10.11.103.11
+...
+</code>
+My deploy script (give an entire line from above as arguments)
+<code>
+#!/bin/bash
+ # deploy a n0.chroot node via PXE golden image transfer
+ # dynamic files are always in stateless CHROOT/centos-6/root/wwtemplates
+ node=$1
+ hwaddr0=$2
+ ipaddr0=$3
+ hwaddr1=$4
+ ipaddr1=$5
+ ipaddri=$6
+ if [ $# != 6 ]; then
+         echo "missing args: node hwaddr0 ipaddr0 hwaddr1 ipaddr1 ipaddri"
+         exit
+ fi
+ wwsh object delete $node -y
+ sleep 3
+ wwsh node new $node --netdev=eth0 \
+ --hwaddr=$hwaddr0 --ipaddr=$ipaddr0 \
+ --netmask=255.255.0.0  --network=255.255.0.0 -y
+ wwsh node set $node --netdev=eth1 \
+ --hwaddr=$hwaddr1 --ipaddr=$ipaddr1 \
+ --netmask=255.255.0.0  --network=255.255.0.0 -y
+ wwsh node set $node --netdev=ib0 \
+ --ipaddr=$ipaddri \
+ --netmask=255.255.0.0  --network=255.255.0.0 -y
+ wwsh provision set $node --fileadd passwd,shadow,group -y
+ wwsh provision set $node --fileadd hosts,bashrc,profile -y
+ wwsh provision set $node --fileadd network.ww,ifcfg-eth1.ww,ifcfg-ib0.ww -y
+# note: no diskpartition, already exists and fails on this hardware,
+# otherwise add diskpartion=sda so that "mkbootable" succeeds
+ wwsh object modify -s bootloader=sda $node -y
+ wwsh object modify -s diskformat=sda1,sda4 $node -y
+ if [ "$node" == 'n22' ]; then
+ # golden images with 3 partitions
+ wwsh object modify -s filesystems="mountpoint=/boot:dev=sda1:type=ext4:size=1024,dev=sda2:type=swap:size=6144,mountpoint=/:dev=sda3:type=ext4:size=+" $node -y
+ else
+ # hp blade 4 partitions
+ wwsh object modify -s filesystems="mountpoint=/boot:dev=sda1:type=ext4:size=1024,dev=sda2:type=swap:size=6144,mountpoint=/:dev=sda4:type=ext4:size=+" $node -y
+ fi
+ wwsh provision set $node --vnfs=n0.chroot -y
+ wwsh provision set $node --bootstrap=2.6.32-573.12.1.el6.x86_64 -y
+ wwsh provision set --bootlocal=UNDEF $node -y
+ wwsh pxe update
+ wwsh dhcp update
+ service dhcpd restart
+ service httpd restart
+ echo "now reboot: $node"
+ echo "wwsh provision set --bootlocal=EXIT $node -y"
+</code>
 \\
 **[[cluster:0|Back]]**

DokuWiki

User Tools

Site Tools

Differences

Page Tools