Hey Guys,

Im pretty new to RabbitMQ and the whole AMPQ protocol.

I have read and done a few tutorials on RabbitMQ using a python
client. But i keep seeing people creating vhosts like so :

rabbitmqctl add_vhost myhost

My question is, what are the vhosts for, are they required ? If not
then what are the advantages of using them ?

Search Discussions

  • Alexandru Scvorţov at Jan 26, 2012 at 7:39 am
    what are the vhosts for
    Vhosts are a way to group queues, exchanges, bindings and user
    permissions. They're like namespaces in which the above are declared.
    are they required
    You need at least one vhost. It's usually called '/'.

    Cheers,
    Alex
    On Wed, Jan 25, 2012 at 06:37:00PM -0800, czpython wrote:
    Hey Guys,

    Im pretty new to RabbitMQ and the whole AMPQ protocol.

    I have read and done a few tutorials on RabbitMQ using a python
    client. But i keep seeing people creating vhosts like so :

    rabbitmqctl add_vhost myhost

    My question is, what are the vhosts for, are they required ? If not
    then what are the advantages of using them ?
    _______________________________________________
    rabbitmq-discuss mailing list
    rabbitmq-discuss at lists.rabbitmq.com
    https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss
  • Elias Levy at Jan 26, 2012 at 5:48 pm

    On Thu, Jan 26, 2012 at 4:00 AM, wrote:

    My question is, what are the vhosts for, are they required ? If not
    then what are the advantages of using them ?
    Virtual hosts give you namespace separation. All users, exchanges, queues
    and bindings in each virtual host are unreachable from all other virtual
    hosts. This is a good feature to use if you want ensure
    compartmentalization.

    Alas, if you are like us, you'd like to compartmentalize customer data
    using virtual hosts, but you'd also like to shuffle some data among those
    virtual hosts (e.g. from a a primary VH to individual customer VHs), but
    this becomes impractical in a dynamic environment.

    The only want to shuffle that data among VHs it to use the shovel or
    federation plugins, and both require that you restart the nodes hosting
    them to make any changes (e.g. add a customer and thus add a shovel to
    their own virtual host).

    Elias
    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20120126/b39e6b69/attachment.htm>
  • Alexandru Scvorţov at Jan 26, 2012 at 5:55 pm

    The only want to shuffle that data among VHs it to use the shovel or
    federation plugins,
    That's not entirely true. The shovel is just a thin wrapper around the
    Erlang client. You can use any of the clients to do what the shovel
    does (but that requires a bit of coding).

    Alex
    On Thu, Jan 26, 2012 at 09:48:47AM -0800, Elias Levy wrote:
    On Thu, Jan 26, 2012 at 4:00 AM, <
    rabbitmq-discuss-request at lists.rabbitmq.com> wrote:
    My question is, what are the vhosts for, are they required ? If not
    then what are the advantages of using them ?
    Virtual hosts give you namespace separation. All users, exchanges, queues
    and bindings in each virtual host are unreachable from all other virtual
    hosts. This is a good feature to use if you want ensure
    compartmentalization.

    Alas, if you are like us, you'd like to compartmentalize customer data
    using virtual hosts, but you'd also like to shuffle some data among those
    virtual hosts (e.g. from a a primary VH to individual customer VHs), but
    this becomes impractical in a dynamic environment.

    The only want to shuffle that data among VHs it to use the shovel or
    federation plugins, and both require that you restart the nodes hosting
    them to make any changes (e.g. add a customer and thus add a shovel to
    their own virtual host).

    Elias
    _______________________________________________
    rabbitmq-discuss mailing list
    rabbitmq-discuss at lists.rabbitmq.com
    https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss
  • Elias Levy at Jan 26, 2012 at 6:00 pm

    On Thu, Jan 26, 2012 at 9:55 AM, Alexandru Scvor?ov wrote:
    The only want to shuffle that data among VHs it to use the shovel or
    federation plugins,
    That's not entirely true. The shovel is just a thin wrapper around the
    Erlang client. You can use any of the clients to do what the shovel
    does (but that requires a bit of coding).
    You are correct. I should have said, doing so within Rabbit itself.
    Obviously, an external client can consume and republish wherever it wants.

    Elias
    -------------- next part --------------
    An HTML attachment was scrubbed...
    URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20120126/e8b18e04/attachment.htm>

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
grouprabbitmq-discuss @
categoriesrabbitmq
postedJan 26, '12 at 2:37a
activeJan 26, '12 at 6:00p
posts5
users3
websiterabbitmq.com
irc#rabbitmq

People

Translate

site design / logo © 2022 Grokbase