What’s the best way to secure merge replication over the internet? The cl
ient is running MSDE and could be a home user with broadband. If it’s usi
ng a VPN would this be SSL or IPSec?Either would secure the traffic from MSDE to the Publisher.
VPN would allow the client to use other corporate resources as well and
authenticate using Windows
Authentication.
At a minimum put a server certificates on the Publisher and update the
Trusted Root Authority
on the Subscriber (MSDE) and use SSL.
316898 HOW TO: Enable SSL Encryption for SQL Server 2000 with Microsoft
http://support.microsoft.com/?id=316898
276553 HOW TO: Enable SSL Encryption for SQL Server 2000 with Certificate
Server
http://support.microsoft.com/?id=276553
If you have an ISA Server, you could publish the SQL resource as well.
Here's some docs on using ISA.
http://www.microsoft.com/technet/pr...tain/proxy.mspx
Thanks,
Kevin McDonnell
Microsoft Corporation
This posting is provided AS IS with no warranties, and confers no rights.|||Thanks Kevin - I left out some important details. Actually, the client side
would be the Publisher (MSDE). We would like to use replication to backup
the database with our application for our customers.
"Kevin McDonnell [MSFT]" wrote:
> Either would secure the traffic from MSDE to the Publisher.
> VPN would allow the client to use other corporate resources as well and
> authenticate using Windows
> Authentication.
> At a minimum put a server certificates on the Publisher and update the
> Trusted Root Authority
> on the Subscriber (MSDE) and use SSL.
> 316898 HOW TO: Enable SSL Encryption for SQL Server 2000 with Microsoft
> http://support.microsoft.com/?id=316898
> 276553 HOW TO: Enable SSL Encryption for SQL Server 2000 with Certificate
> Server
> http://support.microsoft.com/?id=276553
> If you have an ISA Server, you could publish the SQL resource as well.
> Here's some docs on using ISA.
> http://www.microsoft.com/technet/pr...tain/proxy.mspx
> Thanks,
> Kevin McDonnell
> Microsoft Corporation
> This posting is provided AS IS with no warranties, and confers no rights.
>
>|||Thanks Kevin - I left out some important details. Actually, the client side
(MSDE) would be the publisher. We would like to use replication to backup
the databases with our application for our customers.
"Kevin McDonnell [MSFT]" wrote:
> Either would secure the traffic from MSDE to the Publisher.
> VPN would allow the client to use other corporate resources as well and
> authenticate using Windows
> Authentication.
> At a minimum put a server certificates on the Publisher and update the
> Trusted Root Authority
> on the Subscriber (MSDE) and use SSL.
> 316898 HOW TO: Enable SSL Encryption for SQL Server 2000 with Microsoft
> http://support.microsoft.com/?id=316898
> 276553 HOW TO: Enable SSL Encryption for SQL Server 2000 with Certificate
> Server
> http://support.microsoft.com/?id=276553
> If you have an ISA Server, you could publish the SQL resource as well.
> Here's some docs on using ISA.
> http://www.microsoft.com/technet/pr...tain/proxy.mspx
> Thanks,
> Kevin McDonnell
> Microsoft Corporation
> This posting is provided AS IS with no warranties, and confers no rights.
>
>|||OK. Then the MSDE machine (Publisher) would require the server certificate
then.
Also, SQL Replication is not really a backup strategy... Many customers use
Transactional Replication
to provide a warm standby for their Published database, but you should also
have a good backup strategy in place in case there
are hardware failures, that require complete recovery.
Thanks,
Kevin McDonnell
Microsoft Corporation
This posting is provided AS IS with no warranties, and confers no rights.|||Kevin,
I'm working on a similar situation (trying to implement SSL between 2
MSSQL 2000 boxes for replication).
Is there any way to implement the envryption (SSL) for the connections
used during the replication, other than the "force encryption" options
in the client and/or server network utility?
TIA.
Mike
*** Sent via Developersdex http://www.codecomments.com ***
Don't just participate in USENET...get rewarded for it!|||Hi Mike,
In SQL 2000 we use protocol encryption to enable SSL. If you are using
Default instances of SQL for Replication you can also use the older network
library called MultiProtocol.
164667 INF: Replication Setup Over a Firewall
http://support.microsoft.com/?id=164667
Other options are: IPSec between the two machines or a VPN connection
Thanks,
Kevin McDonnell
Microsoft Corporation
This posting is provided AS IS with no warranties, and confers no rights.
Showing posts with label user. Show all posts
Showing posts with label user. Show all posts
Monday, March 26, 2012
Monday, March 12, 2012
Replication or Update via Trigger ?
Hi SQL Gurus,
I have an application where the end user insist that there should be nine(9)
databases, same server within the system.
There are 'shared tables' on one of the database where any
insert/update/delete on those 'shared tables' must appear immediately on
other databases.
In this case, should I use replication or doing updates via Trigger ?
Thanks for your comments,
KristI am not sure why he/she is insisting for identical databases? Perhaps,
instead of creating tables in all these databases, you can opt to create
views which simply SELECT from a table in a single a database. You don't
have to worry about the data being in sync either.
Replication is not a solution for addressing such requirements. You can opt
for a trigger, but with the information from your post, I rather would
conclude it is an overkill to support redundant data for no apparent
reasons.
--
- Anith
( Please reply to newsgroups only )|||Hi Anith,
I used to think of View.
But I need to create FK from transaction table to these 'shared tables'
e.g : SalesOrder must have FK to Salesman and Area table, where Salesman
and Area are 'shared tables'
How can I do this with view ?
Thanks,
Krist
"Anith Sen" <anith@.bizdatasolutions.com> wrote in message
news:#uM6uAfxDHA.1680@.TK2MSFTNGP12.phx.gbl...
> I am not sure why he/she is insisting for identical databases? Perhaps,
> instead of creating tables in all these databases, you can opt to create
> views which simply SELECT from a table in a single a database. You don't
> have to worry about the data being in sync either.
> Replication is not a solution for addressing such requirements. You can
opt
> for a trigger, but with the information from your post, I rather would
> conclude it is an overkill to support redundant data for no apparent
> reasons.
> --
> - Anith
> ( Please reply to newsgroups only )
>|||Krist, first find out the reason that your user needs 9 databases.
1. If is for security then perhaps create 9 different users or roles instead
and assign permissions accordingly.
2. If it's a limitation in the front end application then perhaps you'll
need to stick with several databases
3. If the user later wants to distribute the 9 databases to 9 different
servers then cross-database views become more tricky.
It's a rather strange requirement, and the underlying reason for it probably
constrains your options even further.
"tristant" <krislioe@.cbn.net.id> wrote in message
news:eAJAYgfxDHA.4064@.tk2msftngp13.phx.gbl...
> Hi Anith,
> I used to think of View.
> But I need to create FK from transaction table to these 'shared tables'
> e.g : SalesOrder must have FK to Salesman and Area table, where Salesman
> and Area are 'shared tables'
> How can I do this with view ?
> Thanks,
> Krist
> "Anith Sen" <anith@.bizdatasolutions.com> wrote in message
> news:#uM6uAfxDHA.1680@.TK2MSFTNGP12.phx.gbl...
> > I am not sure why he/she is insisting for identical databases? Perhaps,
> > instead of creating tables in all these databases, you can opt to create
> > views which simply SELECT from a table in a single a database. You don't
> > have to worry about the data being in sync either.
> >
> > Replication is not a solution for addressing such requirements. You can
> opt
> > for a trigger, but with the information from your post, I rather would
> > conclude it is an overkill to support redundant data for no apparent
> > reasons.
> >
> > --
> > - Anith
> > ( Please reply to newsgroups only )
> >
> >
>|||Hi Anthony,
You are absolutely right : the reason the number (3)
> 3. If the user later wants to distribute the 9 databases to 9 different
> servers then cross-database views become more tricky.
So, that is my reason, what options do I have now ?
Thanks,
Krist
"Anthony Faull" <anthony.faull@.NOSPAMsanlam.co.za> wrote in message
news:egFyt5fxDHA.536@.tk2msftngp13.phx.gbl...
> Krist, first find out the reason that your user needs 9 databases.
> 1. If is for security then perhaps create 9 different users or roles
instead
> and assign permissions accordingly.
> 2. If it's a limitation in the front end application then perhaps you'll
> need to stick with several databases
> 3. If the user later wants to distribute the 9 databases to 9 different
> servers then cross-database views become more tricky.
> It's a rather strange requirement, and the underlying reason for it
probably
> constrains your options even further.
> "tristant" <krislioe@.cbn.net.id> wrote in message
> news:eAJAYgfxDHA.4064@.tk2msftngp13.phx.gbl...
> > Hi Anith,
> > I used to think of View.
> > But I need to create FK from transaction table to these 'shared tables'
> > e.g : SalesOrder must have FK to Salesman and Area table, where
Salesman
> > and Area are 'shared tables'
> >
> > How can I do this with view ?
> >
> > Thanks,
> > Krist
> >
> > "Anith Sen" <anith@.bizdatasolutions.com> wrote in message
> > news:#uM6uAfxDHA.1680@.TK2MSFTNGP12.phx.gbl...
> > > I am not sure why he/she is insisting for identical databases?
Perhaps,
> > > instead of creating tables in all these databases, you can opt to
create
> > > views which simply SELECT from a table in a single a database. You
don't
> > > have to worry about the data being in sync either.
> > >
> > > Replication is not a solution for addressing such requirements. You
can
> > opt
> > > for a trigger, but with the information from your post, I rather would
> > > conclude it is an overkill to support redundant data for no apparent
> > > reasons.
> > >
> > > --
> > > - Anith
> > > ( Please reply to newsgroups only )
> > >
> > >
> >
> >
>|||Your options:
1. Partitioned views
2. Replication
3. Log shipping (backup & restore)
"tristant" <krislioe@.cbn.net.id> wrote in message
news:OQxaWDgxDHA.1760@.TK2MSFTNGP10.phx.gbl...
> Hi Anthony,
> You are absolutely right : the reason the number (3)
> > 3. If the user later wants to distribute the 9 databases to 9 different
> > servers then cross-database views become more tricky.
> So, that is my reason, what options do I have now ?
> Thanks,
> Krist
> "Anthony Faull" <anthony.faull@.NOSPAMsanlam.co.za> wrote in message
> news:egFyt5fxDHA.536@.tk2msftngp13.phx.gbl...
> > Krist, first find out the reason that your user needs 9 databases.
> > 1. If is for security then perhaps create 9 different users or roles
> instead
> > and assign permissions accordingly.
> > 2. If it's a limitation in the front end application then perhaps you'll
> > need to stick with several databases
> > 3. If the user later wants to distribute the 9 databases to 9 different
> > servers then cross-database views become more tricky.
> >
> > It's a rather strange requirement, and the underlying reason for it
> probably
> > constrains your options even further.
> >
> > "tristant" <krislioe@.cbn.net.id> wrote in message
> > news:eAJAYgfxDHA.4064@.tk2msftngp13.phx.gbl...
> > > Hi Anith,
> > > I used to think of View.
> > > But I need to create FK from transaction table to these 'shared
tables'
> > > e.g : SalesOrder must have FK to Salesman and Area table, where
> Salesman
> > > and Area are 'shared tables'
> > >
> > > How can I do this with view ?
> > >
> > > Thanks,
> > > Krist
> > >
> > > "Anith Sen" <anith@.bizdatasolutions.com> wrote in message
> > > news:#uM6uAfxDHA.1680@.TK2MSFTNGP12.phx.gbl...
> > > > I am not sure why he/she is insisting for identical databases?
> Perhaps,
> > > > instead of creating tables in all these databases, you can opt to
> create
> > > > views which simply SELECT from a table in a single a database. You
> don't
> > > > have to worry about the data being in sync either.
> > > >
> > > > Replication is not a solution for addressing such requirements. You
> can
> > > opt
> > > > for a trigger, but with the information from your post, I rather
would
> > > > conclude it is an overkill to support redundant data for no apparent
> > > > reasons.
> > > >
> > > > --
> > > > - Anith
> > > > ( Please reply to newsgroups only )
> > > >
> > > >
> > >
> > >
> >
> >
>
I have an application where the end user insist that there should be nine(9)
databases, same server within the system.
There are 'shared tables' on one of the database where any
insert/update/delete on those 'shared tables' must appear immediately on
other databases.
In this case, should I use replication or doing updates via Trigger ?
Thanks for your comments,
KristI am not sure why he/she is insisting for identical databases? Perhaps,
instead of creating tables in all these databases, you can opt to create
views which simply SELECT from a table in a single a database. You don't
have to worry about the data being in sync either.
Replication is not a solution for addressing such requirements. You can opt
for a trigger, but with the information from your post, I rather would
conclude it is an overkill to support redundant data for no apparent
reasons.
--
- Anith
( Please reply to newsgroups only )|||Hi Anith,
I used to think of View.
But I need to create FK from transaction table to these 'shared tables'
e.g : SalesOrder must have FK to Salesman and Area table, where Salesman
and Area are 'shared tables'
How can I do this with view ?
Thanks,
Krist
"Anith Sen" <anith@.bizdatasolutions.com> wrote in message
news:#uM6uAfxDHA.1680@.TK2MSFTNGP12.phx.gbl...
> I am not sure why he/she is insisting for identical databases? Perhaps,
> instead of creating tables in all these databases, you can opt to create
> views which simply SELECT from a table in a single a database. You don't
> have to worry about the data being in sync either.
> Replication is not a solution for addressing such requirements. You can
opt
> for a trigger, but with the information from your post, I rather would
> conclude it is an overkill to support redundant data for no apparent
> reasons.
> --
> - Anith
> ( Please reply to newsgroups only )
>|||Krist, first find out the reason that your user needs 9 databases.
1. If is for security then perhaps create 9 different users or roles instead
and assign permissions accordingly.
2. If it's a limitation in the front end application then perhaps you'll
need to stick with several databases
3. If the user later wants to distribute the 9 databases to 9 different
servers then cross-database views become more tricky.
It's a rather strange requirement, and the underlying reason for it probably
constrains your options even further.
"tristant" <krislioe@.cbn.net.id> wrote in message
news:eAJAYgfxDHA.4064@.tk2msftngp13.phx.gbl...
> Hi Anith,
> I used to think of View.
> But I need to create FK from transaction table to these 'shared tables'
> e.g : SalesOrder must have FK to Salesman and Area table, where Salesman
> and Area are 'shared tables'
> How can I do this with view ?
> Thanks,
> Krist
> "Anith Sen" <anith@.bizdatasolutions.com> wrote in message
> news:#uM6uAfxDHA.1680@.TK2MSFTNGP12.phx.gbl...
> > I am not sure why he/she is insisting for identical databases? Perhaps,
> > instead of creating tables in all these databases, you can opt to create
> > views which simply SELECT from a table in a single a database. You don't
> > have to worry about the data being in sync either.
> >
> > Replication is not a solution for addressing such requirements. You can
> opt
> > for a trigger, but with the information from your post, I rather would
> > conclude it is an overkill to support redundant data for no apparent
> > reasons.
> >
> > --
> > - Anith
> > ( Please reply to newsgroups only )
> >
> >
>|||Hi Anthony,
You are absolutely right : the reason the number (3)
> 3. If the user later wants to distribute the 9 databases to 9 different
> servers then cross-database views become more tricky.
So, that is my reason, what options do I have now ?
Thanks,
Krist
"Anthony Faull" <anthony.faull@.NOSPAMsanlam.co.za> wrote in message
news:egFyt5fxDHA.536@.tk2msftngp13.phx.gbl...
> Krist, first find out the reason that your user needs 9 databases.
> 1. If is for security then perhaps create 9 different users or roles
instead
> and assign permissions accordingly.
> 2. If it's a limitation in the front end application then perhaps you'll
> need to stick with several databases
> 3. If the user later wants to distribute the 9 databases to 9 different
> servers then cross-database views become more tricky.
> It's a rather strange requirement, and the underlying reason for it
probably
> constrains your options even further.
> "tristant" <krislioe@.cbn.net.id> wrote in message
> news:eAJAYgfxDHA.4064@.tk2msftngp13.phx.gbl...
> > Hi Anith,
> > I used to think of View.
> > But I need to create FK from transaction table to these 'shared tables'
> > e.g : SalesOrder must have FK to Salesman and Area table, where
Salesman
> > and Area are 'shared tables'
> >
> > How can I do this with view ?
> >
> > Thanks,
> > Krist
> >
> > "Anith Sen" <anith@.bizdatasolutions.com> wrote in message
> > news:#uM6uAfxDHA.1680@.TK2MSFTNGP12.phx.gbl...
> > > I am not sure why he/she is insisting for identical databases?
Perhaps,
> > > instead of creating tables in all these databases, you can opt to
create
> > > views which simply SELECT from a table in a single a database. You
don't
> > > have to worry about the data being in sync either.
> > >
> > > Replication is not a solution for addressing such requirements. You
can
> > opt
> > > for a trigger, but with the information from your post, I rather would
> > > conclude it is an overkill to support redundant data for no apparent
> > > reasons.
> > >
> > > --
> > > - Anith
> > > ( Please reply to newsgroups only )
> > >
> > >
> >
> >
>|||Your options:
1. Partitioned views
2. Replication
3. Log shipping (backup & restore)
"tristant" <krislioe@.cbn.net.id> wrote in message
news:OQxaWDgxDHA.1760@.TK2MSFTNGP10.phx.gbl...
> Hi Anthony,
> You are absolutely right : the reason the number (3)
> > 3. If the user later wants to distribute the 9 databases to 9 different
> > servers then cross-database views become more tricky.
> So, that is my reason, what options do I have now ?
> Thanks,
> Krist
> "Anthony Faull" <anthony.faull@.NOSPAMsanlam.co.za> wrote in message
> news:egFyt5fxDHA.536@.tk2msftngp13.phx.gbl...
> > Krist, first find out the reason that your user needs 9 databases.
> > 1. If is for security then perhaps create 9 different users or roles
> instead
> > and assign permissions accordingly.
> > 2. If it's a limitation in the front end application then perhaps you'll
> > need to stick with several databases
> > 3. If the user later wants to distribute the 9 databases to 9 different
> > servers then cross-database views become more tricky.
> >
> > It's a rather strange requirement, and the underlying reason for it
> probably
> > constrains your options even further.
> >
> > "tristant" <krislioe@.cbn.net.id> wrote in message
> > news:eAJAYgfxDHA.4064@.tk2msftngp13.phx.gbl...
> > > Hi Anith,
> > > I used to think of View.
> > > But I need to create FK from transaction table to these 'shared
tables'
> > > e.g : SalesOrder must have FK to Salesman and Area table, where
> Salesman
> > > and Area are 'shared tables'
> > >
> > > How can I do this with view ?
> > >
> > > Thanks,
> > > Krist
> > >
> > > "Anith Sen" <anith@.bizdatasolutions.com> wrote in message
> > > news:#uM6uAfxDHA.1680@.TK2MSFTNGP12.phx.gbl...
> > > > I am not sure why he/she is insisting for identical databases?
> Perhaps,
> > > > instead of creating tables in all these databases, you can opt to
> create
> > > > views which simply SELECT from a table in a single a database. You
> don't
> > > > have to worry about the data being in sync either.
> > > >
> > > > Replication is not a solution for addressing such requirements. You
> can
> > > opt
> > > > for a trigger, but with the information from your post, I rather
would
> > > > conclude it is an overkill to support redundant data for no apparent
> > > > reasons.
> > > >
> > > > --
> > > > - Anith
> > > > ( Please reply to newsgroups only )
> > > >
> > > >
> > >
> > >
> >
> >
>
Wednesday, March 7, 2012
Replication Monitor SqlServerCE
I have developed a smart device application using SqlServerCE that replicates data to the backend server. I need to provide to the user a status of this replication process. I read about the "MergeSubscriberMonitor" in SQL Server but can't find it in SqlServerCe. Is this possible? Can someone please point me in the right direction?You need to wait for synchronisation to finish and then you can query to MergeReplication object to report on the data transferred.
If you were using SQLMobile (the CE version for SQL 2005), there is messaging available that will give you progress updates.|||I'm not sure I understand what your comment "messaging" is referring to. Do you have or can point me to some sample code that uses messaging? I am using SQL CE on the mobile device.|||Sorry, by messaging I mean events that you can catch within code and then react to. With SQL CE I don't believe this is happening. You can only query the replication object at the completion of the synchronisation process in order to determine what the outcome was (data sent and received)|||Thanks for your response. I did try asynchronous replication and it is working well so far. I need to implement some additional error trapping but I think I have 95% of what I need. It allows me to get a "percent of completion" during the sync process. I can then display this on the screen to my users. The database is VERY large and was taking forever to sync. I really needed some type of progress bar to show the user. Without that status update, it appeared the unit was locked up. Thanks anyway for your reponse.|||Hi,
I am also trying to get a status update on each article during the web synchronization process. Can you please update me on how to monitor the percent complete progress for the replication.
Thanks in advance.
Apurva
Replication Monitor SqlServerCE
I have developed a smart device application using SqlServerCE that replicates data to the backend server. I need to provide to the user a status of this replication process. I read about the "MergeSubscriberMonitor" in SQL Server but can't find it in SqlServerCe. Is this possible? Can someone please point me in the right direction?You need to wait for synchronisation to finish and then you can query to MergeReplication object to report on the data transferred.
If you were using SQLMobile (the CE version for SQL 2005), there is messaging available that will give you progress updates.|||I'm not sure I understand what your comment "messaging" is referring to. Do you have or can point me to some sample code that uses messaging? I am using SQL CE on the mobile device.|||Sorry, by messaging I mean events that you can catch within code and then react to. With SQL CE I don't believe this is happening. You can only query the replication object at the completion of the synchronisation process in order to determine what the outcome was (data sent and received)|||Thanks for your response. I did try asynchronous replication and it is working well so far. I need to implement some additional error trapping but I think I have 95% of what I need. It allows me to get a "percent of completion" during the sync process. I can then display this on the screen to my users. The database is VERY large and was taking forever to sync. I really needed some type of progress bar to show the user. Without that status update, it appeared the unit was locked up. Thanks anyway for your reponse.|||Hi,
I am also trying to get a status update on each article during the web synchronization process. Can you please update me on how to monitor the percent complete progress for the replication.
Thanks in advance.
Apurva
Replication Monitor
Hi !
Now I have SQL Server 2000. Which could be the minimun
privilegies (roles for server) for a user that only need
to review the Replication Monitor ? It doesn't need to
create, eliminate and modify a Publication or Suscription.
In SQL 7.0, I had to use System Administrator Role, but is
was so dangerous (it was a SA), I would know if SQL 2000
have something better about this.
Thanks in advance
Salvador
sdlreyes@.hotmail.comFrom SQL Books online;
users that are not members of the sysadmin fixed server role can monitor
replication if they are defined with the replmonitor role in the
distribution database. A system administrator can add any user to the
replmonitor role, which allows that user to view replication activity with
the Replication Monitor node in SQL Server Enterprise Manager; however, the
user will not be able to administer replication.
Thanks,
Kevin McDonnell
Microsoft Corporation
This posting is provided AS IS with no warranties, and confers no rights.|||If it's for the user, I wouldn't let them use Replicaiton
Monitor at all. They are mostly concerned with the latency
of their data and whether or not replication is working
end to end.
You can add a dummy single-row table to your publication,
and regularly update that table on the publisher (say once
every minute). Make sure that you timestamp the datetime
the row is updated on the publisher and timestampt when
the row is inserted on the subscriber. Expose the table to
the users on both the publisher and the subscriber and let
them see the time it takes to replicate that row from the
publisher to the subscriber.
This will save you a lot of time explaining to the user
what Replication Monitor reports.
Linchi
is
Is Perfect to me!!!
role can monitor
in the
user to the
replication activity with
Manager; however, the
confers no rights.
Now I have SQL Server 2000. Which could be the minimun
privilegies (roles for server) for a user that only need
to review the Replication Monitor ? It doesn't need to
create, eliminate and modify a Publication or Suscription.
In SQL 7.0, I had to use System Administrator Role, but is
was so dangerous (it was a SA), I would know if SQL 2000
have something better about this.
Thanks in advance
Salvador
sdlreyes@.hotmail.comFrom SQL Books online;
users that are not members of the sysadmin fixed server role can monitor
replication if they are defined with the replmonitor role in the
distribution database. A system administrator can add any user to the
replmonitor role, which allows that user to view replication activity with
the Replication Monitor node in SQL Server Enterprise Manager; however, the
user will not be able to administer replication.
Thanks,
Kevin McDonnell
Microsoft Corporation
This posting is provided AS IS with no warranties, and confers no rights.|||If it's for the user, I wouldn't let them use Replicaiton
Monitor at all. They are mostly concerned with the latency
of their data and whether or not replication is working
end to end.
You can add a dummy single-row table to your publication,
and regularly update that table on the publisher (say once
every minute). Make sure that you timestamp the datetime
the row is updated on the publisher and timestampt when
the row is inserted on the subscriber. Expose the table to
the users on both the publisher and the subscriber and let
them see the time it takes to replicate that row from the
publisher to the subscriber.
This will save you a lot of time explaining to the user
what Replication Monitor reports.
Linchi
quote:
>--Original Message--
>Hi !
>Now I have SQL Server 2000. Which could be the minimun
>privilegies (roles for server) for a user that only need
>to review the Replication Monitor ? It doesn't need to
>create, eliminate and modify a Publication or Suscription.
>In SQL 7.0, I had to use System Administrator Role, but
is
quote:|||Thank you so much !!!
>was so dangerous (it was a SA), I would know if SQL 2000
>have something better about this.
>Thanks in advance
>Salvador
>sdlreyes@.hotmail.com
>.
>
Is Perfect to me!!!
quote:
>--Original Message--
>From SQL Books online;
>users that are not members of the sysadmin fixed server
role can monitor
quote:
>replication if they are defined with the replmonitor role
in the
quote:
>distribution database. A system administrator can add any
user to the
quote:
>replmonitor role, which allows that user to view
replication activity with
quote:
>the Replication Monitor node in SQL Server Enterprise
Manager; however, the
quote:
>user will not be able to administer replication.
>
>Thanks,
>Kevin McDonnell
>Microsoft Corporation
>This posting is provided AS IS with no warranties, and
confers no rights.
quote:
>
>.
>
Replication Model
Hello,
I am just playing about with replication and trying to work the best way
to keep a central database updated.
We have a laptop user who will need to be able to work in disconnected mode
(MSDE) then connect up and replicate changes back to a server from which
others can run reports. The server copy will never be updated besides from
the changes replicated down from the 1 laptop user.
What would be the best model for this? Seems I have a few options but was
hoping someone could point me in the right direction of which machine would
be the publisher, the type (merge, snapshot) and whether push or pull. I
understand the concepts of publisher, subscriber etc, but am just unsure of
how it all fits together.
Thanks is advance
- EdHi
It sounds like your application would probably be a candidate for merge
replication, especially if there is a future need for more than one client
updating the data and if you are replicating larger volumes of data.
Snapshot replication would be more applicable if the data was not updatable
and the frequency of replication low.
You don't say which version of SQL Server is running on the server, this may
effect how you administer the replication and you may want to check out the
topic "Understanding SQL Server 2000 Desktop Engine (MSDE 2000)" in Books
Online.
John
"Ed." wrote:
> Hello,
> I am just playing about with replication and trying to work the best way
> to keep a central database updated.
> We have a laptop user who will need to be able to work in disconnected mode
> (MSDE) then connect up and replicate changes back to a server from which
> others can run reports. The server copy will never be updated besides from
> the changes replicated down from the 1 laptop user.
> What would be the best model for this? Seems I have a few options but was
> hoping someone could point me in the right direction of which machine would
> be the publisher, the type (merge, snapshot) and whether push or pull. I
> understand the concepts of publisher, subscriber etc, but am just unsure of
> how it all fits together.
> Thanks is advance
> - Ed
>
>
I am just playing about with replication and trying to work the best way
to keep a central database updated.
We have a laptop user who will need to be able to work in disconnected mode
(MSDE) then connect up and replicate changes back to a server from which
others can run reports. The server copy will never be updated besides from
the changes replicated down from the 1 laptop user.
What would be the best model for this? Seems I have a few options but was
hoping someone could point me in the right direction of which machine would
be the publisher, the type (merge, snapshot) and whether push or pull. I
understand the concepts of publisher, subscriber etc, but am just unsure of
how it all fits together.
Thanks is advance
- EdHi
It sounds like your application would probably be a candidate for merge
replication, especially if there is a future need for more than one client
updating the data and if you are replicating larger volumes of data.
Snapshot replication would be more applicable if the data was not updatable
and the frequency of replication low.
You don't say which version of SQL Server is running on the server, this may
effect how you administer the replication and you may want to check out the
topic "Understanding SQL Server 2000 Desktop Engine (MSDE 2000)" in Books
Online.
John
"Ed." wrote:
> Hello,
> I am just playing about with replication and trying to work the best way
> to keep a central database updated.
> We have a laptop user who will need to be able to work in disconnected mode
> (MSDE) then connect up and replicate changes back to a server from which
> others can run reports. The server copy will never be updated besides from
> the changes replicated down from the 1 laptop user.
> What would be the best model for this? Seems I have a few options but was
> hoping someone could point me in the right direction of which machine would
> be the publisher, the type (merge, snapshot) and whether push or pull. I
> understand the concepts of publisher, subscriber etc, but am just unsure of
> how it all fits together.
> Thanks is advance
> - Ed
>
>
Replication Model
Hello,
I am just playing about with replication and trying to work the best way
to keep a central database updated.
We have a laptop user who will need to be able to work in disconnected mode
(MSDE) then connect up and replicate changes back to a server from which
others can run reports. The server copy will never be updated besides from
the changes replicated down from the 1 laptop user.
What would be the best model for this? Seems I have a few options but was
hoping someone could point me in the right direction of which machine would
be the publisher, the type (merge, snapshot) and whether push or pull. I
understand the concepts of publisher, subscriber etc, but am just unsure of
how it all fits together.
Thanks is advance
- Ed
Hi
It sounds like your application would probably be a candidate for merge
replication, especially if there is a future need for more than one client
updating the data and if you are replicating larger volumes of data.
Snapshot replication would be more applicable if the data was not updatable
and the frequency of replication low.
You don't say which version of SQL Server is running on the server, this may
effect how you administer the replication and you may want to check out the
topic "Understanding SQL Server 2000 Desktop Engine (MSDE 2000)" in Books
Online.
John
"Ed." wrote:
> Hello,
> I am just playing about with replication and trying to work the best way
> to keep a central database updated.
> We have a laptop user who will need to be able to work in disconnected mode
> (MSDE) then connect up and replicate changes back to a server from which
> others can run reports. The server copy will never be updated besides from
> the changes replicated down from the 1 laptop user.
> What would be the best model for this? Seems I have a few options but was
> hoping someone could point me in the right direction of which machine would
> be the publisher, the type (merge, snapshot) and whether push or pull. I
> understand the concepts of publisher, subscriber etc, but am just unsure of
> how it all fits together.
> Thanks is advance
> - Ed
>
>
I am just playing about with replication and trying to work the best way
to keep a central database updated.
We have a laptop user who will need to be able to work in disconnected mode
(MSDE) then connect up and replicate changes back to a server from which
others can run reports. The server copy will never be updated besides from
the changes replicated down from the 1 laptop user.
What would be the best model for this? Seems I have a few options but was
hoping someone could point me in the right direction of which machine would
be the publisher, the type (merge, snapshot) and whether push or pull. I
understand the concepts of publisher, subscriber etc, but am just unsure of
how it all fits together.
Thanks is advance
- Ed
Hi
It sounds like your application would probably be a candidate for merge
replication, especially if there is a future need for more than one client
updating the data and if you are replicating larger volumes of data.
Snapshot replication would be more applicable if the data was not updatable
and the frequency of replication low.
You don't say which version of SQL Server is running on the server, this may
effect how you administer the replication and you may want to check out the
topic "Understanding SQL Server 2000 Desktop Engine (MSDE 2000)" in Books
Online.
John
"Ed." wrote:
> Hello,
> I am just playing about with replication and trying to work the best way
> to keep a central database updated.
> We have a laptop user who will need to be able to work in disconnected mode
> (MSDE) then connect up and replicate changes back to a server from which
> others can run reports. The server copy will never be updated besides from
> the changes replicated down from the 1 laptop user.
> What would be the best model for this? Seems I have a few options but was
> hoping someone could point me in the right direction of which machine would
> be the publisher, the type (merge, snapshot) and whether push or pull. I
> understand the concepts of publisher, subscriber etc, but am just unsure of
> how it all fits together.
> Thanks is advance
> - Ed
>
>
Replication Model
Hello,
I am just playing about with replication and trying to work the best way
to keep a central database updated.
We have a laptop user who will need to be able to work in disconnected mode
(MSDE) then connect up and replicate changes back to a server from which
others can run reports. The server copy will never be updated besides from
the changes replicated down from the 1 laptop user.
What would be the best model for this? Seems I have a few options but was
hoping someone could point me in the right direction of which machine would
be the publisher, the type (merge, snapshot) and whether push or pull. I
understand the concepts of publisher, subscriber etc, but am just unsure of
how it all fits together.
Thanks is advance
- EdHi
It sounds like your application would probably be a candidate for merge
replication, especially if there is a future need for more than one client
updating the data and if you are replicating larger volumes of data.
Snapshot replication would be more applicable if the data was not updatable
and the frequency of replication low.
You don't say which version of SQL Server is running on the server, this may
effect how you administer the replication and you may want to check out the
topic "Understanding SQL Server 2000 Desktop Engine (MSDE 2000)" in Books
Online.
John
"Ed." wrote:
> Hello,
> I am just playing about with replication and trying to work the best way
> to keep a central database updated.
> We have a laptop user who will need to be able to work in disconnected mod
e
> (MSDE) then connect up and replicate changes back to a server from which
> others can run reports. The server copy will never be updated besides from
> the changes replicated down from the 1 laptop user.
> What would be the best model for this? Seems I have a few options but was
> hoping someone could point me in the right direction of which machine woul
d
> be the publisher, the type (merge, snapshot) and whether push or pull. I
> understand the concepts of publisher, subscriber etc, but am just unsure o
f
> how it all fits together.
> Thanks is advance
> - Ed
>
>
I am just playing about with replication and trying to work the best way
to keep a central database updated.
We have a laptop user who will need to be able to work in disconnected mode
(MSDE) then connect up and replicate changes back to a server from which
others can run reports. The server copy will never be updated besides from
the changes replicated down from the 1 laptop user.
What would be the best model for this? Seems I have a few options but was
hoping someone could point me in the right direction of which machine would
be the publisher, the type (merge, snapshot) and whether push or pull. I
understand the concepts of publisher, subscriber etc, but am just unsure of
how it all fits together.
Thanks is advance
- EdHi
It sounds like your application would probably be a candidate for merge
replication, especially if there is a future need for more than one client
updating the data and if you are replicating larger volumes of data.
Snapshot replication would be more applicable if the data was not updatable
and the frequency of replication low.
You don't say which version of SQL Server is running on the server, this may
effect how you administer the replication and you may want to check out the
topic "Understanding SQL Server 2000 Desktop Engine (MSDE 2000)" in Books
Online.
John
"Ed." wrote:
> Hello,
> I am just playing about with replication and trying to work the best way
> to keep a central database updated.
> We have a laptop user who will need to be able to work in disconnected mod
e
> (MSDE) then connect up and replicate changes back to a server from which
> others can run reports. The server copy will never be updated besides from
> the changes replicated down from the 1 laptop user.
> What would be the best model for this? Seems I have a few options but was
> hoping someone could point me in the right direction of which machine woul
d
> be the publisher, the type (merge, snapshot) and whether push or pull. I
> understand the concepts of publisher, subscriber etc, but am just unsure o
f
> how it all fits together.
> Thanks is advance
> - Ed
>
>
Subscribe to:
Posts (Atom)