----
Top Drawer DQDPYDK
DQDPYDK-P1-C1 mdcapvio10 FC cable 4
DQDPYDK-P1-C3-Port1 mdcapvio11 u8/11 10.189.41.8 etherchannel
DQDPYDK-P1-C3-Port2 mdcapvio11 v9/47 206.245.38.62
DQDPYDK-P1-C4-Port1 Open
DQDPYDK-P1-C4-Port2 mdcapvio10 v9/46 206.245.38.61 (PROBLEM ON PORT/CABLE)
DQDPYDK-T6 mdcapvio10 v9/33 10.189.41.7 etherchannel
DQDPYDK-T7 Open
----
Middle Drawer DQDPKWA
DQDPKWA-P1-C1 mdcapvio10 FC cable 2
DQDPKWA-P1-C3-Port1 mdcapvio11 u8/12 10.189.41.8 etherchannel
DQDPKWA-P1-C3-Port2 Open
DQDPKWA-P1-C4 mdcapvio11 FC cable 3
DQDPKWA-T6 mdcapvio10 v9/34 10.189.41.7 etherchannel
DQDPKWA-T7 Open
----
Bottom Drawer DQDPRNR
DQDPRNR-P1-C3-Port1 mdcapvio11 u8/12 10.189.41.8 etherchannel
DQDPRNR-P1-C3-Port2 Open
DQDPRNR-P1-C4 mdcapvio11 FC cable 1
DQDPRNR-T6 mdcapvio10 v9/34 10.189.41.7 etherchannel
DQDPRNR-T7 Open
# mdclphmc10: Establish passwordless ssh access from desktop to the DEFS HMC hscroot user
# mdcapnim01: The following script is configured for DEFS and ready to run to create the VIO servers, assuming your NIM server login ID has password-less access to the "hscroot" account on the DEFS HMC:
mdcapnim01:/home/s45x/dukeEnergy/mkviolpar.mdcapvio10.ksh
# mdcapnim01: Execute this script to create both VIO servers on the p570, this requires password-less ssh access between your current login on the NIM server and the "hscroot" login on the HMC.
# mdcapnim01: You may also create the client LPAR's at this time as well, by running the following script:
mdcapnim01:/home/s45x/dukeEnergy/mklpar.mdcapdef00.ksh
# mdcapnim01: This script requires the same password-less ssh access between your current login on the NIM server and the "hscroot" login on the HMC. This script will create 5 client LPAR's on the p570.
# mdclphmc10: If you ran the above scripts you should now see 7 total LPAR's on the p570: 2 VIO servers and 5 client LPAR's.
# mdcapnim01: REMOVING LPARs: If an error occurs and you need to remove the LPAR's and start over, using the following scripts for that purpose:
mdcapnim01:/home/s45x/dukeEnergy/rmlpar.mdcapdef00.ksh # removes all client LPARs
mdcapnim01:/home/s45x/dukeEnergy/rmlpar.mdcapvio10.ksh # removes both VIO Servers
# mdclphmc10: You may choose to perform these procedures manually if you want, the standards for creating these LPAR's are at the following URL:
http://ddceperf/css/GlobalSysAdmin/Power5/standards.shtml
# mdclphmc10: Check each VIO server to ensure that all virtual adapters on mdcapvio10 are even numbered, and all virtual adpaters on mdcapvio11 are odd numbered. The purpose of the numbering scheme is to make it easier to identify which VIO server is providing which adapter to a client LPAR.
# mdclphmc10: The internal hard drives in the MDC p570 are attached as follows:
hdisk0 U7879.001.DQDPYDK-P1-T12-L4-L0 16 Bit LVD SCSI Disk Drive (73400 MB)
hdisk1 U7879.001.DQDPYDK-P1-T12-L5-L0 16 Bit LVD SCSI Disk Drive (73400 MB)
hdisk2 U7879.001.DQDPKWA-P1-T12-L5-L0 16 Bit LVD SCSI Disk Drive (73400 MB)
hdisk3 U7879.001.DQDPRNR-P1-T12-L5-L0 16 Bit LVD SCSI Disk Drive (73400 MB)
# mdclphmc10: Modify both VIO server profiles on the HMC to include the physical adapters as detailed below:
DQDPYDK-P1-C1 mdcapvio10 Fiber Channal Adapter
DQDPYDK-P1-C4 mdcapvio10 Ethernet Adapter (dual port)
DQDPYDK-T6 mdcapvio10 Ethernet Adapter
DQDPYDK-T7 mdcapvio11 Ethernet Adapter
DQDPKWA-P1-C1 mdcapvio10 Fiber Channal Adapter
DQDPKWA-T6 mdcapvio10 Ethernet Adapter
DQDPKWA-T7 mdcapvio10 Ethernet Adapter
DQDPKWA-T12 mdcapvio10 Storage Controller
DQDPRNR-T6 mdcapvio10 Ethernet Adapter
DQDPRNR-T7 mdcapvio10 Ethernet Adapter
DQDPYDK-P1-C3 mdcapvio11 Ethernet Adapter (dual port)
DQDPKWA-P1-C3 mdcapvio11 Ethernet Adapter (dual port)
DQDPKWA-P1-C4 mdcapvio11 Fiber Channal Adapter
DQDPRNR-P1-C3 mdcapvio11 Ethernet Adapter (dual port)
DQDPRNR-P1-C4 mdcapvio11 Fiber Channal Adapter
DQDPRNR-T12 mdcapvio11 Storage Controller
# mdclphmc10: Once the VIO Server LPARs have been created you may then load operating systems on the VIO servers. This is best done by performing a remote OS load from the NIM server. Boot a VIO server LPAR to the SMS menu using the HMC generated console terminal.
# mdclphmc10: Configure the VIO Server ethernet adapters as follows:
# mdclphmc10: For the VIO server named "mdcapvio10", use the following ethernet adapter as the boot device, this will also be the MC2 network adapter:
Port 2 - IBM 2 PORT 10/100/100 U7879.001.DQDPYDK-P1-C4-T2 00096b6e71ef
# mdclphmc10: For the VIO server named "mdcapvio11", use the following ethernet adapter as the boot device, this will also be the MC2 network adapter:
Port 2 - IBM 2 PORT 10/100/100 U7879.001.DQDPYDK-P1-C3-T2 00145ec56881
# mdclphmc10: From the HMC, boot the system to SMS and define the network boot IP information (use MC2 network IP addressing).
Client IP: 206.245.38.61
Server IP: 206.245.37.100
Gateway IP: 206.245.38.1
NetMask: 255.255.255.0
# mdclphmc10: From the HMC, boot the system to SMS and define the network boot IP information (use MC2 network IP addressing).
Client IP: 206.245.38.62
Server IP: 206.245.37.100
Gateway IP: 206.245.38.1
NetMask: 255.255.255.0
# mdclphmc10: Test ping from each VIO server LPAR to ensure communication with the NIM server.
# mdcapnim01: How to create VIO server NIM objects
cd /etc/security
vi limit
fsize = -1
reboot
scp root@ddcapnim01:/tftpboot/viospot_1300.chrp.mp.ent /tftpboot
scp root@ddcapnim01:/export/mksysb/mksysb_gold_vio_1300 /export/mksysb
scp root@ddcapnim01:/export/mksysb/bosinst_data_vio_1300 /export/nim
nim -o define -t mksysb \
-a location=/export/mksysb/mksysb_gold_vio_1300 \
-a server=master mksysb_gold_vio_1300
mkdir -p /export/spot
nim -o define -t spot -a server=master \
-a location=/export/spot \
-a source=mksysb_gold_vio_1300 viospot_1300
nim -o define -t bosinst_data \
-a location=/export/nim/bosinst_data_vio_1300 \
-a server=master bosinst_data_vio_1300
# mdcapnim01: NIM Configuration
smitty nim
# mdcapnim01: Define a new network for the 206.245.38.0 network.
# mdcapnim01: Add 206.245.38.0 network as an install network.
# mdcapnim01: Add mdcapvio10-mc2 as a machine.
nim -o remove mdcapvio10-mc2
nim -o define -t standalone -a if1='38_Network mdcapvio10-mc2 0' \
-a platform=chrp -a netboot_kernel=mp -a cable_type1='N/A' mdcapvio10-mc2
# mdcapnim01: Add mdcapvio11-mc2 as a machine.
nim -o remove mdcapvio11-mc2
nim -o define -t standalone -a if1='38_Network mdcapvio11-mc2 0' \
-a platform=chrp -a netboot_kernel=mp -a cable_type1='N/A' mdcapvio11-mc2
# mdcapnim01: Add mdcapdef00-mc2 as a machine.
nim -o remove mdcapdef00-mc2
nim -o define -t standalone -a if1='38_Network mdcapdef00-mc2 0' \
-a platform=chrp -a netboot_kernel=mp -a cable_type1='N/A' mdcapdef00-mc2
# mdcapnim01: Add mdcapdef01-mc2 as a machine.
nim -o remove mdcapdef01-mc2
nim -o define -t standalone -a if1='38_Network mdcapdef01-mc2 0' \
-a platform=chrp -a netboot_kernel=mp -a cable_type1='N/A' mdcapdef01-mc2
# mdcapnim01: Add mdcapdef02-mc2 as a machine.
nim -o remove mdcapdef02-mc2
nim -o define -t standalone -a if1='38_Network mdcapdef02-mc2 0' \
-a platform=chrp -a netboot_kernel=mp -a cable_type1='N/A' mdcapdef02-mc2
# mdcapnim01: Add mdcapdef03-mc2 as a machine.
nim -o remove mdcapdef03-mc2
nim -o define -t standalone -a if1='38_Network mdcapdef03-mc2 0' \
-a platform=chrp -a netboot_kernel=mp -a cable_type1='N/A' mdcapdef03-mc2
# mdcapnim01: Add mdcapdef04-mc2 as a machine.
nim -o remove mdcapdef04-mc2
nim -o define -t standalone -a if1='38_Network mdcapdef04-mc2 0' \
-a platform=chrp -a netboot_kernel=mp -a cable_type1='N/A' mdcapdef04-mc2
# mdcapnim01: Configure the NIM server to perform a remote operating system load of the VIO server "mdcapvio10":
smitty nim
Perform NIM Software Installation and Maintenance Tasks
Install and Update Software
Install the Base Operating System on Standalone Clients
Select Target Machine - mdcapvio10-mc2
Select the installation TYPE - mksysb
Select the MKSYSB to use for the installation - mksysb_gold_vio_1300
Select the SPOT to use for the installation - viospot_1300
Install the Base Operating System on Standalone Clients
BOSINST_DATA to use during installation [bosinst_data_vio_1300]
ACCEPT new license agreements? [yes]
Initiate reboot and installation now? [no]
ACCEPT new license agreements? [yes]
# mdcapnim01: Configure the NIM server to perform a remote operating system load of the VIO server "mdcapvio11":
smitty nim
Perform NIM Software Installation and Maintenance Tasks
Install and Update Software
Install the Base Operating System on Standalone Clients
Select Target Machine - mdcapvio11-mc2
Select the installation TYPE - mksysb
Select the MKSYSB to use for the installation - mksysb_gold_vio_1300
Select the SPOT to use for the installation - viospot_1300
Install the Base Operating System on Standalone Clients
BOSINST_DATA to use during installation [bosinst_data_vio_1300]
ACCEPT new license agreements? [yes]
Initiate reboot and installation now? [no]
ACCEPT new license agreements? [yes]
# mdclphmc10: Boot each VIO Server LPAR from the configured MC2 network interface and load the VIO operating system from the NIM server. For the initial install, select an internal SCSI disk for the OS load, do not select a SAN disk at this time. After the VIO operating system is loaded, some additional drivers will have to be added to enable MPIO communications with the SAN system.
# mdcapvio10: Login to the VIO server "mdcapvio10" using the "padmin" account and accept the license agreement:
license -accept
# mdcapvio10: Then run the "oem_setup_env" command to gain "root" access to the VIO server system, and start a Korn Shell 93 shell
ksh93
# mdcapvio10: Set the root password to "password" for now.
# mdcapvio10: Change the hostname to "mdcapvio10" and modify the /etc/rc.net file to reflect this hostname change:
smitty tcpip
vi /etc/rc.net
# mdcapvio10: Setup the password-less ssh login between the NIM server and the VIO server for the "root" user.
# mdcapvio10: Begin by loading the additional MPIO drivers necessary to communicate with the SAN system.
# mdcapvio10: Login to the VIO server "mdcapvio10-mc2" as root
# mdcapvio10: Copy the XP array MPIO tar file from the NIM server, extract and install:
mkdir /tmp/xpmpio
cd /tmp/xpmpio
scp root@206.245.37.100:/home/s45x/dukeEnergy/XPMPIO1.tar .
tar -xvf ./XPMPIO1.tar
inutoc .
smitty install_all
INPUT device / directory for software [.]
Install and Update from ALL Available Software
SOFTWARE to install [all]
PREVIEW only? (install operation will NOT occur) no
COMMIT software updates? no
SAVE replaced files? yes
ACCEPT new license agreements? no
# mdcapvio10: Remove all SAN disks, fscsi, and fcs adapters and reboot the VIO server in order for the new drivers to take effect:
rmdev -Rdl fcs0
rmdev -Rdl fcs1
shutdown -Fr
# mdcapvio10: After the system reboots, login to the VIO server using the "padmin" account, then run the "oem_setup_env" command to gain "root" access to the VIO server system, and start a Korn Shell 93 shell
ksh93
# mdcapvio10: Upgrade VIO IOS to latest level
mount 206.245.37.100:/export/lpp_source/aix /mnt
su - padmin
emgr -P # identify and remove any efixes
emgr -e 'efix label'
updateios -dev /mnt/vio_1301 --accept --install
# mdcapvio10: Reboot VIO server.
shutdown -Fr
# mdcapvio10: After the system reboots, login to the VIO server using the "padmin" account, then run the "oem_setup_env" command to gain "root" access to the VIO server system, and start a Korn Shell 93 shell
ksh93
# mdcapvio10: verify WWN's and SAN connectivity with Mike wilson.
# mdcapvio10: identify each disk by PVID and allocate for mdcapvio11 and client LPAR operating system usage.
# mdcapvio10: Remove all FC attached disks
for i in $( lsdev -Cc disk -s fcp -F name | grep hdisk )
do
print -- rmdev -Rdl ${i}
rmdev -Rdl ${i}
done
# mdcapvio10: Set fast fail parameter for SCSI adapters and reconfigure FC adapters
for i in 0 1
do
chdev -l fcs${i} -a init_link=pt2pt -P
chdev -l fscsi${i} -a dyntrk=yes -P
chdev -l fscsi${i} -a fc_err_recov=fast_fail -P
done
cfgmgr -vl fcs0
cfgmgr -vl fcs1
# mdcapvio10: Change reserve policy on hdisks to "no_reserve"
for i in $( lsdev -Cc disk -s fcp -F name | grep hdisk )
do
chdev -l ${i} -a reserve_policy=no_reserve -P
chdev -l ${i} -a algorithm=round_robin -P
chdev -l ${i} -a hcheck_interval=20 -P
chdev -l ${i} -a queue_depth=16 -P
chdev -l ${i} -a pv=yes -P
done
# mdcapvio10: copy the vscsiPriority script from the NIM server to the VIO server and set the path priority for each disk.
mkdir /usr/local/scripts
scp root@206.245.37.100:/home/s45x/dukeEnergy/vscsiPriority.ksh /usr/local/scripts
/usr/local/scripts/vscsiPriority.ksh -voi 20
for i in $( lsdev -Cc disk -s fcp -F name )
do
P0=$( lspath -Al ${i} -p fscsi0 -F value -a priority )
P1=$( lspath -Al ${i} -p fscsi1 -F value -a priority )
print "${i} : fscsi0 : priority = ${P0}"
print "${i} : fscsi1 : priority = ${P1}"
done
# mdcapvio10: Allocate each SAN disk for use as VIO operating system storage and mirror from internal hard drive. Change system to boot from SAN.
extendvg rootvg hdisk1
mirrorvg rootvg hdisk1
bosboot -ad /dev/hdisk1
bootlist -m normal hdisk1 hdisk0
bootlist -m normal -o
# mdcapvio10: remove the internal SCSI disk from the rootvg and the system
unmirrorvg rootvg hdisk0
reducevg roovg hdisk0
rmdev -Rdl hdisk0
# mdcapvio10: Reconfigure ethernet adapters for etherchannel configuration, begin by unconfiguring all ethernet adapters
lscfg -l ent*
for i in 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
do
ifconfig en${i} down
ifconfig en${i} detach
ifconfig et${i} down
ifconfig et${i} detach
rmdev -Rdl ent${i}
rmdev -Rdl en${i}
rmdev -Rdl et${i}
done
ifconfig -a
lsdev -Cc adapter | grep ent
# mdcapvio10: Rediscover all ethernet adapters
cfgmgr
lsdev -Cc adapter | grep ent
lsdev -C | grep inet
lsdev -C | grep lo
lscfg -l ent*
# mdcapvio10: turn off the chksum_offload parameter on all adapters
for i in 10 9 8 7 6 5 4 3 2 1 0
do
lsattr -El ent${i} | grep chksum_offload
chdev -l ent${i} -a chksum_offload=no
chdev -l ent${i} -a chksum_offload=no -P
lsattr -El ent${i} | grep chksum_offload
done
# mdcapvio10: identify the application ethernet adapters in slot T6
lsdev -Cc adapter | grep ent
ETHADAPT='ent0,ent4,ent6'
# mdcapvio10: Create the application etherchannel adapter
if mkdev -c adapter -s pseudo -t ibm_ech \
-a adapter_names="${ETHADAPT}" \
-a netaddr=10.189.41.1
then
/usr/lib/methods/defif
fi
# mdcapvio10: identify the MC2 ethernet adapter in slot P1-C4 port 2
lsdev -Cc adapter | grep ent
ETHADAPT='ent3'
# mdcapvio10: Create the MC2 etherchannel adapter
if mkdev -c adapter -s pseudo -t ibm_ech \
-a adapter_names="${ETHADAPT}" \
-a netaddr=206.245.38.1
then
/usr/lib/methods/defif
fi
lsdev -Cc adapter | grep ent
chdev -l ent12 -a mode=round_robin
lsattr -El ent12
chdev -l ent13 -a mode=standard
lsattr -El ent13
# mdcapvio10: build the SEA's (shared ethernet adapters) Use the virtual ethernet adapters associated with slots 500 and 850.
/usr/ios/cli/ioscli mkvdev -sea ent12 -vadapter ent8 -default ent8 -defaultid 2012
/usr/ios/cli/ioscli mkvdev -sea ent13 -vadapter ent10 -default ent10 -defaultid 520
# mdcapvio10: configure IP addresses on virtual ethernet adapters associated with slots 510 and 860
smitty tcpip
# mdcapvio10: en9: inet 10.189.41.7 netmask 255.255.255.0
# mdcapvio10: en11: inet 206.245.38.61 netmask 255.255.255.0
chdev -l inet0 -a delroute='net,-hopcount,0,,0,206.245.38.1'
route add 206.245.37.0 206.245.38.1
vi /etc/rc.net
route add 206.245.37.0 206.245.38.1
# mdcapvio10: Shutdown VIO LPAR and remove the SCSI adapters from HMC profile for this LPAR.
# mdcapvio10: DQDPKWA-T12 mdcapvio10 Storage Controller
# mdcapvio10: Reboot VIO server from SAN disk.
# mdcapvio11: Login to the VIO server "mdcapvio11" using the "padmin" account and accept the license agreement:
license -accept
# mdcapvio11: Then run the "oem_setup_env" command to gain "root" access to the VIO server system, and start a Korn Shell 93 shell
ksh93
# mdcapvio11: Set the root password to "password" for now.
# mdcapvio11: Change the hostname to "mdcapvio11" and modify the /etc/rc.net file to reflect this hostname change:
smitty tcpip
vi /etc/rc.net
# mdcapvio11: Setup the password-less ssh login between the NIM server and the VIO server for the "root" user.
# mdcapvio11: Begin by loading the additional MPIO drivers necessary to communicate with the SAN system.
# mdcapvio11: Copy the XP array MPIO tar file from the NIM server, extract and install:
mkdir /tmp/xpmpio
cd /tmp/xpmpio
scp root@206.245.37.100:/home/s45x/dukeEnergy/XPMPIO1.tar .
tar -xvf ./XPMPIO1.tar
inutoc .
smitty install_all
INPUT device / directory for software [.]
Install and Update from ALL Available Software
SOFTWARE to install [all]
PREVIEW only? (install operation will NOT occur) no
COMMIT software updates? no
SAVE replaced files? yes
ACCEPT new license agreements? no
# mdcapvio11: Now remove all SAN disks, fscsi, and fcs adapters and reboot the VIO server in order for the new drivers to take effect:
rmdev -Rdl fcs0
rmdev -Rdl fcs1
shutdown -Fr
# mdcapvio11: After the system reboots, login to the VIO server using the "padmin" account, then run the "oem_setup_env" command to gain "root" access to the VIO server system, and start a Korn Shell 93 shell
ksh93
# mdcapvio11: Upgrade VIO IOS to latest level
mount 206.245.37.100:/export/lpp_source/aix /mnt
su - padmin
emgr -P # identify and remove any efixes
emgr -e 'efix label'
updateios -dev /mnt/vio_1301 --accept --install
# mdcapvio11: Reboot VIO server
shutdown -Fr
# mdcapvio11: After the system reboots, login to the VIO server using the "padmin" account, then run the "oem_setup_env" command to gain "root" access to the VIO server system, and start a Korn Shell 93 shell
ksh93
# mdcapvio11: verify WWN's and SAN connectivity with Mike wilson.
# mdcapvio11: identify hdisk to use for this VIO operating system SAN boot.
# mdcapvio11: Remove all FC attached disks
for i in $( lsdev -Cc disk -s fcp -F name | grep hdisk )
do
print -- rmdev -Rdl ${i}
rmdev -Rdl ${i}
done
# mdcapvio11: Set fast fail parameter for SCSI adapters and reconfigure FC adapters
chdev -l fscsi0 -a fc_err_recov=fast_fail
chdev -l fscsi1 -a fc_err_recov=fast_fail
cfgmgr -vl fcs0
cfgmgr -vl fcs1
# mdcapvio11: Change reserve policy on hdisks to "no_reserve"
for i in $( lsdev -Cc disk -s fcp -F name | grep hdisk )
do
chdev -l ${i} -a reserve_policy=no_reserve
chdev -l ${i} -a algorithm=round_robin
chdev -l ${i} -a hcheck_interval=20
chdev -l ${i} -a queue_depth=16
chdev -l ${i} -a pv=yes
done
# mdcapvio11: copy the vscsiPriority script from the NIM server to the VIO server and set the path priority for each disk.
mkdir /usr/local/scripts
scp root@206.245.37.100:/home/s45x/dukeEnergy/vscsiPriority.ksh /usr/local/scripts
/usr/local/scripts/vscsiPriority.ksh -voi 20
for i in $( lsdev -Cc disk -s fcp -F name )
do
P0=$( lspath -Al ${i} -p fscsi0 -F value -a priority )
P1=$( lspath -Al ${i} -p fscsi1 -F value -a priority )
print "${i} : fscsi0 : priority = ${P0}"
print "${i} : fscsi1 : priority = ${P1}"
done
# mdcapvio11: Allocate each SAN disk for use as VIO operating system storage and mirror from internal hard drive. Change system to boot from SAN.
extendvg rootvg hdisk2
mirrorvg rootvg hdisk2
bosboot -ad /dev/hdisk2
bootlist -m normal hdisk2 hdisk0
bootlist -m normal -o
# mdcapvio11: remove the internal SCSI disk from the rootvg and the system
unmirrorvg rootvg hdisk0
reducevg roovg hdisk0
rmdev -Rdl hdisk0
# mdcapvio11: Reconfigure ethernet adapters for etherchannel configuration, begin by unconfiguring all ethernet adapters
lscfg -l ent*
for i in 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
do
ifconfig en${i} down
ifconfig en${i} detach
ifconfig et${i} down
ifconfig et${i} detach
rmdev -Rdl ent${i}
rmdev -Rdl en${i}
rmdev -Rdl et${i}
done
ifconfig -a
lsdev -Cc adapter | grep ent
# mdcapvio11: Rediscover all ethernet adapters
cfgmgr
lsdev -Cc adapter | grep ent
lsdev -C | grep inet
lsdev -C | grep lo
lscfg -l ent*
# mdcapvio11: turn off the chksum_offload parameter on all adapters
for i in 10 9 8 7 6 5 4 3 2 1 0
do
lsattr -El ent${i} | grep chksum_offload
chdev -l ent${i} -a chksum_offload=no
chdev -l ent${i} -a chksum_offload=no -P
lsattr -El ent${i} | grep chksum_offload
done
# mdcapvio11: identify the application ethernet adapters in slot P1-C3
lsdev -Cc adapter | grep ent
ETHADAPT='ent0,ent2,ent4'
# mdcapvio11: Create the application etherchannel adapter
if mkdev -c adapter -s pseudo -t ibm_ech -a adapter_names="${ETHADAPT}"
then
/usr/lib/methods/defif
fi
# mdcapvio11: identify the MC2 ethernet adapter in second port of the top drawer in slot P1-C3
lsdev -Cc adapter | grep ent
ETHADAPT='ent1'
# mdcapvio11: Create the MC2 etherchannel adapter
if mkdev -c adapter -s pseudo -t ibm_ech -a adapter_names="${ETHADAPT}"
then
/usr/lib/methods/defif
fi
lsdev -Cc adapter | grep ent
chdev -l ent10 -a mode=round_robin
chdev -l ent11 -a mode=standard
lsattr -El ent10
lsattr -El ent11
# mdcapvio11: build the SEA's (shared ethernet adapters) Use the virtual ethernet adapters associated with slots 500 and 850.
/usr/ios/cli/ioscli mkvdev -sea ent10 -vadapter ent6 -default ent6 -defaultid 2012
/usr/ios/cli/ioscli mkvdev -sea ent11 -vadapter ent8 -default ent8 -defaultid 520
# mdcapvio11: configure IP addresses on virtual ethernet adapters associated with slots 510 and 860
smitty tcpip
# mdcapvio11: en7: inet 10.189.41.8 netmask 255.255.255.0
# mdcapvio11: en9: inet 206.245.38.62 netmask 255.255.255.0
chdev -l inet0 -a delroute='net,-hopcount,0,,0,206.245.38.1'
route add 206.245.37.0 206.245.38.1
vi /etc/rc.net
route add 206.245.37.0 206.245.38.1
# mdcapvio11: Shutdown VIO LPAR and remove the SCSI adapters from HMC # profile for this LPAR.
# mdcapvio11: DQDPRNR-T12 mdcapvio11 Storage Controller
# mdcapvio11: Reboot VIO server from SAN disk.
# mdcapvio10: login to VIO server as padmin and change to "oem_setup_env" and # start a Korn shell 93 shell.
oem_setup_env
ksh93
lspv
/usr/ios/cli/ioscli lsmap -all | grep vhost | head -30
vhost adapter will probably be "vhost0"
vhost adapter will probably be "vhost3"
vhost adapter will probably be "vhost6"
vhost adapter will probably be "vhost9"
vhost adapter will probably be "vhost12"
# mdcapvio10: Ensure the vhost adapters are empty, assuming the above # vhost adapter numbers.
/usr/ios/cli/ioscli rmdev -pdev vhost0
/usr/ios/cli/ioscli rmdev -pdev vhost3
/usr/ios/cli/ioscli rmdev -pdev vhost6
/usr/ios/cli/ioscli rmdev -pdev vhost9
/usr/ios/cli/ioscli rmdev -pdev vhost12
# mdcapvio10: Add the SAN boot disk to the vhost adapter for each LPAR
/usr/ios/cli/ioscli mkvdev -vdev hdisk3 -vadapter vhost0 -dev Amdcapdef00003
/usr/ios/cli/ioscli mkvdev -vdev hdisk4 -vadapter vhost3 -dev Amdcapdef01004
/usr/ios/cli/ioscli mkvdev -vdev hdisk5 -vadapter vhost6 -dev Amdcapdef02005
/usr/ios/cli/ioscli mkvdev -vdev hdisk6 -vadapter vhost9 -dev Amdcapdef03006
/usr/ios/cli/ioscli mkvdev -vdev hdisk7 -vadapter vhost12 -dev Amdcapdef04007
# mdcapvio11: login to VIO server as padmin and change to "oem_setup_env" and # start a Korn shell 93 shell.
oem_setup_env
ksh93
lspv
/usr/ios/cli/ioscli lsmap -all | grep vhost | head -30
vhost adapter will probably be "vhost0"
vhost adapter will probably be "vhost3"
vhost adapter will probably be "vhost6"
vhost adapter will probably be "vhost9"
vhost adapter will probably be "vhost12"
# mdcapvio11: Ensure the vhost adapters are empty, assuming the above vhost adapter numbers.
/usr/ios/cli/ioscli rmdev -pdev vhost0
/usr/ios/cli/ioscli rmdev -pdev vhost3
/usr/ios/cli/ioscli rmdev -pdev vhost6
/usr/ios/cli/ioscli rmdev -pdev vhost9
/usr/ios/cli/ioscli rmdev -pdev vhost12
# mdcapvio11: Add the SAN boot disk to the vhost adapter for each LPAR
/usr/ios/cli/ioscli mkvdev -vdev hdisk3 -vadapter vhost0 -dev Amdcapdef00003
/usr/ios/cli/ioscli mkvdev -vdev hdisk4 -vadapter vhost3 -dev Amdcapdef01004
/usr/ios/cli/ioscli mkvdev -vdev hdisk5 -vadapter vhost6 -dev Amdcapdef02005
/usr/ios/cli/ioscli mkvdev -vdev hdisk6 -vadapter vhost9 -dev Amdcapdef03006
/usr/ios/cli/ioscli mkvdev -vdev hdisk7 -vadapter vhost12 -dev Amdcapdef04007
# mdcapnim01: Create the client LPAR boot image on the NIM Server
scp root@ddcapnim01:/export/mksysb/mksysb_gold_5300-05-04 /export/mksysb
nim -o define -t mksysb \
-a location=/export/mksysb/mksysb_gold_5300-05-04 \
-a server=master mksysb_gold_5300-05-04
mkdir -p /export/spot
nim -o define -t spot -a server=master \
-a location=/export/spot \
-a source=mksysb_gold_5300-05-04 aixspot_gold_5300-05-04
# mdcapdef00: From the HMC, boot the system to SMS and define the network boot IP information.
Client IP: 206.245.38.63
Server IP: 206.245.37.100
Gateway IP: 206.245.38.1
NetMask: 255.255.255.0
# mdcapnim01: Define the LPAR as a machine to NIM using the node name "mdcapdef00-mc2"
# mdcapnim01: Configure the NIM server to perform a remote operating system load of the client LPAR "mdcapdef00-mc2":
smitty nim
Perform NIM Software Installation and Maintenance Tasks
Install and Update Software
Install the Base Operating System on Standalone Clients
Select Target Machine - mdcapdef00-mc2
Select the installation TYPE - mksysb
Select the MKSYSB to use for the installation - mksysb_gold_5300-05-04
Select the SPOT to use for the installation - aixspot_gold_5300-05-04
Install the Base Operating System on Standalone Clients
ACCEPT new license agreements? [yes]
Initiate reboot and installation now? [no]
ACCEPT new license agreements? [yes]
# mdcapdef00: Continue the network boot.
# mdcapdef01: From the HMC, boot the system to SMS and define the network boot IP information.
Client IP: 206.245.38.64
Server IP: 206.245.37.100
Gateway IP: 206.245.38.1
NetMask: 255.255.255.0
# mdcapnim01: Define the LPAR as a machine to NIM using the node name "mdcapdef01-mc2"
# mdcapnim01: Configure the NIM server to perform a remote operating system load of the client LPAR "mdcapdef01-mc2":
smitty nim
Perform NIM Software Installation and Maintenance Tasks
Install and Update Software
Install the Base Operating System on Standalone Clients
Select Target Machine - mdcapdef01-mc2
Select the installation TYPE - mksysb
Select the MKSYSB to use for the installation - mksysb_gold_5300-05-04
Select the SPOT to use for the installation - aixspot_gold_5300-05-04
Install the Base Operating System on Standalone Clients
ACCEPT new license agreements? [yes]
Initiate reboot and installation now? [no]
ACCEPT new license agreements? [yes]
# mdcapdef01: Continue the network boot.
# mdcapdef02: From the HMC, boot the system to SMS and define the network boot IP information.
Client IP: 206.245.38.65
Server IP: 206.245.37.100
Gateway IP: 206.245.38.1
NetMask: 255.255.255.0
# mdcapnim01: Define the LPAR as a machine to NIM using the node name "mdcapdef02-mc2"
# mdcapnim01: Configure the NIM server to perform a remote operating system load of the client LPAR "mdcapdef02-mc2":
smitty nim
Perform NIM Software Installation and Maintenance Tasks
Install and Update Software
Install the Base Operating System on Standalone Clients
Select Target Machine - mdcapdef02-mc2
Select the installation TYPE - mksysb
Select the MKSYSB to use for the installation - mksysb_gold_5300-05-04
Select the SPOT to use for the installation - aixspot_gold_5300-05-04
Install the Base Operating System on Standalone Clients
ACCEPT new license agreements? [yes]
Initiate reboot and installation now? [no]
ACCEPT new license agreements? [yes]
# mdcapdef02: Continue the network boot.
# mdcapdef03: From the HMC, boot the system to SMS and define the network boot IP information.
Client IP: 206.245.38.66
Server IP: 206.245.37.100
Gateway IP: 206.245.38.1
NetMask: 255.255.255.0
# mdcapnim01: Define the LPAR as a machine to NIM using the node name "mdcapdef03-mc2"
# mdcapnim01: Configure the NIM server to perform a remote operating system load of the client LPAR "mdcapdef03-mc2":
smitty nim
Perform NIM Software Installation and Maintenance Tasks
Install and Update Software
Install the Base Operating System on Standalone Clients
Select Target Machine - mdcapdef03-mc2
Select the installation TYPE - mksysb
Select the MKSYSB to use for the installation - mksysb_gold_5300-05-04
Select the SPOT to use for the installation - aixspot_gold_5300-05-04
Install the Base Operating System on Standalone Clients
ACCEPT new license agreements? [yes]
Initiate reboot and installation now? [no]
ACCEPT new license agreements? [yes]
# mdcapdef03: Continue the network boot.
# mdcapdef04: From the HMC, boot the system to SMS and define the network boot IP information.
Client IP: 206.245.38.67
Server IP: 206.245.37.100
Gateway IP: 206.245.38.1
NetMask: 255.255.255.0
# mdcapnim01: Define the LPAR as a machine to NIM using the node name "mdcapdef04-mc2"
# mdcapnim01: Configure the NIM server to perform a remote operating system load of the client LPAR "mdcapdef04-mc2":
smitty nim
Perform NIM Software Installation and Maintenance Tasks
Install and Update Software
Install the Base Operating System on Standalone Clients
Select Target Machine - mdcapdef04-mc2
Select the installation TYPE - mksysb
Select the MKSYSB to use for the installation - mksysb_gold_5300-05-04
Select the SPOT to use for the installation - aixspot_gold_5300-05-04
Install the Base Operating System on Standalone Clients
ACCEPT new license agreements? [yes]
Initiate reboot and installation now? [no]
ACCEPT new license agreements? [yes]
# mdcapdef04: Continue the network boot.
# mdcapdef0?: On each client LPAR: copy the vscsiPriority script from the NIM server to the VIO server and set the path priority for each disk.
mkdir /usr/local/scripts
scp root@206.245.37.100:/home/s45x/dukeEnergy/vscsiPriority.ksh /usr/local/scripts
/usr/local/scripts/vscsiPriority.ksh -voi 20
for i in $( lsdev -Cc disk -s vscsi -F name )
do
P0=$( lspath -Al ${i} -p vscsi0 -F value -a priority )
P1=$( lspath -Al ${i} -p vscsi1 -F value -a priority )
print "${i} : vscsi0 : priority = ${P0}"
print "${i} : vscsi1 : priority = ${P1}"
done
# mdcapdef0?: On each client LPAR: clean-up any leftover info from mksysb install
rmuser oradef0
rmuser oradef1
rmuser oradef2
rmuser oradef3
rmuser oradef4
rmuser oradef5
rmuser oradef6
rmuser oradef7
rmuser oradef8
rmuser oraedn1
rmuser oraedn2
rmuser shre
rmuser organ
rm -rf /home/organ
rm -rf /home/oradef
rm -rf /home/oraedn1
rm -rf /home/oraedn2
rm -rf /home/shre
rm -f /home/tibco
rm -rf /ednddtu1
rm -rf /ednddtu2
rm -rf /aacal01
rm -rf /adcal01
rm -rf /mnt1
rm -rf /mnt2
rm -rf /mnt3
rm -rf /mnt4
rm -rf /mnt5
rm -rf /nohup.out
rm -rf smit.*
rm -rf /dolvs.*
rm /usr/local/scripts/defdd*
rm /usr/local/hascripts/edn*
rm /usr/local/hascripts/start_*
rm /usr/local/hascripts/stop_*
# mdcapdef0?: On each Client LPAR: configure the node name and host name
uname -S mdcapdef00
hostname mdcapdef00
uname -S mdcapdef01
hostname mdcapdef01
uname -S mdcapdef02
hostname mdcapdef02
uname -S mdcapdef03
hostname mdcapdef03
uname -S mdcapdef04
hostname mdcapdef04
# mdcapdef0?: On each Client LPAR: Configure LPAR etherchannel devices
# mdcapdef0?: On each client LPAR: configure a static route to the mdcapnim01 network
lsattr -El inet0
chdev -l inet0 -a delroute='net,-hopcount,0,,0,206.245.38.1'
lsattr -El inet0
route add 206.245.37.0 206.245.38.1
netstat -rn
# mdcapdef0?: ADMINISTRATORS NOTE: It is usually easier to remove all
ethernet information and start from the beginning than to attempt to fix
a flawed configuration.
# mdcapdef0?: Unconfigure all ethernet adapters
lscfg -l ent*
for i in 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
do
ifconfig en${i} down
ifconfig en${i} detach
ifconfig et${i} down
ifconfig et${i} detach
rmdev -Rdl ent${i}
rmdev -Rdl en${i}
rmdev -Rdl et${i}
done
ifconfig -a
lsdev -Cc adapter | grep ent
# mdcapdef0?: Rediscover all ethernet adapters
cfgmgr
lsdev -Cc adapter | grep ent
lsdev -C | grep inet
lsdev -C | grep lo
lscfg -l ent*
# mdcapdef0?: turn off the chksum_offload parameter on all adapters
for i in 10 9 8 7 6 5 4 3 2 1 0
do
lsattr -El ent${i} | grep chksum_offload
chdev -l ent${i} -a chksum_offload=no
chdev -l ent${i} -a chksum_offload=no -P
lsattr -El ent${i} | grep chksum_offload
done
# mdcapdef0?: Even numbered hosts should use the even numbered virtual ethernet adapter as primary network path.
# mdcapdef0?: Odd numbered hosts should use the odd numbered virtual ethernet adapter as primary network path.
NETADDR="10.189.40.1"
MC2ADDR="206.245.38.1"
ZERO=0
ONE=1
TWO=2
THREE=3
# mdcapdef0?: If the hostname is odd numbered, reverse the adapter order
if [[ "$( hostname )" == *[13579] ]]
then
ZERO=1
ONE=0
TWO=3
THREE=2
fi
if mkdev -c adapter -s pseudo -t ibm_ech \
-a adapter_names="ent${ZERO}" \
-a backup_adapter="ent${ONE}" \
-a netaddr="${NETADDR}"
then
/usr/lib/methods/defif
fi
if mkdev -c adapter -s pseudo -t ibm_ech \
-a adapter_names="ent${TWO}" \
-a backup_adapter="ent${THREE}" \
-a netaddr="${MC2ADDR}"
then
/usr/lib/methods/defif
fi
lsdev -Cc adapter | grep ent
lsattr -El ent4
lsattr -El ent5
# mdcapdef0?: On each client LPAR: configure IP addresses on etherchannels
smitty tcpip
# mdcapdef0?: use the following information specific to each client LPAR
mdcapdef00-pers: en4: inet 10.189.41.10 netmask 0xffffff00
mdcapdef00-mc2: en5: inet 206.245.38.63 netmask 0xfffffe00
mdcapdef01-pers: en4: inet 10.189.41.11 netmask 0xffffff00
mdcapdef01-mc2: en5: inet 206.245.38.64 netmask 0xfffffe00
mdcapdef02-pers: en4: inet 10.189.41.12 netmask 0xffffff00
mdcapdef02-mc2: en5: inet 206.245.38.65 netmask 0xfffffe00
mdcapdef03-pers: en4: inet 10.189.41.13 netmask 0xffffff00
mdcapdef03-mc2: en5: inet 206.245.38.66 netmask 0xfffffe00
mdcapdef04-pers: en4: inet 10.189.41.14 netmask 0xffffff00
mdcapdef04-mc2: en5: inet 206.245.38.67 netmask 0xfffffe00
# mdcapdef0?: On each client LPAR: configure a static route to the mdcapnim01 network
lsattr -El inet0
chdev -l inet0 -a delroute='net,-hopcount,0,,0,206.245.38.1'
lsattr -El inet0
route add 206.245.37.0 206.245.38.1
netstat -rn
vi /etc/rc.net
/usr/sbin/route add 206.245.37.0 206.245.38.1
/bin/hostname mdcapdef0?
# mdcapdef0?: On each client LPAR: Add the following entries to the /etc/hosts file on all client LPAR's, VIO servers, and mdcapnim01
127.0.0.1 loopback localhost
206.245.37.100 mdcapnim01-mc2.def.ad mdcapnim01-mc2 mdcapnim01
206.245.38.60 mdclphmc10-mc2.def.ad mdclphmc10-mc2
206.245.38.61 mdcapvio10-mc2.def.ad mdcapvio10-mc2
206.245.38.62 mdcapvio11-mc2.def.ad mdcapvio11-mc2
206.245.38.63 mdcapdef00-mc2.def.ad mdcapdef00-mc2
206.245.38.64 mdcapdef01-mc2.def.ad mdcapdef01-mc2
206.245.38.65 mdcapdef02-mc2.def.ad mdcapdef02-mc2
206.245.38.66 mdcapdef03-mc2.def.ad mdcapdef03-mc2
206.245.38.67 mdcapdef04-mc2.def.ad mdcapdef04-mc2
10.189.41.6 mdclphmc10.def.ad mdclphmc10
10.189.41.7 mdcapvio10.def.ad mdcapvio10
10.189.41.8 mdcapvio11.def.ad mdcapvio11
10.189.41.10 mdcapdef00-pers.def.ad mdcapdef00-pers
10.189.41.11 mdcapdef01-pers.def.ad mdcapdef01-pers
10.189.41.12 mdcapdef02-pers.def.ad mdcapdef02-pers
10.189.41.13 mdcapdef03-pers.def.ad mdcapdef03-pers
10.189.41.14 mdcapdef04-pers.def.ad mdcapdef04-pers
10.189.41.20 mdcapdef00.def.ad mdcapdef00
10.189.41.21 mdcapdef01.def.ad mdcapdef01
10.189.41.22 mdcapdef02.def.ad mdcapdef02
10.189.41.23 mdcapdef03.def.ad mdcapdef03
10.189.41.24 mdcapdef04.def.ad mdcapdef04
10.189.41.30 mdcapdef00-rg01.def.ad mdcapdef00-rg01
10.189.41.31 mdcapdef01-rg01.def.ad mdcapdef01-rg01
10.189.41.32 mdcapdef02-rg01.def.ad mdcapdef02-rg01
10.189.41.33 mdcapdef03-rg01.def.ad mdcapdef03-rg01
10.189.41.34 mdcapdef04-rg01.def.ad mdcapdef04-rg01
# mdcapdef0?: Configure groups
# mdcapdef0?: On each client LPAR: Configure users and set passwords
mkdir /usr/local/bin
scp root@mdcapnim01:/usr/local/bin/setpwd /usr/local/bin
UID=$( echo avanduyn | sum -r )
UID="${UID//[!0-9]/}"
mkuser id=${UID} gecos="Al Vanduyn|CGE|816.459.3059|Alvin.VanDuyne@capgemini.com" avanduyn
/usr/local/bin/setpwd -u avanduyn -p 'yrXB8jKX'
grep avanduyn /etc/passwd
UID=$( echo skuditin | sum -r )
UID="${UID//[!0-9]/}"
mkuser id=${UID} gecos="Sanjay Kuditini|CGE|816.347.7602|sanjay.kuditini@capgemini.com" skuditin
/usr/local/bin/setpwd -u skuditin -p 'mixM5jm='
grep skuditin /etc/passwd
UID=$( echo avonnied | sum -r )
UID="${UID//[!0-9]/}"
mkuser id=${UID} gecos="Adam vonNieda|CGE|816.347.7941|adam.vonnieda@capgemini.com" avonnied
/usr/local/bin/setpwd -u avonnied -p 'uA@Lj44&'
grep avonnied /etc/passwd
mkgroup -A id=369751 dbaora
mkgroup -A id=574551 defs
mkuser id=214 gecos="People Soft Account" groups='staff,defs' psoft
chuser fsize='-1' psoft
chuser data='-1' psoft
chuser stack='-1' psoft
/usr/local/bin/setpwd -u psoft -p password
id psoft
mkuser id=3553 gecos="Oracle DB Account" groups='dbaora,defs' oracle
chuser fsize='-1' oracle
chuser data='-1' oracle
chuser stack='-1' oracle
/usr/local/bin/setpwd -u oracle -p password
id oracle
mkgroup -A id=369751 dbaora
chuser pgrp='dbaora' groups='dbaora,defs' psoft
chuser pgrp='dbaora' groups='dbaora,defs' oracle
egrep '^dbaora:|^defs:' /etc/group
# mdcapdef0?: On each client LPAR: Configure performance tools/nfs directory
smitty mknfsmnt
Add a File System for Mounting
Pathname of mount point [/prfpmce0]
Pathname of remote directory [/prfpmce0]
Host where remote directory resides [206.245.37.100]
Security method [sys]
Mount now, add entry to /etc/filesystems or both? both
/etc/filesystems entry will mount the directory yes
on system restart.
Mode for this NFS file system read-write
Attempt mount in background or foreground background
Mount file system soft or hard hard
rmfs /prfdmce0
rmfs /prfpmce0
/usr/sbin/mknfsmnt -f /prfpmce0 -d /prfpmce0 -h 206.245.37.100 -M sys \
-B -A -t rw -w bg -Y -Z -X -H -j -q -g
lsfs /prfpmce0
df
mkdir -p /prfpmce1/mdcapdef00/Top
mkdir -p /prfpmce0/mdcapdef00/wlm
mkdir -p /prfpmce0/mdcapdef01/Top
mkdir -p /prfpmce0/mdcapdef01/wlm
mkdir -p /prfpmce0/mdcapdef02/Top
mkdir -p /prfpmce0/mdcapdef02/wlm
mkdir -p /prfpmce0/mdcapdef03/Top
mkdir -p /prfpmce0/mdcapdef03/wlm
mkdir -p /prfpmce0/mdcapdef04/Top
mkdir -p /prfpmce0/mdcapdef04/wlm
# mdcapdef0?: On each client LPAR: Configure NFS installation NFS directory
mount 206.245.37.100:/export/defs /mnt
# mdcapdef0?: Request application storage from SAN Group - Mike Wilson
# mdcapvio??: Configure application storage from SAN on both VIO servers
# mdcapvio10: discover the application SAN storage disks
cfgmgr
# mdcapvio10: Change reserve policy on hdisks to "no_reserve"
for i in $( lsdev -Cc disk -s fcp -F name | grep hdisk )
do
chdev -l ${i} -a pv=yes
chdev -l ${i} -a reserve_policy=no_reserve
chdev -l ${i} -a algorithm=round_robin
chdev -l ${i} -a hcheck_interval=20
chdev -l ${i} -a queue_depth=16
done
for i in $( lsdev -Cc disk -s fcp -F name | grep hdisk )
do
lsattr -El ${i} -a reserve_policy -F name=value
lsattr -El ${i} -a algorithm -F name=value
lsattr -El ${i} -a hcheck_interval -F name=value
lsattr -El ${i} -a queue_depth -F name=value
done
# mdcapvio10: set the path priority for each disk.
/usr/local/scripts/vscsiPriority.ksh -vi 20
/usr/local/scripts/vscsiPriority.ksh -voi 20
for i in $( lsdev -Cc disk -s fcp -F name )
do
P0=$( lspath -Al ${i} -p fscsi0 -F value -a priority )
P1=$( lspath -Al ${i} -p fscsi1 -F value -a priority )
print "${i} : fscsi0 : priority = ${P0}"
print "${i} : fscsi1 : priority = ${P1}"
done
# mdcapvio10: copy the storage allocation script from the NIM server to mdcapvio10 and execute it on mdcapvio10.
# mdcapvio10: You will need to EDIT mkvdev.mdcapvio10.ksh to have the correct hdisk numbers
# mdcapvio10: allocate the hdisks to the vhost adapters using a prewritten script
scp root@mdcapnim01:/home/s45x/dukeEnergy/mkvdev.mdcapvio10.ksh /home/padmin
cd /home/padmin
vi ./mkvdev.mdcapvio10.ksh # EDIT hdisk numbers
./mkvdev.mdcapvio10.ksh
# mdcapvio10: reboot the VIO server so that path priority settings take effect
# mdcapvio11: discover the application SAN storage disks
cfgmgr
# mdcapvio11: Change reserve policy on hdisks to "no_reserve"
for i in $( lsdev -Cc disk -s fcp -F name | grep hdisk )
do
chdev -l ${i} -a reserve_policy=no_reserve
chdev -l ${i} -a algorithm=round_robin
chdev -l ${i} -a hcheck_interval=20
chdev -l ${i} -a queue_depth=16
chdev -l ${i} -a pv=yes
done
# mdcapvio11: set the path priority for each disk.
/usr/local/scripts/vscsiPriority.ksh -vi 20
/usr/local/scripts/vscsiPriority.ksh -voi 20
for i in $( lsdev -Cc disk -s fcp -F name )
do
P0=$( lspath -Al ${i} -p fscsi0 -F value -a priority )
P1=$( lspath -Al ${i} -p fscsi1 -F value -a priority )
print "${i} : fscsi0 : priority = ${P0}"
print "${i} : fscsi1 : priority = ${P1}"
done
# mdcapvio11: copy the storage allocation script from the NIM server to mdcapvio11 and execute it on mdcapvio11.
# mdcapvio11: You will need to EDIT mkvdev.mdcapvio11.ksh to have the correct hdisk numbers
# mdcapvio11: allocate the hdisks to the vhost adapters using a prewritten script
scp root@mdcapnim01:/home/s45x/dukeEnergy/mkvdev.mdcapvio11.ksh /home/padmin
cd /home/padmin
vi ./mkvdev.mdcapvio11.ksh # EDIT hdisk numbers
./mkvdev.mdcapvio11.ksh
# mdcapvio11: reboot the VIO server so that path priority settings take effect
shutdown -Fr
# mdcapdef0?: On each client LPAR: configure application storage on LPAR's
# mdcapdef0?: On each client LPAR: Configure path priorities for virtual disks
# mdcapnim01: copy the vscsiPriority script from the NIM server to each client LPAR and set the path priority for each disk.
if cd /home/s45x/dukeEnergy
then
for i in 0 1 2 3 4
do
ssh -x root@mdcapdef0${i}-mc2 'mkdir -p /usr/local/scripts'
scp ./vscsiPriority.ksh root@mdcapdef0${i}-mc2:/usr/local/scripts
ssh -x mdcapdef0${i}-mc2 '/usr/local/scripts/vscsiPriority.ksh -voi 20'
done
for i in 0 1 2 3 4
do
print -- "# mdcapdef0${i}-mc2"
ssh -x mdcapdef0${i}-mc2 'for i in $( lsdev -Cc disk -s vscsi -F name )
do
P0=$( lspath -Al ${i} -p vscsi0 -F value -a priority )
P1=$( lspath -Al ${i} -p vscsi1 -F value -a priority )
print "${i} : vscsi0 : priority = ${P0}"
print "${i} : vscsi1 : priority = ${P1}"
done'
print
done
fi
# mdcapnim01: Configure VG, LVs, LG, FSs (attach dolvs.defpdcg?.sh)
# mdcapdef0?: # You may create the Volume groups, logical volumes, log volumes, and filesystems manually, or you can use the provided scripts:
# mdcapnim01: transfer the "dolvs" script to each client LPAR and execute it to create the VGs, LVs, LGs, and FSs
if cd /home/s45x/dukeEnergy
then
for i in 0 1 2 3 4
do
ssh -x root@mdcapdef0${i}-mc2 "mkdir -p /usr/local/scripts"
scp ./dolvs.defpdcg${i}.sh mdcapdef0${i}-mc2:/usr/local/scripts
ssh -x mdcapdef0${i}-mc2 "/usr/local/scripts/dolvs.defpdcg${i}.sh"
# mdcapdef00: dolvs.defpdcg0.sh should be run on mdcapdef00
# mdcapdef01: dolvs.defpdcg1.sh should be run on mdcapdef01
# mdcapdef02: dolvs.defpdcg2.sh should be run on mdcapdef02
# mdcapdef03: dolvs.defpdcg3.sh should be run on mdcapdef03
# mdcapdef04: dolvs.defpdcg4.sh should be run on mdcapdef04
ssh -x mdcapdef0${i}-mc2 "chown -R psoft:defs /defddcg*"
ssh -x mdcapdef0${i}-mc2 "chmod -R 775 /defddcg*"
ssh -x mdcapdef0${i}-mc2 "lsuser psoft"
ssh -x mdcapdef0${i}-mc2 "lsuser oracle"
ssh -x mdcapdef0${i}-mc2 "lsgroup defs"
ssh -x mdcapdef0${i}-mc2 "find /defddcg* -exec ls -ld {} \;"
ssh -x mdcapdef0${i}-mc2 "rm /etc/ora*"
ssh -x mdcapdef0${i}-mc2 "touch /etc/oratab"
ssh -x mdcapdef0${i}-mc2 "chown oracle:defs /etc/oratab"
ssh -x mdcapdef0${i}-mc2 "chmod 775 /etc/oratab"
ssh -x mdcapdef0${i}-mc2 "ls -l /etc/oratab"
done
fi
# mdcapnim01: Configure application server start/stop scripts
DD_HASCR="/usr/local/hascripts"
DD_LOC="/home/s45x/dukeEnergy"
if cd "${DD_LOC}"
then
for i in 0 1 2 3 4
do
print -- "#!/usr/bin/ksh93
################################################################
ifconfig en4 10.189.41.2${i} netmask 255.255.255.0 alias
ifconfig en4 10.189.41.3${i} netmask 255.255.255.0 alias
varyonvg defpdcg${i}01vg
mount /defpdcg${i}
init c
cd /defpdcg${i}/mflmf
/defpdcg${i}/mflmf/mflmman
exit 0" > ${DD_LOC}/defpdcg${i}_appl01_start.sh
print -- "#!/usr/bin/ksh93
################################################################
ifconfig en4 10.189.41.2${i} netmask 255.255.255.0 delete
ifconfig en4 10.189.41.3${i} netmask 255.255.255.0 delete
cd /defpdcg${i}/mflmf
/defpdcg${i}/mflmf/lmfgetpv k
umount -f /defpdcg${i}
varyoffvg defpdcg${i}01vg
exit 0" > ${DD_LOC}/defpdcg${i}_appl01_stop.sh
chmod 755 ${DD_LOC}/defpdcg${i}_appl01_start.sh
ls -l ${DD_LOC}/defpdcg${i}_appl01_start.sh
chmod 755 ${DD_LOC}/defpdcg${i}_appl01_stop.sh
ls -l ${DD_LOC}/defpdcg${i}_appl01_stop.sh
ssh -x root@mdcapdef0${i}-mc2 "mkdir -p ${DD_HASCR}"
scp ${DD_LOC}/defpdcg${i}_appl01_start.sh root@mdcapdef0${i}-mc2:${DD_HASCR}
scp ${DD_LOC}/defpdcg${i}_appl01_stop.sh root@mdcapdef0${i}-mc2:${DD_HASCR}
scp ${DD_LOC}/dolvs.defpdcg${i}.sh root@mdcapdef0${i}-mc2:${DD_HASCR}
ssh -x root@mdcapdef0${i}-mc2 "ls -alrt ${DD_HASCR}; echo"
print -- "####"
done
fi
# mdcapnim01: Configure PLM control of LPARs
cd /etc/plm/setup
for i in 0 1 2 3 4
do
print -- ./plmsetup mdcapdef0${i}-mc2.def.ad root
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.HostPublic
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.LPAR
print -- ./plmsetup mdcapdef0${i}-mc2.def.ad root
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.HostPublic
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.LPAR
print -- ./plmsetup mdcapdef0${i}-mc2.def.ad root
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.HostPublic
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.LPAR
print -- ./plmsetup mdcapdef0${i}-mc2.def.ad root
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.HostPublic
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.LPAR
print -- ./plmsetup mdcapdef0${i}-mc2.def.ad root
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.HostPublic
CT_CONTACT=mdcapdef0${i}-mc2 lsrsrc IBM.LPAR
done
# mdcapdef0?: Miscellaneous Tasks to perform on each client LPAR
# mdcapdef0?: On each client LPAR: Add directory exclusions for mksysb backups to /etc/exclude.rootvg.
print '^./tmp/\n./usr/websm/pc_client/\n./usr/sys/inst.images' > /etc/exclude.rootvg
# mdcapdef0?: On each client LPAR: Set ulimits to unlimited as follows:
UNAMES[0]='root'
UNAMES[1]='oracle'
UNAMES[2]='psoft'
for UNAME in "${UNAMES[@]}"
do
chuser rss_hard=-1 ${UNAME}
chuser stack_hard=-1 ${UNAME}
chuser fsize=-1 ${UNAME}
chuser cpu=-1 ${UNAME}
chuser data=-1 ${UNAME}
chuser rss=-1 ${UNAME}
chuser stack=-1 ${UNAME}
lsuser -a rss_hard stack_hard fsize cpu data rss stack ${UNAME}
done
# mdcapdef0?: On each client LPAR: AIX tuning parameters
chdev -l sys0 -a maxuproc=4096
vmo -po maxclient%=45
vmo -po maxperm%=45
vmo -po minperm%=3
vmo -po lru_file_repage=0
vmo -po strict_maxclient=1
vmo -po strict_maxperm=0
vmo -po maxfree=10880
vmo -po minfree=9600
vmo -po v_pinshm=1
vmo -ro memory_affinity=0
vmo -o maxclient%
vmo -o maxperm%
vmo -o minperm%
vmo -o lru_file_repage
vmo -o strict_maxclient
vmo -o strict_maxperm
vmo -o minfree
vmo -o maxfree
vmo -o v_pinshm
vmo -o memory_affinity
# mdcapdef0?: On each client LPAR: Asynchronous I/O and System Tuning Parameters
chdev -l sys0 -a maxuproc=3000 -P
chdev -l aio0 -a maxreqs=16386 -P
chdev -l aio0 -a maxservers=300 -P
chdev -l aio0 -a minservers=150 -P
chdev -l aio0 -a autoconfig=available -P
mkdev -l aio0
lsdev -l aio0
lsattr -El aio0
# mdcapdef0?: On each client LPAR: Verify settings on each DEFS client LPAR
ulimit -a
lsps -a
lsdev -l aio0
lsattr -El aio0
for i in $( lsdev -Cc disk -F name )
do
print "${i}: $( lsattr -El ${i} | grep queue)"
done
lsattr -El sys0 -a maxuproc
vmo -o maxclient%
vmo -o maxperm%
vmo -o minperm%
vmo -o lru_file_repage
vmo -o strict_maxclient
vmo -o strict_maxperm
vmo -o minfree
vmo -o maxfree
vmo -o memory_affinity
# mdcapdef0?: On each client LPAR: reboot each system to insure all changes take effect
shutdown -Fr
# mdcapvio10: create password-less ssh login to mdclphmc10-mc2
# mdcapvio10: Copy the "mapluns.sh" script from the NIM server and execute it to create a DISKMAP of the LUNs configured by this VIO server:
scp root@206.245.37.100:/home/s45x/dukeEnergy/mapluns.sh /home/padmin
cd /home/padmin
./mapluns.sh -v \
-h mdclphmc10-mc2 \
-u hscroot \
-s Server-9117-570-SN1013D6E 2>&1 | tee DISKMAP.$( date +"%Y%m%d" )
# mdcapvio11: create password-less ssh login to mdclphmc10-mc2
# mdcapvio11: Copy the "mapluns.sh" script from the NIM server and execute it to create a DISKMAP of the LUNs configured by this VIO server:
scp root@206.245.37.100:/home/s45x/dukeEnergy/mapluns.sh /home/padmin
cd /home/padmin
./mapluns.sh -v \
-h mdclphmc10-mc2 \
-u hscroot \
-s Server-9117-570-SN1013D6E 2>&1 | tee DISKMAP.$( date +"%Y%m%d" )
# mdcapnim01: export the mksysb storage directory
exportfs -i -o 'root=*' /export/mksysb
# mdcapdef0?: mount the mksysb NFS directory and backup the system
umount /mnt
mount 206.245.37.100:/export/mksysb /mnt
ls -l /mnt/mksysb_[md]dcapdef0[0-9]_*
mksysb -Xive /mnt/mksysb_$( hostname )_$( oslevel -s )_$( date +"%y%m%d" )
ls -l /mnt/mksysb_[md]dcapdef0[0-9]_*
df
umount /mnt
mount 206.245.37.100:/export/defs /mnt
df
# mdcapnim01: export the mksysb storage directory
exportfs -i -o 'root=*' /export/mksysb
# mdcapvio??: mount the mksysb NFS directory and backup the system
umount /mnt
mount 206.245.37.100:/export/mksysb /mnt
ls -l /mnt/backupios_*
su - padmin
OUTFILE="/mnt/backupios"
OUTFILE="${OUTFILE}_$( hostname )"
OUTFILE="${OUTFILE}_$( ioslevel )"
OUTFILE="${OUTFILE}_$( date +'%y%m%d' )"
backupios -file "${OUTFILE}" -mksysb
unset OUTFILE
exit # exit from the "padmin" shell
ls -l /mnt/backupios_*
df
umount /mnt