Contents

OCI Storage Environment

In the OCI environment, you can, for example,

create a new storage volume (a block volume),
attach an existing storage volume to your instance,
detach a storage volume from your instance.

These steps are shown below.

For more details, please refer to the Oracle Cloud documentation.

Creating a New Volume

Step	Details
*Open the Block Storage* configuration from the Core Infrastructure menu section.**	This will open the block volume list.
*Click on Create Block Volume* to create a new virtual disk.**	This will open the block volume creation screen.
Enter the details of the new volume.	Add at least name, compartment, availability zone, and size and performance. Adapt the other settings as required. When done, click on Create Block Volume. If you chose to view the volume details after the volume has been created, the detail screen will be displayed.

Return to the instance list to attach the new volume to your instance.

Attaching an Existing Volume to an Instance

After returning to the instance list (via Compute > Instances), select your instance to display the instance details. Then perform the following steps to attach the block volume to your instance:

Step	Details
*Select Attached Block Volumes* from the Resources section on the left.**	In the example above, no volumes have been attached so far. Click on Attach Block Volume to select the volume that is to be attached.
Select the parameters for the volume and attach it to the instance.	Select which block volume to attach, the parameters for attaching the device (paravirtualized, read/write), the compartment, and the device path presented to the Charon host. Once all parameters have been set, click on Attach.
Verify that the device has been attached.	The instance details list now shows the attached volume.

Detaching a Volume from an Instance

First, unmount the volume in the Charon host system (see below).

Then detach the volume from your instance:

Go to the Attached Block Volumes list in the instance details display.
Click on the three dots next to the volume to be detached and select Detach.

Steps on the Charon Host System

Mounting a Newly Attached Volume Using the Storage Manager (SSP AL Images Only)

After the volume has been attached to the instance, it must be included in the Charon-SSP host system configuration. On Charon-SSP AL instances, this is achieved via the Charon-SSP Manager.

Open the Charon-SSP Manager on your local system and connect to your Charon cloud instance.
Select Tools > Oracle Cloud > Storage Manager.
In the Storage Manager window, perform the following steps:
1. Select the new device.
2. Click on the cog-wheel symbol.
3. Only if required, select Format Volume to create a filesystem on the new device.
  Please note: this will delete all data on the volume.
4. Click on the cog-wheel symbol and select Mount the Filesystem.

This will mount the new volume under /charon/storage/media/<UUID>/. The following image shows a sample:

Once the filesystem has been mounted, the space is available to the Charon-SSP host system. After the first mount via the Storage Manager, the filesystem will be automatically mounted after a restart of the Charon host instance.

Mounting a Newly Attached Volume Manually

This is an example of how to mount (and if necessary partition) an additional disk on a Charon host system. Please refer to the Linux manual pages for details.

The general tasks on the Charon host system require to identify the disk, add a file system to it (if this has not been done before), and mount the disk on a suitable mount-point.

Please note: the different cloud environments may offer the disk volumes to the Linux instance using different names. However, the basic steps will be the same as in the examples below.

Step 1: Identify new disk

After logging in on the system, you can identify the new disk using the lsblk command:

# lsblk
NAME   MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
sda      8:0    0   20G  0 disk 
├─sda1   8:1    0  200M  0 part /boot/efi
└─sda2   8:2    0 19,8G  0 part /
sdb      8:16   0  200G  0 disk

In the example above, the new disk is /dev/sdb. The output shows no mount-point, i.e., the disk is not mounted yet. It also does not have any partitions.

Please note:

A disk without partitions can also have a filesystem and data on it. Hence be sure that the disk really does not have any important data on it before you partition it.
If a system has many disks, it is helpful to run the lsblk command before the new disk is added. This makes it easy to identify the new disk in the output after it has been added.

Step 2: Partition disk (fdisk or parted) - only if required

Please note: This step is only meant for new disks or to re-partition an existing disk. It will destroy all data on an existing disk.

Please refer to the manual pages ($ man parted and $ man fdisk) of your Linux distribution for details on the disk-partitioning commands. If the whole disk is used for one filesystem, it is not strictly required to create a partition. The decision of which disk layout is required depends on the customer requirements is the responsibility of the user.

After creating one partition on disk with fdisk (# fdisk /dev/sdb), the lsblk output shows the new partition:

# lsblk
NAME   MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
sda      8:0    0   30G  0 disk
├─sda1   8:1    0  500M  0 part /boot
└─sda2   8:2    0 29,5G  0 part /
sdb      8:32   0   64G  0 disk
└─sdb1   8:33   0   64G  0 part

Step3: Create a filesystem on the new partition(s)

Use the mkfs command to create a new filesystem. The selection of a filesystem depends on customer requirements. For example, to create an XFS filesystem, use

# mkfs.xfs /dev/sdb1

Please refer to the documentation of your Linux distribution for details about the mkfs command.

Step 4: Create a mount-point and mount the new filesystem

The following example shows how to create a mount-point and mount the file system.
To keep the example consistent with the sample outputs above, /dev/sdb1 is used in this example. However, as the /dev/sdX device names are not guaranteed to be persistent across reboots, it is strongly recommended to use names from the /dev/disk/by-* hierarchy (for example by-uuid) for permanent, production use.

# mkdir /space# mount /dev/sdb1 /space

The df command shows the mounted filesystem:

# df
Filesystem     1K-blocks    Used Available Use% Mounted on
devtmpfs         4065684       0   4065684   0% /dev
tmpfs            4077556      16   4077540   1% /dev/shm
tmpfs            4077556    9224   4068332   1% /run
tmpfs            4077556       0   4077556   0% /sys/fs/cgroup
/dev/sda2       30929148 1677416  29251732   6% /
/dev/sda1         508580   65512    443068  13% /boot
tmpfs             815512       0    815512   0% /run/user/1000
/dev/sdb1       65923628   53272  62498580   1% /space

Step 5: Mount the disk automatically at system boot

To mount the disk automatically when the system boots, you must add it to the file /etc/fstab.

Please note: The device naming /dev/sdXN (e.g., /dev/sdb1) is not guaranteed to be persistent across reboots. Hence, it is advisable to use a persistent name from the /dev/disk/by* hierarchy (for example, the UUID).

You can use the ls or the blkid command to identify the UUID. Examples:

$ ls -l /dev/disk/by-uuid/
total 0
lrwxrwxrwx. 1 root root 10 2020-08-14 21:14 0c523909-fb78-48cb-9dc8-e7a08197a673 -> ../../dm-4
lrwxrwxrwx. 1 root root 10 2020-08-14 21:14 31fa8e8c-a6c0-45f7-9892-da13ba81e0e5 -> ../../sdb1

$ blkid |grep sdb1
/dev/sdb1: UUID="31fa8e8c-a6c0-45f7-9892-da13ba81e0e5" BLOCK_SIZE="4096" TYPE="xfs" PARTUUID="db62deaa-f25f-43d4-b958-700c1c13d844"

To add the device to /etc/fstab perform the following steps:

As the root user, open the file /etc/fstab with a text editor.
Add the mount command to the file. Please note: The following is for illustration only. The exact options depend on your requirements.
Sample fstab entry:
UUID=31fa8e8c-a6c0-45f7-9892-da13ba81e0e5 /space xfs defaults 1 2
Save the file.
Test if the automatic mount works correctly.

Unmounting a Volume

To unmount a volume before perform the following steps:

Stop all Charon instances that might use the volume that is about to be unmounted.
On host systems based on AL images:
- In Charon Manager go to Tools > Oracle Cloud > Storage Manager.
- Select the volume.
- Click on the cogwheel symbol and select Unmount the Filesystem.
On other systems:
- Use the command # umount <device-path> or # umount <mount-point>
- To make this permanent, remove the corresponding entry in /etc/fstab.

Browser not supported