Storage

Two kinds of storage on The Wahda Cloud:

1. Block storage — persistent SSD volumes you attach to a VM. Behaves like a physical disk: format it, mount it, put a database on it. Snapshot it for point-in-time copies, back it up for durable off-instance recovery.
2. Object storage — S3-compatible bucket storage for anything you'd normally put in AWS S3: static site assets, media libraries, application uploads, backup archives, data-lake dumps. Reach it with any S3-compatible client.

Both live under Storage in the left navigation of console.thewahda.com.

---

Block storage volumes

Every VM you launch through the Create VM wizard already gets one block volume — its System Disk — created automatically. You can attach additional volumes at any time for data that outlives the instance or needs to move between VMs.

The volume type

The public catalog has one tier today:

Type	Media	Use case
SSD-GP1	General-purpose SSD	Almost everything — root disks, database storage, application data, media libraries.

Where the volumes list lives

From the left navigation: Storage → Volumes.

Storage › Volumes

Volumes list. Each row shows Size, Status, Type, which VM the volume is attached to, whether it's bootable, and when it was created.

The header offers Create Volume, Accept Volume Transfer (paired with another project sending it), and Delete. Row actions include Edit and More (attach/detach, resize, snapshot, transfer).

The row's Status values you'll see day-to-day:

Status	What it means
Available	Not attached to any VM. Safe to attach, snapshot, or delete.
In-use	Attached to a VM. The Attached To column tells you which one and at which device (/dev/vda, /dev/vdb).
Reserved	Being attached — transient.
Creating / Deleting	Provisioning or teardown — transient.
Error	Something went wrong. Email info@thewahda.com with the volume ID.

Create a volume

Click Create Volume at the top of the list.

Storage › Volumes › Create Volume

Blank volume, image, or snapshot as data source; SSD-GP1 as the only public type; capacity from 1 GiB to 920 GiB via the slider.

Field	What to enter
Available Zone	Leave Not select unless you need to pin the volume to a specific AZ (multi-AZ layouts).
Data Source Type	Blank Volume for a fresh empty disk, Image to seed from an OS image (used automatically when the Create VM wizard makes a boot volume), or Volume Snapshot to restore from an existing snapshot.
Volume Type	Pick SSD-GP1.
Capacity (GiB)	The size you want. The slider goes from 1 GiB to 920 GiB; type an exact number in the box on the right.
Name	1–128 chars, letters/digits/-/_/.. Pick something you'll recognize — webapp-data-2026, not vol1.
Count	Leave 1 unless you're batch-creating identical disks.

Click Confirm. The volume moves through Creating → Available in a few seconds.

Attach a volume to a VM

From the volume's row → More → Attach, pick the VM, and confirm. The device name (/dev/vdb, /dev/vdc, …) is picked by the platform.

On the VM, format the fresh volume once and mount it:

sudo mkfs.ext4 /dev/vdb
sudo mkdir -p /mnt/data
sudo mount /dev/vdb /mnt/data

Persist the mount across reboots by adding to /etc/fstab — get the volume's UUID with sudo blkid /dev/vdb and add:

UUID=<uuid> /mnt/data ext4 defaults,nofail 0 2

nofail matters — if the volume is ever detached, the VM will still boot without dropping into an emergency shell.

Resize (grow) a volume

From the volume's row → More → Extend Volume. Type the new (larger) size and confirm. You can only grow — never shrink.

After the platform grows the underlying storage, tell the guest OS:

# ext4
sudo resize2fs /dev/vdb

# xfs
sudo xfs_growfs /mnt/data

Detach and reattach

Detach frees the volume from the current VM without deleting it. Reattach it to any other VM in the same project. Great for one-shot data migrations: create fresh VM → detach old VM's data volume → attach to new VM → cut over.

Delete a volume

The volume must be Available (detached) before you can delete it. Tick the row → Delete in the header. If you might want the data later, take a snapshot first — deleting is irreversible.

---

Snapshots vs. backups

Two similar-sounding but distinct concepts. Pick based on what you need to survive.

	Volume Snapshot	Volume Backup
What it is	Point-in-time reference to a volume's data, stored on the same underlying storage system.	Full copy of the volume's data written to durable object storage, independent of the source volume.
Speed	Near-instant to create.	Slower — proportional to volume size.
Where the data sits	Same storage backend as the source. If that backend has a bad day, both go together.	Separate object storage. Survives loss of the source volume.
Cost	Small — only differential blocks are stored.	Larger — full data copy, billed per GB of backup storage.
Restore path	Create a new volume with Data Source Type = Volume Snapshot.	Restore into a new volume from Storage → Volume Backups → Restore.
When to use	Fast, cheap rollback points during a risky config change or before a deploy.	Disaster recovery. Compliance retention. Off-site copy for a critical database.

Rule of thumb: snapshots for "I might want to roll this back in the next hour"; backups for "I need this data next month if everything else is gone."

Both are managed under Storage → Volume Snapshots and Storage → Volume Backups.

---

Object storage (S3-compatible)

For unstructured data — static site assets, images, videos, application uploads, backup archives, log dumps — use object storage. Buckets look and feel like AWS S3, and any S3-compatible tool works against them.

Where buckets live

From the left navigation: Storage → Object Storage. The page lists every bucket in the current project. Fresh project → empty list.

The endpoint

https://api.thewahda.com:6780/swift/v1

This one endpoint serves the S3-compatible API. Point any AWS SDK, aws-cli, or third-party S3 tool at it.

Configure aws-cli

Create a profile that talks to the endpoint above with your credentials:

aws configure --profile wahda
# AWS Access Key ID:     <your access key>
# AWS Secret Access Key: <your secret>
# Default region name:   in-north-1
# Default output format: json

You get access credentials from the console — see Application credentials for the recommended way to create machine credentials.

Then override the endpoint on every call:

aws --profile wahda --endpoint-url https://api.thewahda.com:6780/swift/v1 s3 ls

Common recipes

Create a bucket and upload a file:

aws --profile wahda --endpoint-url https://api.thewahda.com:6780/swift/v1 s3 mb s3://my-assets
aws --profile wahda --endpoint-url https://api.thewahda.com:6780/swift/v1 s3 cp \
    ./photo.jpg s3://my-assets/photo.jpg

Host a static site's assets:

aws --profile wahda --endpoint-url https://api.thewahda.com:6780/swift/v1 s3 sync \
    ./dist/ s3://my-assets/site/ --delete

Nightly offsite database dump:

# in a cron / systemd timer on a jump VM
pg_dump -h db-endpoint -U analytics analytics | gzip | \
  aws --profile wahda --endpoint-url https://api.thewahda.com:6780/swift/v1 \
    s3 cp - s3://backups/pg-$(date +%F).sql.gz

Download an entire bucket:

aws --profile wahda --endpoint-url https://api.thewahda.com:6780/swift/v1 s3 sync \
    s3://my-assets ./local-copy/

Public vs. private objects

By default, every object you upload is private — only requests carrying your credentials can read it. If you want a bucket's contents publicly readable (e.g. static site assets), grant public read via ACL or bucket policy on that specific bucket. Do not make a bucket world-writeable — that's how you end up hosting other people's malware.

---

Region and availability

Both block volumes and object storage buckets are region-scoped. Today the only region is in-north-1 (Hyderabad datacenter, single AZ in-north-az1). Multi-region replication is not exposed yet — if you need offsite copies today, run a scheduled sync into a remote provider from a jump VM.

---

Cost model

Pay-as-you-go:

• Block volumes — billed per GB per month for the allocated capacity, whether attached or detached, whether the VM is running or stopped. Delete volumes you're not using.
• Volume backups — billed per GB of backup storage consumed.
• Object storage — billed per GB stored + egress out of the region. Ingress and intra-region transfer are free.

Invoiced in INR with GST-compliant billing.

---

Troubleshooting

Symptom	Where to look
Volume stuck in Creating for more than 60 seconds	Rare. Email info@thewahda.com with the volume ID.
Cannot detach — "volume in use"	The VM's OS is still writing to the mount. sudo umount /mnt/data first, then detach.
resize2fs says "nothing to do" after Extend	The extend didn't propagate yet. Check the volume shows the new size in the console; wait a moment; retry.
aws s3 ls returns Could not connect	The --endpoint-url is missing or the client machine can't reach api.thewahda.com:6780. Confirm both.
aws s3 ls returns 403 Forbidden	The credentials profile is wrong or lacks permission on the bucket. Regenerate the application credential and retry.
Uploaded object isn't publicly reachable	Objects are private by default. Set a bucket policy or per-object ACL for public read.

---

Next steps

• Create a VM → — the compute that attaches your volumes.
• Backups & restore → — how managed databases use object storage for backups.
• Application credentials → — machine access keys for CI and scripts.