You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cloudstack.apache.org by David Knuth <dk...@ussignal.com.INVALID> on 2025/03/10 14:33:29 UTC
Apache Cloudstack question
Good morning! I have what I hope is a quick and simple question. My company is working with Cloudstack and we're looking at some of the features called out in the documentation, such as the new B&R plugin, as well as the snapshot copy functionality as we are working on setting up a feasible disaster recovery solution for our clients.
We've looked into some of the options with snapshot copying and Linbit's DRBD solution for replication of snapshots/data, and I was curious about something.
In the documentation for snapshotting, it says that the snapshots are taken and then immediately copied off to secondary storage. Is it possible to define what secondary storage is targeted on a per-domain/instance basis for this? E.g., could we set up separate 'secondary storage' targets per customer domain, such that snapshots for one customer's domain remain separate from general storage/the rest of the customers?
[David Knuth]<https://www.bqignite.com/api/SignatureMetadata/GetRedirect?Guid=1e8fc43aed8e48428e08dd490aced0d4>
RE: Apache Cloudstack question
Posted by David Knuth <dk...@ussignal.com.INVALID>.
Alex,
We're a managed service provider, and we segment out customers via domain/"Tenant". We're working on a means of replicating features we have for our other offerings which allow us to do cross-market replication of VMs, with historic checkpoints, so that in the event of a market outage or issue with the systems in the customer's primary production pool (e.g., a pool running in location A), we can quickly and easily reconstitute and bring online their protected instances in a separate market that is geodiverse (e.g., resource pool running in location B), with the ability to choose the point at which the machines can be restored from by their checkpoint. We would then ideally be able to replicate data back to their original location so that their workloads can be brought up in the original location after the original disaster is over.
There are a number of toolsets and approaches that discuss HA of Cloudstack as a whole across multiple sites, however there's little in the way of easy approaches, it seems, to be able to selectively protect instances, etc. within Cloudstack to allow for a per-customer type DR protection schema.
Thanks!
-----Original Message-----
From: Alex Mattioli <Al...@shapeblue.com>
Sent: Monday, March 10, 2025 11:13 AM
To: users@cloudstack.apache.org
Subject: Re: Apache Cloudstack question
EXTERNAL
Hi David,
Currently there isn't a way to segregate secondary storages per domain.
What's the use case there?
Alex.
________________________________
From: David Knuth <dk...@ussignal.com.INVALID>
Sent: Monday, March 10, 2025 7:33:29 AM
To: users@cloudstack.apache.org <us...@cloudstack.apache.org>
Subject: Apache Cloudstack question
Good morning! I have what I hope is a quick and simple question. My company is working with Cloudstack and we're looking at some of the features called out in the documentation, such as the new B&R plugin, as well as the snapshot copy functionality as we are working on setting up a feasible disaster recovery solution for our clients.
We've looked into some of the options with snapshot copying and Linbit's DRBD solution for replication of snapshots/data, and I was curious about something.
In the documentation for snapshotting, it says that the snapshots are taken and then immediately copied off to secondary storage. Is it possible to define what secondary storage is targeted on a per-domain/instance basis for this? E.g., could we set up separate 'secondary storage' targets per customer domain, such that snapshots for one customer's domain remain separate from general storage/the rest of the customers?
[David Knuth]<https://urldefense.com/v3/__https://www.bqignite.com/api/SignatureMetadata/GetRedirect?Guid=1e8fc43aed8e48428e08dd490aced0d4__;!!P9cq_d3Gyw!gMSZNluLy7278OG-6-8iGW53guKYOgDwUHufOzixLZidA9JtdDD1Eahdny1tDkWOfxg2y-S7qZWkKsDNhwRt-pLSJQ$ <https://urldefense.com/v3/__https:/www.bqignite.com/api/SignatureMetadata/GetRedirect?Guid=1e8fc43aed8e48428e08dd490aced0d4__;!!P9cq_d3Gyw!gMSZNluLy7278OG-6-8iGW53guKYOgDwUHufOzixLZidA9JtdDD1Eahdny1tDkWOfxg2y-S7qZWkKsDNhwRt-pLSJQ$%20> >
[David Knuth]<https://www.bqignite.com/api/SignatureMetadata/GetRedirect?Guid=1e8fc43aed8e48428e08dd490aced0d4&Subject=RE%3a+Apache+Cloudstack+question>
Re: Apache Cloudstack question
Posted by Fabricio Duarte <fa...@scclouds.com.br>.
Hey David and Alexandre,
Actually, it is possible to define what secondary storage is targeted on
a per-domain basis using the secondary storage selectors functionality.
You can use the `createSecondaryStorageSelector` API to define a rule to
redirect snapshots to a specific secondary storage depending on the
domain. For instance:
```
(admin) đ˘ > create secondarystorageselector name="<name of the rule>"
description="<description of the rule>" type=SNAPSHOT zoneid=<uuid of
the zone> heuristicrule="if (account.domain.id === '<uuid of domain X>')
{ '<uuid of secondary storage A>' } else { '<uuid of secondary storage
B>' }"
```
With this rule, snapshots taken at domain X will be copied to secondary
storage A, and snapshots taken at other domains will be copied to
secondary storage B.
It is not possible to redirect on a per-instance basis currently though,
but the code can be easily extended to allow it. You can read more about
this functionality, and about other currently available preset variables
at [1].
[1]:
https://docs.cloudstack.apache.org/en/4.20.0.0/adminguide/storage.html#direct-resources-to-a-specific-secondary-storage
Regards,
Fabricio
On 3/10/25 12:13, Alex Mattioli wrote:
> Hi David,
>
> Currently there isnât a way to segregate secondary storages per domain.
>
> Whatâs the use case there?
>
> Alex.
>
>
>
> ________________________________
> From: David Knuth <dk...@ussignal.com.INVALID>
> Sent: Monday, March 10, 2025 7:33:29 AM
> To: users@cloudstack.apache.org <us...@cloudstack.apache.org>
> Subject: Apache Cloudstack question
>
>
> Good morning! I have what I hope is a quick and simple question. My company is working with Cloudstack and we're looking at some of the features called out in the documentation, such as the new B&R plugin, as well as the snapshot copy functionality as we are working on setting up a feasible disaster recovery solution for our clients.
>
>
>
> We've looked into some of the options with snapshot copying and Linbit's DRBD solution for replication of snapshots/data, and I was curious about something.
>
>
>
> In the documentation for snapshotting, it says that the snapshots are taken and then immediately copied off to secondary storage. Is it possible to define what secondary storage is targeted on a per-domain/instance basis for this? E.g., could we set up separate 'secondary storage' targets per customer domain, such that snapshots for one customer's domain remain separate from general storage/the rest of the customers?
>
>
>
>
> [David Knuth]<https://www.bqignite.com/api/SignatureMetadata/GetRedirect?Guid=1e8fc43aed8e48428e08dd490aced0d4>
>
Re: Apache Cloudstack question
Posted by Alex Mattioli <Al...@shapeblue.com>.
Hi David,
Currently there isnât a way to segregate secondary storages per domain.
Whatâs the use case there?
Alex.
________________________________
From: David Knuth <dk...@ussignal.com.INVALID>
Sent: Monday, March 10, 2025 7:33:29 AM
To: users@cloudstack.apache.org <us...@cloudstack.apache.org>
Subject: Apache Cloudstack question
Good morning! I have what I hope is a quick and simple question. My company is working with Cloudstack and we're looking at some of the features called out in the documentation, such as the new B&R plugin, as well as the snapshot copy functionality as we are working on setting up a feasible disaster recovery solution for our clients.
We've looked into some of the options with snapshot copying and Linbit's DRBD solution for replication of snapshots/data, and I was curious about something.
In the documentation for snapshotting, it says that the snapshots are taken and then immediately copied off to secondary storage. Is it possible to define what secondary storage is targeted on a per-domain/instance basis for this? E.g., could we set up separate 'secondary storage' targets per customer domain, such that snapshots for one customer's domain remain separate from general storage/the rest of the customers?
[David Knuth]<https://www.bqignite.com/api/SignatureMetadata/GetRedirect?Guid=1e8fc43aed8e48428e08dd490aced0d4>