Groove Relay Server Question

[Bob Nitrio]

I have been trying to get the answer to the following question through
resources available to Microsoft partners but I have not been making any
headway. Therefore, I will ask for assistance here in the public newsgroup.
I will change some of the details in order to make it appropriate for the
public as well as for those Microsoft partners on whose behalf I am trying
to find this answer.

I would like to start a Groove site that can be used by a group of my peers
for collaboration purposes. All of the users would be utilizing Groove
2007. The question I have is simply this. If I create this Groove site,
will the data be hosted on a public relay server operated by Microsoft or
one of its affiliates? I have found conflicting information on this
question. Some of the information says that there is indeed a public relay
server that will host the data. Other information says that there is no
public relay server. In that case, it appears that either I have to host
the information locally on my computer and keep it available at all times in
order for the Groove data to be accessible, or I need to setup a private
Groove server. Neither of these options is viable for this group of people.

Can you clarify this question about public relay servers? If you need to
take this offline, please contact me directly.

Bob N.

 

[Gregg Johnston [MSFT]]

Hi Bob,
So let’s start with terminology so that we are on the same page. You are
not creating a Groove site but a Groove workspace. This is important because
one of the key concepts to Groove is that almost everything is local. So if
you create a workspace, put some data into that workspace, and then invite
someone to the workspace, after the space arrives to the other user, they
will have the same data stored locally on their computer. What is in the
space resides on all members computers all protected by 192 bit encryption.
So what about those relay servers. If you are using Groove as a standalone
product and do not have any server products involved, then you are using the
Public Relay Servers. The function of those Relay Servers is to transfer
data. So let’s say you and your partner have a Groove Workspace that you are
both members of. You put a document into the space. Groove sees that there
is another member of the workspace that needs this document so it sends it to
the Relay Server. When the other member logs into Groove then that document
is then transferred to his computer and is no longer on the Relay Server.
Again, all of this is using the 192 bit encryption from start to finish.
Data on the Relay servers is never decrypted and is inaccessible except for
the intended recipient of that data (and that is only after they have logged
into Groove and have downloaded the data).

Does this answer your question?
Gregg

 

[Bob Nitrio]

Hi, Gregg,

Sorry about the terminology. I have been working on so many client sites
remotely lately that I completely mishandled the Groove workspace label. I
am completely onboard with everything you said concerning local storage, how
the relay server holds encrypted data until it is delivered to all of the
members who need to receive it, etc.

This is exactly how I understood Groove to work when it is installed as a
standalone product. Thank you for clarifying this for me and many others
who have had a difficult time getting such a great explanation as you
presented.

Bob N.

 

[David Schrag]

Some additional questions:

Do all Groove licenses -- including full packaged product (retail), Groove
Live subscriptions, and volume licenses -- give the licensee free, unlimited
access to the Public Relay Servers?

Do the Public Relay Servers hold data in transit indefinitely, or does the
data disappear after a while even if the additional workspace members never
log in to pick it up?

Suppose I create a 500 MB workspace and add "Joe" as a member. Joe accepts
my invitation to the workspace and acquires the workspace through the Public
Relay Servers. Now I add Jane. Will Joe and/or I have to upload the entire
workspace through the Public Relay Servers again in order for Jane to get
the data? Suppose Joe and I are both online at the moment Jane accepts the
invitation for the first time. How does Groove determine whether to pull the
data from Joe's copy or mine? What happens if we don't realize that Jane is
in the middle of getting the workspace and we disconnect from the Internet
during the first transfer? Can the relay servers pick up where they left of
or will we have to start the upload process all over again?

I am a pretty faithful reader of the Official Groove Blog and I have seen a
lot of discussion about the architecture involved with Groove Enterprise
Services and Groove Server, but I've seen very little written about the
Public Relay Servers. I work with very small businesses (<50 users, many
with <20 potential Groove users) who would probably prefer to use the Public
Relay Servers if possible, but I need to understand the performance,
managability, and security implications of that configuration.

TIA,
David

 

[Jan Talens]

Hi David,

Here some answers to your additional questions.

Q1.
Do all Groove licenses -- including full packaged product (retail),
Groove Live subscriptions, and volume licenses -- give the licensee
free, unlimited access to the Public Relay Servers?
mistaken, Microsoft warrants access for at least 5 years.
Note that Live Groove is an annual subscription.


Q2.
Do the Public Relay Servers hold data in transit indefinitely, or does
the data disappear after a while even if the additional workspace
members never log in to pick it up?
sync. Initially Groove will try to resolve this through a built in
mechanism. If Groove does not suuceed, the user will be marked as out
of sync and will need to be reinvited.


Q3.
Suppose I create a 500 MB workspace and add "Joe" as a member. Joe
accepts my invitation to the workspace and acquires the workspace
through the Public Relay Servers. Now I add Jane. Will Joe and/or I have
to upload the entire workspace through the Public Relay Servers again in
order for Jane to get the data? Suppose Joe and I are both online at the
moment Jane accepts the invitation for the first time. How does Groove
determine whether to pull the data from Joe's copy or mine? What happens
if we don't realize that Jane is in the middle of getting the workspace
and we disconnect from the Internet during the first transfer? Can the
relay servers pick up where they left of or will we have to start the
upload process all over again?
checks for the best source to pull data from. One major criteria is of
course bandwidth.
As the workspace data is cut in smaller bits, Groove can easilly recover
in case of an interruption. Fortunately, all this is transparent to the
users, who just see the data arrive and who do not need to bother about
where to get the data from.
More generally, we advise to have both users stay connected during the
initial recovery. But that is a best practice we have because of
previous Groove releases


Other considerations :
Although the use of public relay servers is free of charge, you may want
to consider some other aspects for Groove.
In case you want to manage your Groove environment, you will need to
deploy Groove servers or subscribe to Groove Enterprise Services (GES).
Managing your Groove environment includes managing users (centralized
authentification of users, ...) as well as security rules (identity and
device policies, ...).
The service level agreement for relay services included in Groove
Enterprise Services logically are higher than the SLA for the public relays.

Best regards,

Jan Talens

http://www.hommesetprocess.com/
http://www.grooveit.biz/

David Schrag a écrit :