- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - HP-UX
- >
- FileSystem Removal in a Cluster Node Pair
Categories
Company
Local Language
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Forums
Discussions
Discussions
Discussions
Forums
Discussions
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Community
Resources
Forums
Blogs
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-15-2004 03:41 PM
09-15-2004 03:41 PM
I will need to remove a filesytem /mnt/KCCAFPRD2/data04 on /dev/vg02/lvol4 from my cluster package named kcb2bprd.
This are the steps I will be performing:-
1.Stop the package kccafprd
cmhaltpkg -v kcb2bprd
2.Unmark vg02 from cluster and activate vg02
vgchange –c n vg02
vgchange –a y vg02
3. Remove logical volume /dev/vg02/lvol4
lvremove /dev/vg02/lvol4
4. Modify the /etc/cmcluster/kccafprd2/kccafprd2.cntl file in both the primary and secondary node, by removing the entry LV[3]=/dev/vg02/lvol4; FS[3]=/oracle/KCCAFPRD2/data04; FS_MOUNT_OPT[3]="-o rw,suid,largefiles"
5. I presume I need to do a vgexport of vg02 in the primary node and then a vgimport vg02 in the secondary node.
6.Deactivate vg02 and mark to cluster
vgchange –a n vg02
vgchange –c y vg02
7.Start cluster kcb2bprd
cmrunpkg –v kcb2bprd
Queries:
- Can anyone advise me on step 5 on how to perform a vgexport and vgimport?
- Can I check do I need to modify the /etc/fstab file cause I dont see any entry there for filesytem /mnt/KCCAFPRD2/data04.
- Finally did I miss out any steps needed?
Thanks in advance for the render help.
regards/cliff
Solved! Go to Solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-15-2004 04:30 PM
09-15-2004 04:30 PM
SolutionYou may not need to unmount the filesystem unless it is heavily used by the application. If there are no processes associated with that filesystem, you can remove it while the package is up and running. In case if you have to bring down the package, then I would do 'vgchange -a e vg02' instead of those two vgchange steps.
Regarding point 5, yes you will need to do a vgexport and import of vg02. The easier way is
5a. Generate the map file on the primary node.
Primary Node:
vgexport -p -v -s -m /tmp/vg02.map
Copy /tmp/vg02.map file to the alternate node
Secondary Node:
ll /dev/vg02/group (note the minor number say 0x020000)
vgexport vg02
mkdir /dev/vg02
mknod /dev/vg02/group c 64 0x020000
vgimport -v -s -m /tmp/vg02.map vg02
When you get a chance, run the package on secondary node and make sure things work. Don't forget to sync the control files. There wont' be any entry for ServiceGuarded filesystems in /etc/fstab.
-Sri
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-15-2004 04:59 PM
09-15-2004 04:59 PM
Re: FileSystem Removal in a Cluster Node Pair
Thanks for the advise.
Need to verify with u the following:-
- If I can remove the filesystem together with the logical volume while the package is up and running, there is still a need to modify the package control file in both nodes and vgexport and reimport procedure am I right?
- vgchange -a e is applicable even with the package down? By running vgchange -c n and vgchange -a y is it allowed too? If vgchange -a e is applicable for step 6 I will run vgchange -a n vg02 to deactivate it?
- U mentioned sync control files meaning to manually edit the /etc/cmcluster/kccafprd2/kccafprd2.cntl file as in step 4 on both primary and secondary node. Am I correct?
thanks n regards/cliff
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-15-2004 05:14 PM
09-15-2004 05:14 PM
Re: FileSystem Removal in a Cluster Node Pair
Yes. You can remove the logical volume if the cluster is running provided stopping the processes on that filesystem will not affect serviceguard functionality (like monitoring those processes through serviceguard etc.,). Be careful while editing the control script. Syntax errors can result the package halt to fail subsequently.
Regarding modifying the control files, you have to modify them in any case. on the same note, I copy the serviceguard files from the other node instead of editing the file on the node itself. That's what I meant by sync'ing up.
You can run vgchange -a e when the package is down. It's better to run vgchange -a n once you are done as 'vgchange -a e' may not necessarily be the activation policy in your package configuration. You can run vgchange -c n steps too but they are nothing but few extra steps.
-Sri
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-15-2004 07:52 PM
09-15-2004 07:52 PM
Re: FileSystem Removal in a Cluster Node Pair
Copying from the modified control file of primary node over to the secondary node is easier I presume.
- Check with u all I need is to remove the filesystem entry 'LV[3]=/dev/vg02/lvol4; FS[3]=/oracle/KCCAFPRD2/data04; FS_MOUNT_OPT[3]="-o rw,suid,largefiles"'
from the package control file will do? Is there any other package file I might have missed out?
- vgexport -p -v -s -m /tmp/vg02.map in primary node in preview mode -m to generate the map file, does not matter if vg is currently activated. Am I correct?
- vgexport and vgimport in the secondary node should have no concern with the vg activation since it is not activated at all. Am I right?
thanks n regards/cliff
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-15-2004 07:56 PM
09-15-2004 07:56 PM
Re: FileSystem Removal in a Cluster Node Pair
// Check with u all I need is to remove the filesystem entry 'LV[3]=/dev/vg02/lvol4; FS[3]=/oracle/KCCAFPRD2/data04; FS_MOUNT_OPT[3]="-o rw,suid,largefiles"'
from the package control file will do? Is there any other package file I might have missed out?//
Yes. That's the only place where you will specify the filesystem.
//- vgexport -p -v -s -m /tmp/vg02.map in primary node in preview mode -m to generate the map file, does not matter if vg is currently activated. Am I correct?//
Correct. It may give out errors like "vg is active" etc., but still will generate the map file.
//- vgexport and vgimport in the secondary node should have no concern with the vg activation since it is not activated at all. Am I right? //
Yes. Only import the VG. Do not activate it.
-Sri
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
09-15-2004 11:22 PM
09-15-2004 11:22 PM
Re: FileSystem Removal in a Cluster Node Pair
cheers/cliff