Filesystem getting mounted from /etc/fstab while the same is in cluster package control file...

[dattu_1]

hi guys,
i have a volume group in cluster which mounts the mountpoint from /etc/fstab and not from package control file after running cmrunpkg.
i am really not getting how its happening..when i hash the entry in /etc/fstab it doesnt get mount ..
log file says mount point not found in /etc/fstab.

[Don Wilt]

Did you mark the volume group 'exclusive'? Do a man on vgchange, 'vgchange -a e vgXX', the os will not activate the vg during startup and the filesystems won't be mounted. Then SG will have control of the vg and filesystems.

[melvyn burnard]

You need to make sure ANY file system that is part of a cluster package is NOT enabled to be mounted in /etc/fstab.
You should also check that any cluter package VG does not auto activate using /etc/lvmrc.
The activation of the vg and mount of the files system is done by the package control script, verify you have this configured correctly

[Ralph Grothe]

Mount points for cluster shared filesystems should only appear in the array FS of the cluster package that requires this resource.
If you sticked with the formatting of the package script template you could run this to check what volumes and mount points are involved for a certain package:

# grep ^LV /etc/cmcluster/some_package/some_package.cntl

[Geoff Wild]

What is in /etc/lvmrc ?

You should have (for clusters):

AUTO_VG_ACTIVATE=0

Then under "custom_vg_activation()"
put the vg's that are not in the cluster.

IE:

custom_vg_activation()
{
# e.g. /sbin/vgchange -a y -s
# parallel_vg_sync "/dev/vg00 /dev/vg01"
# parallel_vg_sync "/dev/vg02 /dev/vg03"
/sbin/vgchange -a y -s /dev/vg00
/sbin/vgchange -a y -s /dev/vg01
parallel_vg_sync "/dev/vg00 /dev/vg01"

return 0
}


That way, it doesn't matter what is in /etc/fstab....

Rgds...Geoff

[dattu_1]

Ok guys i will check and let u know on monday..Really thank u for responding immediately....

[Mridul Shrivastava]

Did you make the vg cluster aware by executing vgchange -c y

[dattu_1]

Ok guys,
My package control script attached herewith...
Also my /etc/lvmrc file is as follows :
# @(#)B11.23_LR
# /etc/lvmrc
#
# This file is sourced by /sbin/lvmrc. This file contains the flags
# AUTO_VG_ACTIVATE and RESYNC which are required by the script in /sbin/lvmrc.
# These flags must be set to valid values (see below).
#

#
# The activation of Volume Groups may be customized by setting the
# AUTO_VG_ACTIVATE flag to 0 and customizing the function
# custom_vg_activation()
#

#
# To disable automatic volume group activation,
# set AUTO_VG_ACTIVATE to 0.
#

AUTO_VG_ACTIVATE=1

#
# The variable RESYNC controls the order in which
# Volume Groups are resyncronized. Allowed values
# are:
# "PARALLEL" - resync all VGs at once.
# "SERIAL" - resync VGs one at a time.
#
# SERIAL will take longer but will have less of an
# impact on overall I/O performance.
#

RESYNC="SERIAL"
#
# Add customized volume group activation here.
# A function is available that will synchronize all
# volume groups in a list in parallel. It is
# called parallel_vg_sync.
#
# This routine is only executed if AUTO_VG_ACTIVATE
# equals 0.
#

custom_vg_activation()
{
# e.g. /sbin/vgchange -a y -s
# parallel_vg_sync "/dev/vg00 /dev/vg01"
# parallel_vg_sync "/dev/vg02 /dev/vg03"

return 0
}

#
# The following functions should require no additional customization:
#

parallel_vg_sync()
{
for VG in $*
do
{
if /sbin/vgsync $VG > /dev/null
then
echo "Resynchronized volume group $VG"
fi
} &
done
}


Yes mridul the vg is cluster aware...
regards,
Tejas

[Ralph Grothe]

Hi Tejas,

as Geoff already pointed out,
you should set
AUTO_VG_ACTIVATE=0
in /etc/lvmrc to avoid inadvertent activation of your cluster shared VG vg_mcgb_bancs_prod,
which should only be activated exclusively on package start on the node that is runnung this package.
If you have any other VGs that aren't part of the cluster but which you would like to be activated on a reboot you should list them in the function custom_activation of /etc/lvmrc like Geoff has shown.

From your attached control script I can see these mountpoints, which must not appear in /etc/fstab (maybe only commented out as a reminder).
Btw, are these the conflicting mountpoints?

$ grep ^LV /tmp/286463.cntl|cut -d\; -f2
FS[0]="/fns/p1"
FS[1]="/oradata_p1"
FS[2]="/oradata_pfone"
FS[3]="/comlink"

[dattu_1]

HI Ralph,
The mountpoints pointed by u r no longer used by us..so i have hashed it in my control file....so only setting AUTO_VG parameter to zero will solve my problem..

Regards,
Dattu.