issues with sdd driver

[Patricio Besteiro]

Hi Sysadmins!

I'm having issues with a server connected to a SVC storage box, its an 11.11 using sdd driver. At boot time is throwing this messages:

/sbin/bcheckrc:
Checking for LVM volume groups and Activating (if any exist)
Volume group "/dev/vg00" has been successfully changed.
vgchange: Warning: Couldn't attach to the volume group physical volume "/dev/dsk/vpath51":
A component of the path of the physical volume does not exist.
vgchange: Warning: couldn't query physical volume "/dev/dsk/vpath51":
The specified path does not correspond to physical volume attached to
this volume group
vgchange: Warning: couldn't query all of the physical volumes.
vgchange: Couldn't activate volume group "/dev/vgsap":
Quorum not present, or some physical volume(s) are missing.

vgchange: Warning: Couldn't attach to the volume group physical volume "/dev/dsk/vpath137":
A component of the path of the physical volume does not exist.

vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.
vgchange: A component of the path of the physical volume does not exist.

....

t

and goes on for the rest of the vpaths, looks like the driver its not yet loaded when the system issues the bcheckrc script...

there's an rc script that loads the sdd driver:

----------------------------
       * All configured modules are registered successfully.
       * Module atdd is loaded successfully.
       * Module acdd is loaded successfully.

NOTE:    Module status

Name            ID      Status          Type
=====================================================
krm             1       UNLOADED        WSIO
rng             2       UNLOADED        WSIO
atdd            3       LOADED          WSIO
acdd            4       LOADED          WSIO
vpath           5       UNLOADED        WSIO

Creating /dev/random and /dev/urandom special files
Output from "/sbin/rc1.d/S114krng start":
----------------------------

Starting vpath loading and vpath_config
Output from "/sbin/rc1.d/S200CONFserver start":

[root@usatch06] /root > cat /sbin/rc1.d/S200CONFserver
#!/sbin/sh
#
# @(#) $Revision: 82.1 $
#
# 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.
#    4 = background; causes "BG" to show up in the checklist.
#           Use this value if this script starts a process in background mode.

# 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.


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 "Starting vpath loading and vpath_config"
    ;;

'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 vpath and unloading vpath_config"
    ;;

'start')


    # Execute the commands to start your subsystem
isia64=`uname -a | grep ia64 | wc -l`
isv2pa64=`uname -a | grep 9000 | grep 11.23 | wc -l`
isautovgon=`grep -v '^#' /etc/lvmrc |grep  'AUTO_VG_ACTIVATE[^:blank:]*=[^:blank:]*1' | wc -l`
if (test $isia64 -eq 0 && test $isv2pa64 -eq 0)
then
    kmadmin -L vpath > /dev/null 2>&1
else
    kcmodule vpath=loaded
fi
        if [ -x /sbin/cfgvpath ]
        then
                /sbin/cfgvpath -debug > /var/adm/IBMsdd/cfgvpath.log
                logger "cfgvpath done"
        fi
    if (test $isautovgon -ne 0)
    then
        logger "AUTO_VG_ACTIVATE is on"
            if [ -x /opt/IBMsdd/bin/hd2vp ]
            then
            cat /var/adm/IBMsdd/hd2vp.errlog >> /var/adm/IBMsdd/hd2vp_bak.errlog
            rm /var/adm/IBMsdd/hd2vp.errlog
            if [ -r  /var/adm/IBMsdd/hd2vp.log ]
            then
                cp  /var/adm/IBMsdd/hd2vp.log  /var/adm/IBMsdd/hd2vp_bak.log
            fi
                       /opt/IBMsdd/bin/hd2vp > /var/adm/IBMsdd/hd2vp.log
            fi
    fi
#
#       remove vgchange, it causes slow boot with MC Service Guard VGs
#       The control of active VG should be specified in /etc/lvmrc
#       see AUTO_VG_ACTIVATE and custom_vg_activation()
#    /sbin/vgchange -a y -s
    ;;

'stop')
    # source the system configuration variables
    # Execute the commands to stop your subsystem
    # remove conversion to avoid long shutdown and reboot time
    if [ -x /opt/IBMsdd/bin/vp2hd ]
    then
        cat /var/adm/IBMsdd/vp2hd.errlog >> /var/adm/IBMsdd/vp2hd_bak.errlog
        rm /var/adm/IBMsdd/vp2hd.errlog
        if [ -r /var/adm/IBMsdd/vp2hd.log ]
        then
        cp /var/adm/IBMsdd/vp2hd.log /var/adm/IBMsdd/vp2hd_bak.log
        fi
#REMOVED    rm /var/adm/IBMsdd/vp2hd.log
#REMOVED     /opt/IBMsdd/bin/vp2hd > /var/adm/vp2hd.log
    fi
    ;;

*)
    echo "usage: $0 {start|stop|start_msg|stop_msg}"
    rval=1
    ;;
esac

exit $rval

so my understanding is that the driver its being loaded at rc scripts time, so bcheckrc doesn't know what a vpath is and it givesup trying to activate the VGs...

there are 2 lines commented on the last version of sdd driver at the "stop" option:

#REMOVED    rm /var/adm/IBMsdd/vp2hd.log
#REMOVED     /opt/IBMsdd/bin/vp2hd > /var/adm/vp2hd.log

that script exchanges the vpath devices to c*t*d* devices on lvmtab, on the manual says it was removed because it takes to long to run at server shutdown.

should I add  "kmadmin -L vpath > /dev/null 2>&1" to bcheckrc as workaround?

hope I find an admin that got this kind of issues in the past.

thanks in advance!