add new vxfs file system in cluster package

[hp-ux_2]

Hi,

I hv hp-ux 11i with hp9000 server with mc/serviceguars.
i have created new file system and now i want to add into package, tell me steps.

Regards
Anit Jha

[melvyn burnard]

Halt the package, edit your control script to add in the new file system, distribute the control script to the other nodes, make sure you have the mount point on all nodes, then start the package.
I would also suggest manually moving the package to each node to confirm it starts and stops correctly.

All in the manual

[Chris Watson]

You said that you have created a new filesystem, was this also a new lvol and vg? If so you will need to import the maps into the other nodes.

[Abhijit P.]

Hi ...

Here are the stpes to create a new VG in a cluster setup and add it to the other node.

On one node in the cluster, create the volume group.
# pvcreate /dev/dsk/____
Repeat as needed.

Create the volume group directory:
# mkdir /dev/vg07
# mknod /dev/vg07 group c 64 0x070000
# vgcreate vg07 /dev/dsk/___ ...

With the node in cluster state (cmviewcl), imprint the cluster ID on the
VG. Doing so prevents accidental VG activation and permits the VG to be
activated by the package control script:
# vgchange -c y vg07

Activate the VG (note exclusive activation mode now required):
# vgchange -a e vg07

Create logical volumes in the new VG.

Create a file system on the new logical volume if needed. Example:
# newfs -F vxfs /dev/vg07/rlvol1
# mkdir (do this on all nodes)
Repeat for each logical volume as needed.

If it's host package is up, mount the logical volumes to the file
systems:
# mount /dev/vg07/lvol1 /vg07/lvol1_mount
...

The VG and logical volumes are now ready for use.

If the VG was created with the package down, deactivate the VG:
# vgchange -a n vg07

Create a map file prior to vgimport'ing the VG on other nodes.

# vgexport -pvs -m /etc/lvmconf/map.vg07 vg07

The results of this command will produce a file of the following
format:

VGID 2c80715b3462331e (-s option add this unique VG ID)
2 lv2
3 lv3
4 lv4
1 lv1
5 lv5
\ \_ custom lvol names
\
\_ lvol numbers

Copy the map file to the other nodes. Example:
# rcp /etc/lvmconf/map.vg07 (othernode):/etc/lvmconf/map.vg07
NOTE: "othernode" is a reference to the hostname of the destination
server

8. On the other nodes in the cluster, prepare to vgimport the new VG:
# mkdir /dev/vg07
# mknod /dev/vg07/group c 64 0x0N0000
/|\
... where N is a unique number among group files
NOTE NFS restriction for minor number in step 1 above.

Import the new VG on the adoptive node:
# vgimport -vs -m /etc/lvmconf/map.vg07 vg07

To insure that future cmapplyconf operations do no uncluster the VG,
locate and edit the cluster configuration file and add the new volume
group name.

[Locating the cluster configuration file]
There is is no naming convention for the file. The SAM utility names
it /etc/cmcluster/cmclconfig.ascii. Admins sometimes call it
cluster.ascii. >If< the file cannot be found on one of the
nodes, reconstitute it with: # cmgetconf cluster.ascii

Add a reference to the cluster configuration file:

VOLUME_GROUP /dev/vg07

Copy the file to the other nodes as backup.

11. Add the VG, LVOL and mount points to the package control script that
controls the new VG.

Example lines added to package control script:

VG[7]="vg07"
and
LV[4]="/dev/vg01/lvol1"; FS[4]="/sg1"; FS_MOUNT_OPT[4]="-o rw"
LV[5]="/dev/vg01/lvol2"; FS[5]="/sg2"; FS_MOUNT_OPT[5]="-o rw"
LV[6]="/dev/vg01/lvol3"; FS[6]="/dump5"; FS_MOUNT_OPT[6]="-o rw"
LV[7]="/dev/vg01/lvol4"; FS[7]="/depot"; FS_MOUNT_OPT[7]="-o rw"
\ / /
Note consecutive incremented index value

Check the script for syntax errors:
# sh -n
Copy the updated control script to the adoptive node(s).

Ensure the modified package control script works by testing package
startup and stop when downtime is available.

To stop a currently running package:
# cmhaltpkg

To start a package on a specific node:
# cmrunpkg -n
Drop the '-n ' if the package is to be started on the
current node.

Regards,
Abhijit