Veritas Disk Expansion
Notes for disk expansion when storage may be managed by Veritas.
Important: do not run these commands blindly. Veritas disk expansion can affect clustered applications, disk groups, volumes, mounts, filesystems, and service groups.
Strong warning
This page is not a production procedure.
Veritas disk expansion depends on the exact environment, disk group, volume layout, filesystem, cluster ownership, active node, and company procedure.
Use this page only for:
checking
learning
asking better questions
documenting verified internal procedures
Do not use this page as a blind command recipe.
Only run expansion commands after confirming:
1. Correct server
2. Correct active node
3. Correct service group
4. Correct disk group
5. Correct volume
6. Correct filesystem
7. Correct mount point
8. Correct approved size
9. Correct rollback or recovery plan
10. Approval from the responsible team
Safety rule
Before doing anything with Veritas storage, check:
1. Am I on the correct server?
2. Is this production or test?
3. Is the system clustered?
4. Which service group owns the disk, mount, or application?
5. Which node is active?
6. Is the service group online, offline, partial, or faulted?
7. Is the disk group imported?
8. Is the filesystem mounted?
9. Is the volume managed by Veritas?
10. Do I have approval and a rollback plan?
If unsure, stop and ask.
First cluster checks
Check cluster summary
hastatus -sum
Check service groups
hagrp -state
Check resources
hares -state
Check one service group
hagrp -state SERVICE_GROUP
hagrp -display SERVICE_GROUP
Check if service group is frozen
hagrp -display SERVICE_GROUP | grep -i frozen
First OS/storage checks
hostnamectl
df -h
df -Th
findmnt
lsblk
lsblk -f
blkid
Veritas disk group checks
Show disk groups
vxdg list
Show one disk group
vxdg list DISK_GROUP
Show disks
vxdisk list
Rescan Veritas disks
Use only if storage was added or changed and this is part of the approved procedure.
vxdctl enable
vxdisk scandisks
vxdisk list
Show Veritas volumes and objects
vxprint -ht
For a specific disk group:
vxprint -g DISK_GROUP -ht
Check volume and filesystem
Check mounted filesystem
df -h /mount/point
df -Th /mount/point
findmnt /mount/point
Check Veritas volume
vxprint -g DISK_GROUP -ht
Look for the volume name connected to the mount or application.
Example placeholders:
Disk group: dg_example
Volume: vol_example
Mount: /data
Typical high-level flow
This is a general flow only.
1. Confirm affected filesystem.
2. Confirm whether it is Veritas-managed.
3. Confirm disk group.
4. Confirm volume.
5. Confirm service group and active node.
6. Confirm if free space already exists in the disk group.
7. Confirm if new storage/LUN was added instead.
8. Extend Veritas volume using approved procedure.
9. Grow filesystem if required.
10. Verify filesystem size.
11. Verify application/service group health.
12. Check logs.
If free space already exists in the Veritas disk group
First verify available space and layout:
vxdg list DISK_GROUP
vxprint -g DISK_GROUP -ht
df -h /mount/point
df -Th /mount/point
Then use the approved internal procedure to grow the volume.
Generic command patterns may look like:
vxassist -g DISK_GROUP growto VOLUME_NAME NEW_SIZE
or:
vxassist -g DISK_GROUP growby VOLUME_NAME SIZE_TO_ADD
Example with placeholders:
vxassist -g dg_example growby vol_example 10g
Do not use this without verifying:
correct disk group
correct volume
correct current size
correct target size
correct filesystem
correct active node
approved change window
Grow filesystem after volume expansion
After the Veritas volume is expanded, the filesystem may still need to be grown.
Check filesystem type:
df -Th /mount/point
Grow XFS filesystem
For XFS:
xfs_growfs /mount/point
Example:
xfs_growfs /data
Grow ext4 filesystem
For ext4:
resize2fs /dev/vx/dsk/DISK_GROUP/VOLUME_NAME
Example:
resize2fs /dev/vx/dsk/dg_example/vol_example
Verify:
df -h /mount/point
If new storage or LUN was added
General checking commands:
lsblk
vxdisk list
vxdctl enable
vxdisk scandisks
vxdisk list
Then follow approved procedure to initialize or add the new disk to the correct disk group.
Generic command pattern:
vxdg -g DISK_GROUP adddisk DISK_NAME=DEVICE_NAME
Do not run this without confirming the exact disk and disk group.
Danger: adding the wrong disk can cause data loss or storage problems.
Verify after expansion
df -h /mount/point
df -Th /mount/point
findmnt /mount/point
vxprint -g DISK_GROUP -ht
vxdg list DISK_GROUP
hastatus -sum
hagrp -state SERVICE_GROUP
hares -state
Also check logs:
journalctl -xe
sudo tail -n 100 /var/log/messages
Common mistakes
Expanding the wrong volume
Using the wrong disk group
Working on the passive node instead of the active node
Forgetting to grow the filesystem after growing the volume
Ignoring cluster ownership
Changing storage while service group is faulted
Running OS LVM commands on Veritas-managed disks
Using a size in the wrong unit
Running storage commands without approval
Dangerous commands
Be careful with:
vxdisksetup
vxdg adddisk
vxdg deport
vxdg import
vxassist growto
vxassist growby
vxedit
vxvol
mkfs
resize2fs
xfs_growfs
manual mount/umount of cluster-managed filesystems
hagrp -offline
hagrp -online
hagrp -switch
These can affect production data and availability.
Safe first command set
When I only need to check:
hastatus -sum
hagrp -state
hares -state
df -h
df -Th
findmnt
vxdg list
vxdisk list
vxprint -ht
Questions to ask before change
Which filesystem needs expansion?
Which application uses it?
Is the mount cluster-managed?
Which service group owns it?
Which node is active?
Which disk group contains it?
Which Veritas volume contains it?
Is there already free space in the disk group?
Was a new LUN added?
What is the approved new size?
Who approved the change?
What is the rollback/recovery plan?
Personal notes
Use this page as a checklist, not as a direct recipe.
Add sanitized environment-specific notes here only after verifying them at work.
Examples:
- Filesystem was Veritas-managed, not LVM-managed.
- Volume was expanded but filesystem still needed xfs_growfs.
- Active node had to be confirmed before checking the mount.
- Disk group free space was checked before any grow command.