I have created the Azure VM and taken successfully the backup of disk using powershell. Now i need to create another VM using that disk?
How to proceed with that. Also i wanted to do same using image also. My main goal is to take backup of vm using either disk or image and then able to create new VMs using those backups or update the existing vm using those backups
The process you need to follow is covered here:
http://blogs.msdn.com/b/amol/archive/2014/03/21/how-to-migrate-windows-azure-vm-from-one-region-to-another.aspx
Note that if you don't "sysprep" the source VM when you attempt to provision a new VM based on the VHD of the source VM it will provision but the Azure management portal will show Running (provisioning) for the host.
Related
I want do the following
Create a VM
Install Mongo
Store Mongo DB data in Data disk
Delete the VM which excludes the Data disk
Then create a VM and use the above existing Data disk
My goal is goal create and delete the Azure VM, but re-use the single data disk.
How can I achieve it using ARM template?
If my understanding is right, on your scenario, you should use Azure Custom Script Extension to do this(Install Mongodb and change Mongodb data path).
You could check this question:How to change the location that MongoDB uses to store its data?.
You need write a script and test it on your VM and then use template to execute it.
hi i think you can buy a extra disk and link it to the vm it and while removing the vm detatch it and you can attach to which ever vm you want it on later please let me know if that is what you wanted.
We have a VM that runs a MongoDB and a web api to it.
Currently there is no backup solution enabled.
My goal is to protect data in MongoDB.
Is it a good idea to snapshot the whole Azure VM instance daily?
Or should I go for a database backup solution?
The Backup feature in the developer console for creating backups is great. I would however like the possibility to automate this. Is there a way to do so from the cf command line app?
Thanks
It's not possible from the cf cli, but there's an API endpoint for triggering backups.
API Docs | Custom Extensions | Swisscom Application Cloud Filter for
Cloud Foundry (CF) Cloud Controller (CC) API. Implements Swisscom
proprietary extensions
POST /custom/service_instances/{service-instance-id}/backups
Creates a backup for a given service instance
See for more Info Service Backup and Restore in docs.developer.swisscom.com
Create Backup To create a backup, navigate to the service instance in
the web console and then to the “Backups” tab. There you can click the
“Create” button to trigger a manual backup.
Note: Backups have to be triggered manually from the web console.
Be aware that you can only keep a set number of backups per service
instance. The actual number is dependent on the service type and
service plan. In case you already have the maximum number, you cannot
create any new backups before deleting one of the existing.
It may take several minutes to backup your service (depending on the
size of your service instance).
Restore Backup You can restore any backup at any time. The current
state of your backup will be overwritten and replaced with the state
saved to the backup. You are advised to create a backup of the current
state before restoring an old state.
Limitations You can only perform one backup or restore action per
service instance at a time. If an action is still ongoing, you cannot
trigger another one. You cannot exceed the maxmimum number of backups
per service instance
We did this by developing a small Node.js application which is running on the cloud in the same space and which backups our maria and mongo db every night automatically.
EDIT:
You can download the code from here:
https://github.com/theonlyandone/cf-backup-app
Fresh from the press: Swisscom Application Cloud cf CLI Plugin can also automate backup and restore.
The official cf CLI plugin for the Swisscom Application Cloud gives
you access to all the additional features of the App Cloud.
cf install-plugin -r CF-Community "Swisscom Application Cloud"
from 0.1.0 release notes
Service Instance Backups
Add cf backups command (list all backups of a service instance)
Add cf create-backup command (create a new backup of a service instance)
Add cf restore-backup command (restore an existing backup of a service instance)
Add cf delete-backup command (delete an existing backup of a service instance)
Despite the answer from Matthias Winzeler saying it's not possible, in fact it's totally possible to automate MariaDB backups through the command line.
I developed a plugin for the CF CLI:
https://github.com/gsmachado/cf-mariadb-backup-plugin
In future I could extend such plugin to backup any kind of service that is supported by the Cloud Foundry Provider's API (in this case, Swisscom AppCloud API).
I have created instance in compute engine with windows server 2012. i cant see any option to take automatic backup for instance disk database everyday. there is option of snapshot but we need to operate this manually. please suggest any way to backup automatically and can be restore able on a single click. if is there any other possibility using cloud SQL storage or any other storage please recommend.
thanks
There's an API to take snapshots, see API section here:
https://cloud.google.com/compute/docs/disks/create-snapshots#create_your_snapshot
You can write a simple app to get triggered from Cron or something to take a snapshot periodically.
You have no provision for automatic back up for compute engine disk. But you can do a manual disk backup by creating a snapshot.
Best alternative way is to create a bucket and move your files there. Google cloud buckets have automated back up facility available.
Cloud storage and cloud SQL are your options for automated back ups in google cloud.
Added more details at the bottom of the question.
We are testing deployment scenarios in Azure VM preview and have run into an issue.
Here is our scenario. We have a software stack that we use in all of our servers. We have created an image with all of that stack installed on an attached data drive. We have created a image of the VM that we can use as a template. Now what we want to do is to to create a VM based on that template and create a copy of the data drive and attach it to the newly created VM in an automated manner.
Our problem is that while we have found lots of information about creating drives, we can't find any guidance on how to copy the data drive using Azure for Powershell.
Any thoughts, code, or RTFMs happily accepted.
Cheers,
Terence
We have sucessfully created an operating system image that we can use to create VM's. But there is a data disk that holds our standard software stack that we want to reuse by copying it across VMs. The scenario that we are trying to implement is:
Create a VM from a standard VM image - PBIMaster
Attach a disk as F to that image called PBIMasterDisk
Install all of the software required for our app on F: (to big for the OS disk and besides sticking it on the OS disk seems messy)
Build an image from PBIMaster call it PBIMasterImage save it.
Create a new image from PBIMaster call it Node1
Copy PBIMasterDisk to a new Azure disk call it Node1Software disk
Attach Node1Softwaredisk to Node1 as F:
Since the image has the correct registry settings from the previous installs our stack is ready to go.
9 Add appropriate endpoints.
Rinse and repeat for each additional node.
Hopefully that makes our scenario clearer.
Thanks.
If I understood your objective correctly you already have uploaded two VHD in your subscription and you have also create a VM based on your OS Disk VHD1:
OS Disk (VHD1)
Data Disk (VHD2)
Now you want to copy VHD2 to VHD3 and then attach VHD3 to your VM (which is based on OS disk) via Powershell.
As of there is no powershell command which will let you copy DataDisk (VHD2) to another data disk (i.e VHD3)..
I haven't tried but you can use the following code to try copying your DataDisk:
http://blogs.msdn.com/b/windowsazurestorage/archive/2012/06/12/introducing-asynchronous-cross-account-copy-blob.aspx
This method does copy blobs directly at cloud storage level so there is no bandwidth usage towards on-premise and potentially zero cost if you are in same DC. Trying using the same subscription and see if that solves your problem.