I'm using vmware
Some actions such as adding secondary ip doesn't trigger hypervisor hooks
On Mon, Feb 25, 2019, 12:31 Dag Sonstebo Alireza - which hypervisor do you use? If you use KVM you can use libvirt
> hooks - https://libvirt.org/hooks.html.
>
> Regards,
> Dag Sonstebo
> Cloud Architect
>
Alireza - which hypervisor do you use? If you use KVM you can use libvirt hooks
- https://libvirt.org/hooks.html.
Regards,
Dag Sonstebo
Cloud Architect
ShapeBlue
On 23/02/2019, 15:27, "Suresh Kumar Anaparti"
wrote:
Hi Alireza,
Other approach I can think of is database
Hi Alireza,
Other approach I can think of is database triggers (can impact performance
based on the jobs frequency).
Take a look at the async_job table's insert and update operations which are
performed on creating a new job (job status 0) and updated when job is
complete with non-zero job
Thank you Ivan!
A novel solution based on existing capabilities!
I'll investigate is for our needs.
On Fri, Feb 22, 2019 at 4:49 PM Ivan Kudryavtsev
wrote:
> Hi, Alireza. Easy solution is to use Kafka or RMQ event bus, which are
> supported by cloudstack.
>
> Take a look on the article in our
Async jobs is not my concern.
Imagine that you have a firewall that is not integrated with CS and you
want to insert new roles in it each time a new VM is created.
In this case you need a hook that triggers an external script to configure
the firewall.
Pulling is time and resource consuming and
Hi Alireza,
Could you elaborate a more on the use case you have in mind? The description
you mention is very generic.
As far as I know we need to use polling to track the status of any async job
(which is another API call) and execute based on response. That seems to be
foundation principle
Hi, Alireza. Easy solution is to use Kafka or RMQ event bus, which are
supported by cloudstack.
Take a look on the article in our blog:
https://bitworks.software/en/2017-07-23-cloudstack-3rd-party-integration-events-subscription-kafka.html
We use it very extensively for accounting, billing, DNS
