HP-UX configuration and attachment

# cd /dev
# insf -e
insf: Installing special files for mux2 instance 0 address 8/0/0
       :           :          :         :
       :           :          :         :
#

Procedure

1. Obtain your Device Data table on which you recorded the data for the new devices. You should have the following information for all new devices:

   • Bus number
   • Bus instance number
   • Disk number
   • Driver
   • Device type
   • Target ID
   • LUN
   

2. Build the device file name for each device, and enter the device file names into your table. Example:

   File name = cXtYdZ, where X = bus instance #, Y = target ID, Z = LUN.

3. Build the minor number for each device, and enter the minor numbers into your table. Example:

   0xXXYZ00, where XX = bus instance #, Y = SCSI target ID, and Z = LUN.

4. Display the driver information for the system using the lsdev command.

   

   This sample screen shows the following system information for the “sdisk” device driver:

   • Major number of driver sdisk for character-type files: 188
   • Major number of driver sdisk for block-type files: 31

5. Enter the major numbers for the drivers into your table. You should now have all required device and driver information in the Device Data table (as shown in step 4).

6. Create the device files for all new devices (SCSI disk and raw/FX devices) using the mknod command. Be sure to create the block-type device files in the /dev/dsk directory and the character-type device files in the /dev/rdsk directory, as shown in the following example:

   

   The character-type device file is required for volumes used as raw devices (for example, 3390-3A). The block-type device file is not required for raw devices. If you need to delete a device file, use the rm -i command.

Parent Topic

• Verifying device files and the driver for HP-UX

Procedure

1. Use the pvcreate command to create the physical volume with the character-type device file as the argument. Specify the /dev/rdsk directory for the character file. You can only create one physical volume at a time.

   Example:

   # pvcreate /dev/rdsk/c2t6d0                                   
                          
   Physical volume “/dev/rdsk/c2t6d0” has been successfully created.
   # pvcreate /dev/rdsk/c2t6d1
   Physical volume “/dev/rdsk/c2t6d1” has been successfully created.
   :

   NoteDo not use the -f (force) option with the pvcreate command. This option creates a new physical volume forcibly and overwrites the existing volume.

2. Repeat step 1 for each new disk device on the storage system.

Parent Topic

• Partitioning disk devices for HP-UX

Procedure

1. Use the ls command to display the existing volume groups.

   # ls /dev
   vg00
    :
   vg05
   #

2. Use the mkdir command to create the directory for the new volume group. Choose a name for the new volume group that is different than all other group names. Do not use an existing volume group name.

   # mkdir /dev/vg06
   

   TipIf you need to delete a directory, use the rmdir command (for example, rmdir /dev/vgnn).

3. Use the ls command to verify the new directory.

   # ls /dev
   vg00
    :
   vg06
   #
   

4. Use the ll command to verify the minor numbers for existing group files with the output piped to grep to display only the files containing “group”.

   # ll /dev/vg* | grep group
   crw-rw-rw     1 root    root    64 0x000000 Nov 7 08:13 group                                     
   :
   # 

   In this example, the minor number of the existing group file is 00 (0x000000).

5. Choose a minor number for the new group file in sequential order (that is, when existing volume groups are vg00-vg05 and next group name is vg06, use minor number 06 for the vg06 group file). Do not to duplicate any minor numbers.

   The minor numbers are hexadecimal (for example, the tenth minor number is 0x0a0000, not 0x100000).

6. Use the mknod command to create the group file for the new directory. Specify the correct volume group name, major number, and minor number. The major number for all group files is 64.

   # mknod /dev/vg06/group c 64 0x060000
   :
   # 

   In this example, group name = vg06, major number of group file = 64, and minor number of new group file = 06.

   TipIf you need to delete a group file, use the rm -r command to delete the group file and the directory at the same time (for example, rm -r /dev/vgnn), and start again at step 2.

7. Repeat steps 5 and 6 for each new volume group.

8. Use the vgcreate command to create the volume group.

   # vgcreate /dev/vg06 /dev/dsk/c2t6d0                                      
                    
   Volume group "/dev/vg06" has been successfully created.
   Volume group configuration for /dev/vg06 has been saved in /etc/1vmconf/vg06.cof.
   # vgcreate -s 8 -e 15845 /dev/vg09 /dev/dsk/c2t7d0                      
               
   Volume group "/dev/vg09" has been successfully created.
   Volume Group configuration for /dev/vg09 has been saved in /etc/lvmconf/vg09.cof

   This example shows the creation of volume group vg06 with device c2t6d0 and the creation of volume group vg09 with LUSE device c2t7d0 (n = 18).

   For LUSE volumes with more than 17 OPEN-8/9 LDEVs (n > 18) or more than 7043 MB (OPEN 8/9*n-CVS), use the -s and -e physical extent (PE) parameters of vgcreate, as shown in the example above (PE Size = -s 8, MPE Size = -e 15845).

   For details about the PE and MPE parameters for LUSE devices, see PE and MPE parameters for LUSE devices for HP-UX.

   Tip

   • To allocate more than one physical volume to the new volume group, add the other physical volumes separated by a space (for example, vgcreate /dev/vg06 /dev/dsk/c0t6d0 /dev/dsk/c0t6d1).
   • If you need to delete a volume group, use the vgremove command (for example, vgremove /dev/vgnn). If the vgremove command does not work because the volume group is not active, use the vgexport command (for example, vgexport /dev/vgnn).

9. Use the vgdisplay command to verify that the volume group was created correctly. The -v option displays the detailed volume group information.

   # vgdisplay /dev/vg06                                         
   --- Volume groups ---
   VG Name             /dev/vg06
   VG Write Access     read/write
   VG Status           available
   Max LV              255
   Cur LV              0
   Open LV             0
   Max PV              16
   Cur PV              1
   Act PV              1
   Max PE per PV       1016                                  
   VGDA                2
   PE Size (Mbytes)    4                                     
   Total PE            586 
   Alloc PE            0
   Free PE             586
   Total PVG           0

   For LUSE devices, verify the values displayed for Max PE per PV and PE Size (Mbytes).

Parent Topic

• Partitioning disk devices for HP-UX

Child Topics

• PE and MPE parameters for LUSE devices for HP-UX

Procedure

1. Use the lvcreate -L command to create the logical volume, and specify the volume size and volume group for the new logical volume.

   For example, to create an OPEN-3 volume (2344 MB):

   # lvcreate -L 2344 /dev/vg06                                
                  
   Logical volume "/dev/vg06/lvol1" has been successfully created with character device "/dev/vg06/rlvol1".
   Logical volume "/dev/vg06/lvol1" has been successfully extended.
   Volume Group configuration for /dev/vg06 has been saved in /etc/1vmconf/vg06.cof.

   The HP-UX system assigns the logical volume numbers automatically (lvol1, lvol2, lvol3, …). Use the capacity values specified in Logical device types for the size parameter (for example, OPEN-3 = 2344, OPEN-V = 61432 in maximum size).

   To calculate S1 for VLL, LUSE, and VLL LUSE volumes, use the vgdisplay command to display the physical extent size (PE Size) and usable number of physical extents (Free PE) for the volume. Calculate the maximum size value (in MB) as follows:

   S1 = ( PE Size) × (Free PE)

   For example:

   # vgdisplay /dev/vg01
   
   --- Volume groups ---
   VG Name             /dev/vg01
   VG Write Access     read/write
   VG Status           available
   Max LV              255
   Cur LV              0
   Open LV             0
   Max PV              16
   Cur PV              1
   Act PV              1
   Max PE per PV       1016
   VGDA                2
   PE Size (Mbytes)    4                                        
   Total PE            586
   Alloc PE            0
   Free PE             586                                   
   Total PVG           0

   This example shows the following information for /dev/vg01:

   • Physical extent size = 4
   • Usable number of physical extents = 586

   Therefore, maximum size value = 4 × 586 = 2344

2. Use the lvdisplay command to verify that the logical volume was created correctly. If desired, wait until all logical volumes have been created, and then use the * wildcard character with the lvdisplay command to verify all volumes at one time (for example, lvdisplay /dev/vg06/lvol*).

   Example:

   # lvdisplay /dev/vg06/lvol1                                 
   --- Logical volume ---
   LV Name                 /dev/vg06/lvol1
   VG Name                 /dev/vg06
   LV Permission           read/write
   LV Status               available/syncd
   Mirror copies           0
   Consistency Recovery    MWC
   Schedule                parallel
   LV Size (Mbytes)        2344        (7040 for OPEN-9)         
   Current LE              586         (1760 for OPEN-9)              
   Allocated PE            586         (1760 for OPEN-9)             
   Stripes                 0
   Stripe Size (Kbytes)    0
   Bad block               on
   Allocation              strict

   In this example:

   • LV Size 2344 = 586 * 4 = OPEN-3
   • LE = logical extent
   • PE= physical extent

3. Repeat steps 1 and 2 for each logical volume to be created. You can only create one logical volume at a time, but you can verify more than one logical volume at a time.

   Tip

   • If you need to delete a logical volume, use the lvremove command (for example, lvremove /dev/vgnn/lvolx).
   • If you need to increase the size of an existing logical volume, use the lvextend command (for example, lvextend L size /dev/vgnn/lvolx).
   • If you need to decrease the size of an existing logical volume, use the lvreduce command (for example, lvreduce L size /dev/vgnn/lvolx).

Parent Topic

• Partitioning disk devices for HP-UX

Procedure

1. Use the pvdisplay command to verify the current IO time-out value.

   # pvdisplay /dev/dsk/c0t6d0                       
   --- Physical volumes ---
   PV Name               /dev/dsk/c0t6d0
   VG Name               /dev/vg06
   PV Status             available
   Allocatable           yes
   VGDA                  2
   Cur LV                1
   PE Size (Mbytes)      4
   Total PE              586             
   Free PE               0
   Allocated PE          586            
   Stale PE              0
   IO Timeout (Seconds)  default  

   Shown in this example:

   • Total PE: 586 indicates OPEN-3, 1760 indicates OPEN-9
   • Allocated PE: 586 indicates OPEN-3, 1760 indicates OPEN-9
   • IO Timeout: default

2. Use the pvchange -t command to change the IO time-out value to 60.

   # pvchange -t 60 /dev/dsk/c0t6d0                            
   Physical volume “/dev/dsk/c0t6d0” has been successfully changed.
   Volume Group configuration for /dev/vg06 has been saved in /etc/lvmconf/vg06.cof

3. Use the pvdisplay command to verify that the new IO time-out value is 60 seconds.

   # pvdisplay /dev/dsk/c0t6d0                             
   --- Physical volumes ---
   PV Name               /dev/dsk/c0t6d0
   VG Name               /dev/vg06
   PV Status             available
    :
   Stale PE              0
   IO Timeout (Seconds)  60

4. Repeat steps 1 through 3 for each new disk device on the storage system.

Parent Topic

• Setting device parameters for HP-UX

Procedure

1. If you cannot shut down and restart the system at this time, use the scsictl command to set the queue depth for each new device. The scsictl commands to set queue depth should be registered as HP-UX start-up script for future reboot.

   # /usr/sbin/scsictl -m queue_depth=32 -a /dev/rdsk/c0t6d0
   # /usr/sbin/scsictl -m queue_depth=32 -a /dev/rdsk/c0t6d1
   # /usr/sbin/scsictl -m queue_depth=32 -a /dev/rdsk/c0t6d2	
   # /usr/sbin/scsictl -m queue_depth=32 -a /dev/rdsk/c0t6d3
    :
    :
   # /usr/sbin/scsictl -m queue_depth=32 -a /dev/rdsk/c0t8d0

2. Check the /sbin/init.d and /sbin/rc1.d directories to see whether the script name "queue" is already used (link name Sxxxqueue or Kxxxqueue).

   # ls /sbin/init.d                                                     
   OspfMib    clean_ex   dfs        hpether    names      nis.server savecore   swconfig
   SnmpHpunix clean_tmps diagnostic iforls     ncs        pd         sendmail   syncer
    :
   clean_adm  ddfa       hparray    mrouted    nis.client rwhod      swcluster  xntpd
   
   # ls /sbin/rc1.d                                    
   K230audio       K340xntpd   K420dfs         K475rarpd       K630named      S420set_date
   K240auditing    K356vjed    K430dce         K480rdpd        K660net        S440savecore
   K250envd        K358egcd    K435OspfMib     K490gated       K700nettl      S500swap_start
   K258diagnostic  K360kks     K435SnmpHpunix  K500inetd       K770ptydaemon  S520syncer
   K270cron        K370vt      K435SnmpMib2    K510mrouted     K780syslogd
   K278pd          K380xfs     K440SnmpMaster  K570nfs.client  K900swagentd
   K280lp          K390rbootd  K450ddfa        K580nis.client  S100localmount
   K290hparray     K400iforls  K460sendmail    K590nis.server  S320hostname
   K300acct        K410ncs     K470rwhod       K600nfs.core    S400set_prvgrp

   Choose a unique name for the start-up script as follows:

   • If there is no script named "queue" and no link file named Sxxxqueue or Kxxxqueue, use the name "queue" for the new script and go to step 3.
   • If the script "queue" and the link file Sxxxqueue or Kxxxqueue exist and the script is used to set the queue depth for other previously installed storage systems, check the script file to see whether the queue depth is set to the desired number and add a line for each new disk device. If necessary, restart the HP-UX system to set the queue depth for the new volumes.
   • If the script queue and the link file Sxxxqueue or Kxxxqueue already exist and the script is not used for setting the queue depth for the storage system, use another name for the new queue-depth script for the storage system (for example, hitachi_q) and go to step 3.
     NoteIf the link Sxxxqueue and/or Kxxxqueue exists, but there is no script file named "queue", delete the link files, use the name "queue" for the new script, and go to step 3.

3. Choose a unique 3-digit number for the link name. This number cannot be used in any other links. The link name is derived as follows: S stands for “start up script,” K stands for “kill script,” the three-digit number is unique to each link, and the script file name follows the three-digit number (for example, S890queue or S890hitachi_q).

4. Create and register the new start-up script for the storage system (see Queue depth start-up script for HP-UX for an example).

5. Shut down and restart the HP-UX system, so the new start-up script sets the queue depth for the disk devices to the specified value.

6. After restarting the system or setting the queue depths manually, use the scsictl command to verify the queue depth for each disk device.

   # /usr/sbin/scsictl -a /dev/rdsk/c0t6d0                         
                                     
   immediate_report = 0; queue_depth = 32                                 
   :
   :
   # /usr/sbin/scsictl -a /dev/rdsk/c0t8d0                        
   immediate_report = 0; queue_depth = 32                                
   

Parent Topic

• Setting device parameters for HP-UX

# cp /sbin/init.d/template /sbin/init.d/queue 
# vi /sbin/init.d/queue
--------------------------file(/sbin/init.d/queue)----------------------
# !/sbin/sh
#
# @(#) $Revision: 78.1 $
# 
# NOTE:    This script is not configurable!  Any changes made to this
#          script will be overwritten when you upgrade to the next
#          release of HP-UX.
#
# WARNING: Changing this script in any way may lead to a system that
#          is unbootable.  Do not modify this script.
#
# <Insert comment about your script here>
#
# Allowed exit values:
#	0 = success; causes "OK" to show up in checklist.
#	1 = failure; causes "FAIL" to show up in checklist.
#	2 = skip; causes "N/A" to show up in the checklist.
#           Use this value if execution of this script is overridden
#	    by the use of a control variable, or if this script is not
#	    appropriate to execute for some other reason.
#       3 = reboot; causes the system to be rebooted after execution.
# Input and output:
#       stdin is redirected from /dev/null
#       stdout and stderr are redirected to the /etc/rc.log file
#       during checklist mode, or to the console in raw mode.

PATH=/usr/sbin:/usr/bin:/sbin
export PATH

# NOTE: If your script executes in run state 0 or state 1,then /usr
#      might not be available.  Do not attempt to access commands or
#      files in /usr unless your script executes in run state 2 or
#      greater.  Other file systems typically not mounted until run
#      state 2 include /var and /opt.

rval=0

# Check the exit value of a command run by this script.  If non-zero,
# the exit code is echoed to the log file and the return value of this
# script is set to indicate failure.
set_return()  {
       x=$?
       if [ $x -ne 0 ]; then
              echo "EXIT CODE: $x"
              rval=1 # script FAILed
       fi
}
# Kill the named process(es).
# $1=<search pattern for your process>

killproc()  {
       pid='ps -el | awk '( )$NF ~ /'"$1"'/) && ($4 !=mypid) && ($5 !=
mypid)  ){ print $4 }' mypid=$$ '
       if [ "X$pid" != "X" ]; then
              if kill "$pid"; then
                     echo "$1 stopped"
              else
                     rval=1
                     echo "Unable to stop $1"
              fi
       fi
}

case $1 in
'start_msg')
        # Emit a _short_ message relating to running this script with
        # the "start" argument; this message appears as part of the
        # checklist.
        echo "Setting the queue value"
        ;;
'stop_msg')
       # Emit a _short_ message relating to running this script with
       # the "stop" argument; this message appears as part of the
       # checklist.
       echo "Stopping the <specific> system"
       ;;

'start')
__________________________________________________________________
       # source the system configuration variables                |
       if [ -f /etc/rc.config ] ; then                            |
              . /etc/rc.config                                    |
       else                                                       |
              echo "ERROR: /etc/rc.config defaults file MISSING"  |
       fi                                                         |
                                                                  |
       # Check to see if this script is allowed to run...         |
       if [ "$CONTROL_VARIABLE" != 1 ]; then                      |
              rval=2                                              |
       else                                                       |
                                                                  |
       # Execute the commands to stop your system                 |
       :                                                          |
       fi                                                         |

       /usr/sbin/scsictl -m queue_depth=8 /dev/rdsk/c0t6d0
       /usr/sbin/scsictl -m queue_depth=8 /dev/rdsk/c0t6d1
       /usr/sbin/scsictl -m queue_depth=8 /dev/rdsk/c0t8d0
        :
       ;;
'stop')
       # source the system configuration variables
       if [ -f /etc/rc.config ] ; then
              . /etc/rc.config
       else
              echo "ERROR: /etc/rc.config defaults file MISSING"
       fi

       # Check to see if this script is allowed to run...
       if [ "$CONTROL_VARIABLE" != 1 ]; then
              rval=2
       else
        :
       # Execute the commands to stop your system

       fi
       ;;
*)
       echo "usage: $0 {start|stop|start_msg|stop_msg}"
       rval=1
       ;;
esac
exit $rval
--------------------------end of file(/sbin/init.d/queue)---------------------

# ls /sbin/rc1.d
K230audio       K340xntpd   K420dfs         K475rarpd       K630named      S420set_date
K240auditing    K356vjed    K430dce         K480rdpd        K660net        S440savecore
K250envd        K358egcd    K435OspfMib     K490gated       K700nettl      S500swap_start
K258diagnostic  K360kks     K435SnmpHpunix  K500inetd       K770ptydaemon  S520syncer
K270cron        K370vt      K435SnmpMib2    K510mrouted     K780syslogd
K278pd          K380xfs     K440SnmpMaster  K570nfs.client  K900swagentd
K280lp          K390rbootd  K450ddfa        K580nis.client  S100localmount
K290hparray     K400iforls  K460sendmail    K590nis.server  S320hostname
K300acct        K410ncs     K470rwhod       K600nfs.core    S400set_prvgrp

# ln -s /sbin/init.d/queue /sbin/rc1.d/S890queue

Parent Topic

• Setting device parameters for HP-UX