What onboarding sets up on Azure
| What | What it is for |
|---|---|
The role “Zoorik Storage Operator” | A custom role, assigned to Zoorik’s app on each subscription you connect. It can read every resource, update VMs, run commands on them, and create and delete disks and snapshots. |
Zoorik’s app in your directory | The identity that holds the role. The connect script adds it to your directory. |
The agent on each VM | Installed through Azure Run Command when you onboard the VM. It reports the VM’s disks and mounts to Zoorik, and carries out the move, the expands and the shrinks. |
Pool disks | The managed disks that hold your data after the move. Amoeba adds, grows and removes them as the pool expands and shrinks. |
Safety snapshots | One incremental snapshot of each source disk, in the disk’s own resource group. Taken before Amoeba deletes the disk, and deleted after a few days. |
- The role is the limit on what Zoorik can do in your Azure. A permission added to the role later reaches your subscription only when you run the script again.
- Onboarding is your go-ahead. After the agent’s first report, the move starts by itself.
- A move only goes forward. Amoeba deletes a source disk only after its own snapshot has succeeded.
What your VMs need
- A running VM with at least one data disk. The data must sit on mounted filesystems.
- A few free data-disk slots. The pool’s disks attach beside the old ones during the move. With too few, Amoeba declines the VM and says why.
- Outbound access from the VM to Zoorik, and to the hosts the agent downloads from.
- On Linux: a recent kernel that supports btrfs. Every service that keeps a data mount open must run under systemd. The agent also needs rsync, btrfs-progs and psmisc, which the installer doesn’t add.
- VMs that use Azure Disk Encryption are not moved as they are.
To connect, you need the rights to add Zoorik’s app to your directory, create a custom role and assign it on each subscription. The product guide has the full list.
Questions about Amoeba on Azure
Which Azure disks does Amoeba work on?
The managed data disks of running Windows and Linux VMs. Amoeba never touches the OS disk. On AWS, it works with Amazon EBS volumes: see Amoeba for AWS.
Which Azure disk type does Amoeba’s pool use?
Premium SSD v2 where the VM’s region and zone offer it, and Premium SSD where they don’t. Data on Standard SSD, Standard HDD or Ultra Disk moves to the pool’s type, which can cost more than Standard disks. Zone-redundant (ZRS) disks map to a locally redundant pool, so the data loses zone redundancy.
Is there downtime on Azure?
Only at the switch. Once the copy onto the pool has caught up, the switch runs by itself, and the services that use those mounts stop and start again. After that, expands and shrinks run while your services keep working.
Do shrinks wait for me?
Only if you choose. Each VM has one setting, “When Zoorik shrinks”: in “Auto” shrinks run on their own, and in “Manual” each one waits for you. Expands run on their own, whatever the setting.
How is the saving worked out on Azure?
At Azure’s public list price for the disk type and region, measured against the source disks. A saving can be negative, for example when a pool has grown past the disks it replaced. Amoeba shows it as it is.