Differences

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

--- cluster:154 [2017/03/09 15:02]
hmeij07 [Deploy]
+++ cluster:154 [2018/07/31 12:38]
hmeij07
@@ Line 2: / Line 2: @@
 **[[cluster:0|Back]]**
-==== OpenHPC ====
+==== OpenHPC page 1====
   * install vanilla CentOS 7.2 on //master//
@@ Line 10: / Line 10: @@
 <code>
+[root@ohpc0-test ~]# systemctl stop NetworkManager
+[root@ohpc0-test ~]# systemctl disable NetworkManager
 [root@ohpc0-test ~]# systemctl stop firewalld
@@ Line 50: / Line 54: @@
 Chain OUTPUT (policy ACCEPT)
 target     prot opt source               destination
+# copy global hpc /etc/hosts in place
+# check hostname is on rpovisionng network
+[root@ohpc0-test ~]# ping `hostname`
+PING ohpc0-test (192.168.1.249) 56(84) bytes of data.
+bytes from ohpc0-test (192.168.1.249): icmp_seq=1 ttl=64 time=0.043 ms
 [root@ohpc0-test ~]# reboot
@@ Line 112: / Line 122: @@
 # -ni not -pi
 perl -ni -e "print unless /^\s+Order allow,deny/" /etc/httpd/conf.d/warewulf-httpd.conf
+# the recipe does not set a mysql root password but we will
+[root@ohpc0-test]# vi /etc/warewulf/database-root.conf
+mysql> set password for 'root'@'localhost' = PASSWORD('some_string');
+Query OK, 0 rows affected (0.00 sec)
+[root@ohcp0-test]# chmod o-r /etc/warewulf/database-root.conf
 # restart/enable services
@@ Line 181: / Line 199: @@
 # finally on **master** issue
 perl -pi -e "s/ControlMachine=\S+/ControlMachine=ohpc0-test/" /etc/slurm/slurm.conf
+# this turned out to be wrong, first I change the hostname to ''ohpc0-slurm'' ControllerMachine=
+# added line in /etc/hosts pointing this to 192.168.1.249, then I defined ''/etc/slurm/slurm.con''
+NodeName=ohpc0-slurm NodeAddr=192.168.1.249 CPUs=2 \
+        RealMemory=8 Sockets=2 CoresPerSocket=4 ThreadsPerCore=2 State=UNKNOWN
+NodeName=n29 NodeAddr=192.168.102.38 CPUs=2 \
+        RealMemory=8 Sockets=2 CoresPerSocket=4 ThreadsPerCore=2 State=UNKNOWN
+NodeName=n31 NodeAddr=192.168.102.40 CPUs=2
+        RealMemory=8 Sockets=2 CoresPerSocket=4 ThreadsPerCore=2 State=UNKNOWN
+PartitionName=test Nodes=n29,n31 Default=YES MaxTime=INFINITE State=UP
 echo "/home *(rw,no_subtree_check,fsid=10,no_root_squash)" >> /etc/exports
 echo "/opt/ohpc/pub *(ro,no_subtree_check,fsid=11)" >> /etc/exports
@@ Line 236: / Line 269: @@
                          mountpoint=/:dev=sda3:type=ext3:size=fill" -t node ohpc0
 wwsh -y object modify -s bootlocal=UNDEF -t node ohpc0
-systemctl restart dhcpd
 wwsh pxe update
+wwsh dhcp update
+systemctl restart dhcpd
+systemctl restart httpd
@@ Line 258: / Line 294: @@
 : NODENAME   = ohpc0
 : VNFSID     = 2
+# Strange that netmask is not listed, but when I recreated the object I used
+wwsh -y node new ohpc0 --ipaddr=192.168.1.248 --hwaddr=00:15:c5:ef:0c:bf \
+     --netdev=enp4s0 --netmask=255.255.0.0  --network=255.255.0.0
 [root@ohpc0-test ~]# wwsh -y bootstrap list
@@ Line 269: / Line 309: @@
 </code>
+Final notes. I now have a 3 node OpenHPC cluster up using CentOS 7.3.1611 ... because of the edit mention above of the provision template URL the CHROOT is at the latest version of CentOS. Thus I updated my SMS master too so that the construct bootstrap=`uname -r` builds an image compatible between SMS and CHROOT.
+On towards testing the tools.
+Made a little script to recreate nodes as we'll do this often. And I also some weirdo situation where eth0/1 change NIC location during PXE boot, so I work around it.
+  * ''deploy.sh''
+<code>
+#!/bin/bash
+# enable both NIC to boot from 501/500
+# provision black on bottom, red on top (handler swithces to this)
+# set bootlocal to EXIT reboot, handler exits
+# switch black to top reboot (no media, fails to hdd)
+# insane
+node=n31
+ipaddr0=192.168.102.40
+hwaddr0=1c:c1:de:19:40:6f
+wwsh object delete $node -y
+wwsh node new $node --netdev=eth0 --hwaddr=$hwaddr0 --ipaddr=$ipaddr0 --netmask=255.255.0.0  --network=255.255.0.0 -y
+wwsh -y provision set $node --vnf=centos7.2 --bootstrap=`uname -r` --files=dynamic_hosts,passwd,group,shadow,slurm.conf,munge.key,network
+wwsh object modify -s bootloader=sda $node -y
+wwsh object modify -s diskpartition=sda $node -y
+wwsh object modify -s diskformat=sda1,sda2,sda3 $node -y
+wwsh object modify -s filesystems="mountpoint=/boot:dev=sda1:type=ext3:size=1024,dev=sda2:type=swap:size=6144,mountpoint=/:dev=sda3:type=ext3:size=+" $node -y
+wwsh provision set --bootlocal=UNDEF $node -y
+node=n31e
+ipaddr0=192.168.102.40
+hwaddr0=1c:c1:de:19:40:6e
+wwsh object delete $node -y
+wwsh node new $node --netdev=eth0 --hwaddr=$hwaddr0 --ipaddr=$ipaddr0 --netmask=255.255.0.0  --network=255.255.0.0 -y
+wwsh -y provision set $node --vnf=centos7.2 --bootstrap=`uname -r` --files=dynamic_hosts,passwd,group,shadow,slurm.conf,munge.key,network
+wwsh object modify -s bootloader=sda $node -y
+wwsh object modify -s diskpartition=sda $node -y
+wwsh object modify -s diskformat=sda1,sda2,sda3 $node -y
+wwsh object modify -s filesystems="mountpoint=/boot:dev=sda1:type=ext3:size=1024,dev=sda2:type=swap:size=6144,mountpoint=/:dev=sda3:type=ext3:size=+" $node -y
+wwsh provision set --bootlocal=UNDEF $node -y
+node=n29
+ipaddr0=192.168.102.38
+hwaddr0=1c:c1:de:1c:88:c3
+wwsh object delete $node -y
+wwsh node new $node --netdev=eth0 --hwaddr=$hwaddr0 --ipaddr=$ipaddr0 --netmask=255.255.0.0  --network=255.255.0.0 -y
+wwsh -y provision set $node --vnf=centos7.2 --bootstrap=`uname -r` --files=dynamic_hosts,passwd,group,shadow,slurm.conf,munge.key,network
+wwsh object modify -s bootloader=sda $node -y
+wwsh object modify -s diskpartition=sda $node -y
+wwsh object modify -s diskformat=sda1,sda2,sda3 $node -y
+wwsh object modify -s filesystems="mountpoint=/boot:dev=sda1:type=ext3:size=1024,dev=sda2:type=swap:size=6144,mountpoint=/:dev=sda3:type=ext3:size=+" $node -y
+wwsh provision set --bootlocal=UNDEF $node -y
+node=n29e
+ipaddr0=192.168.102.38
+hwaddr0=1c:c1:de:1c:88:c2
+wwsh object delete $node -y
+wwsh node new $node --netdev=eth0 --hwaddr=$hwaddr0 --ipaddr=$ipaddr0 --netmask=255.255.0.0  --network=255.255.0.0 -y
+wwsh -y provision set $node --vnf=centos7.2 --bootstrap=`uname -r` --files=dynamic_hosts,passwd,group,shadow,slurm.conf,munge.key,network
+wwsh object modify -s bootloader=sda $node -y
+wwsh object modify -s diskpartition=sda $node -y
+wwsh object modify -s diskformat=sda1,sda2,sda3 $node -y
+wwsh object modify -s filesystems="mountpoint=/boot:dev=sda1:type=ext3:size=1024,dev=sda2:type=swap:size=6144,mountpoint=/:dev=sda3:type=ext3:size=+" $node -y
+wwsh provision set --bootlocal=UNDEF $node -y
+wwsh pxe update
+wwsh dhcp update
+systemctl restart dhcpd
+systemctl restart httpd
+echo "after first boot: wwsh provision set --bootlocal=EXIT $node"
+</code>
+page 1 - [[cluster:155|OpenHPC page 2]] - [[cluster:156|OpenHPC page 3]] - [[cluster:160|OpenHPC page 4]]
 \\
 **[[cluster:0|Back]]**

DokuWiki

User Tools

Site Tools

Differences

Page Tools