Merge rplication - records not being replicated.

We are replicating 5 subscriptions to 500 users with MSDE databases. All
databases are SQL Server 2000 SP4.
The issue we are seeing is that some users are not seeing all the records
they should. (Merge replication) Examples:
User A creates some new records. The records get replicated to the server
and then are not in user A's database (they should still be in the users
database). There are no conflicts. Re-building the snap shot and
re-initializing with the upload changes selected does not fix the issue.
Re-building the snap shot and re-initializing with the upload changes Not
selected does fix the issue.
Any suggestions for tracking this proble down would be appreciated.
Are you using any dynamic filters? If the added rows on A are not allowed
according to the filter, the row is outside of the partition and this could
account for such behaviour.
Cheers,
Paul Ibison SQL Server MVP, www.replicationanswers.com
(recommended sql server 2000 replication book:
http://www.nwsu.com/0974973602p.html)
|||Paul,
We are using dynamic filters. The users are only allowed to add records
that are in the partition.
|||Then this is normal behaviour.
It would help if you explain a little more about what you would like to
happen.
Cheers,
Paul Ibison SQL Server MVP, www.replicationanswers.com
(recommended sql server 2000 replication book:
http://www.nwsu.com/0974973602p.html)

Merge Replication: sp_MSgetmetadatabatch Duration

I'm running SQL 2000 SP4 publisher with approximately 15 MSDE 2000
pull subscribers.
A few of my users are having problems with replication. This is all
done via VPN connections to our site (so they're not on our LAN.)
The merge agent fails with the error "The process could not query row
data at the 'Publisher'."
After running Profiler, it appears that the this is the longest
duration of anything run.
I can't find any real details on this stored procedure, so I'm not
real sure what its doing. I'm also NOT a replication expert. I know
enough to have gotten it running and its been working okay, but these
problems have slowly been creeping up. We can normally get the
synchronization to go through, but its frustrating to my users, and I
may be facing a mutiny! Help! I'll be glad to provide any more details
that anyone requests, and if anyone has any details on what this
stored procedure is doing or how I can find out, that would be most
appreciated!
Try setting querytimeout to a large value or using the slowlink
profile.
On Nov 15, 12:26 pm, dday...@.gmail.com wrote:
> I'm running SQL 2000 SP4 publisher with approximately 15 MSDE 2000
> pull subscribers.
> A few of my users are having problems with replication. This is all
> done via VPN connections to our site (so they're not on our LAN.)
> The merge agent fails with the error "The process could not query row
> data at the 'Publisher'."
> After running Profiler, it appears that the this is the longest
> duration of anything run.
> I can't find any real details on this stored procedure, so I'm not
> real sure what its doing. I'm also NOT a replication expert. I know
> enough to have gotten it running and its been working okay, but these
> problems have slowly been creeping up. We can normally get the
> synchronization to go through, but its frustrating to my users, and I
> may be facing a mutiny! Help! I'll be glad to provide any more details
> that anyone requests, and if anyone has any details on what this
> stored procedure is doing or how I can find out, that would be most
> appreciated!
|||On Nov 15, 1:06 pm, Hilary Cotter <hilary.cot...@.gmail.com> wrote:
> Try setting querytimeout to a large value or using the slowlink
> profile.
> On Nov 15, 12:26 pm, dday...@.gmail.com wrote:
>
>
>
>
> - Show quoted text -
I've done both of those (Set QueryTimeout = 6000) and it still occurs.
|||Index fragmentation could be cuase, you should be rebuilding your merge
system table indexes on a regular basis:
DBCC DBREINDEX (MSmerge_contents, '', 80)
DBCC DBREINDEX (MSmerge_genhistory, '', 80)
DBCC DBREINDEX (MSmerge_tombstone, '', 80)
DBCC DBREINDEX (MSmerge_current_partition_mappings, '', 80)
DBCC DBREINDEX (MSmerge_past_partition_mappings, '', 80)
ChrisB MCDBA
MSSQLConsulting.com
"dday515@.gmail.com" wrote:

> I'm running SQL 2000 SP4 publisher with approximately 15 MSDE 2000
> pull subscribers.
> A few of my users are having problems with replication. This is all
> done via VPN connections to our site (so they're not on our LAN.)
> The merge agent fails with the error "The process could not query row
> data at the 'Publisher'."
> After running Profiler, it appears that the this is the longest
> duration of anything run.
> I can't find any real details on this stored procedure, so I'm not
> real sure what its doing. I'm also NOT a replication expert. I know
> enough to have gotten it running and its been working okay, but these
> problems have slowly been creeping up. We can normally get the
> synchronization to go through, but its frustrating to my users, and I
> may be facing a mutiny! Help! I'll be glad to provide any more details
> that anyone requests, and if anyone has any details on what this
> stored procedure is doing or how I can find out, that would be most
> appreciated!
>
|||On Nov 15, 1:48 pm, Chris <Ch...@.discussions.microsoft.com> wrote:
> Index fragmentation could be cuase, you should be rebuilding your merge
> system table indexes on a regular basis:
> DBCC DBREINDEX (MSmerge_contents, '', 80)
> DBCC DBREINDEX (MSmerge_genhistory, '', 80)
> DBCC DBREINDEX (MSmerge_tombstone, '', 80)
> DBCC DBREINDEX (MSmerge_current_partition_mappings, '', 80)
> DBCC DBREINDEX (MSmerge_past_partition_mappings, '', 80)
> ChrisB MCDBA
> MSSQLConsulting.com
>
> "dday...@.gmail.com" wrote:
>
>
> - Show quoted text -
I've done a full reindex as well prior, still the same problem!
|||Did you try the slow link profile?
Is it possible also that your network link is going down during your sync?
Can you run a ping -t between the publisher and subscriber to verify that
the link stays up during the sync?
RelevantNoise.com - dedicated to mining blogs for business intelligence.
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com
"Daniel Day" <dday515@.gmail.com> wrote in message
news:3d27eaf9-9105-47e0-96bc-c6afe7763da0@.v4g2000hsf.googlegroups.com...
> On Nov 15, 1:06 pm, Hilary Cotter <hilary.cot...@.gmail.com> wrote:
> I've done both of those (Set QueryTimeout = 6000) and it still occurs.

Merge replication question

We are using merge replication with 600 users over a nationwide wan/lan
network. For the most part it works, but occasionally some changes will not
be merged to the agent. All changes make it to the server, but when users
sync the changes do not come down to them. Someitmes if they do multiple
syncs in succession they will make it but other times not. We do a snapshot
of the database nightly that is used by new users to get the initial
database. We have noticed on one user where they installed the snapshot but
did not immediately sync afterwards (we instruct them to so so that they
will get any changes since the snapsot was created). They sync'ed the next
day after a new snapshot was created.. Changes in the database from the
snapshot downloaded one day to the creation of the new sanpshot will not
sync to the database. Changes after the second snapsot is created will
sync. It appears the the client DB thinks it has the records and does not
need to get them. Any thoughts?
Can you run validation on a couple of your subscribers and let us know the
results? Are you using filtering or does each subscriber receive the same
set of data? What build of SQL Server are your subscribers and publishers
on?
Philip Vaughn
This posting is provided "AS IS" with no warranties, and confers no rights.
"Fast Eddie" <e@.bellsouth.net> wrote in message
news:VjA_d.35353$6g7.31842@.bignews1.bellsouth.net. ..
> We are using merge replication with 600 users over a nationwide wan/lan
> network. For the most part it works, but occasionally some changes will
> not be merged to the agent. All changes make it to the server, but when
> users sync the changes do not come down to them. Someitmes if they do
> multiple syncs in succession they will make it but other times not. We do
> a snapshot of the database nightly that is used by new users to get the
> initial database. We have noticed on one user where they installed the
> snapshot but did not immediately sync afterwards (we instruct them to so
> so that they will get any changes since the snapsot was created). They
> sync'ed the next day after a new snapshot was created.. Changes in the
> database from the snapshot downloaded one day to the creation of the new
> sanpshot will not sync to the database. Changes after the second snapsot
> is created will sync. It appears the the client DB thinks it has the
> records and does not need to get them. Any thoughts?
>
|||There is no filtering, all users get all data. We are using SQLServer 2000.
We confirmed today that the issue is the data lost between the creation of
the snapshots. When a new snapshot is created, data that is the new
snapshot but not in the snapshot downloaded will not syncronize if a sync
was not performed immediately after the snapshot is loaded. Any settings
that would cause this behavior?
lip Vaughn" <philipv@.microsoft.com> wrote in message
news:OSAMikALFHA.3552@.TK2MSFTNGP12.phx.gbl...
> Can you run validation on a couple of your subscribers and let us know the
> results? Are you using filtering or does each subscriber receive the same
> set of data? What build of SQL Server are your subscribers and publishers
> on?
> Philip Vaughn
> This posting is provided "AS IS" with no warranties, and confers no
> rights.
>
> "Fast Eddie" <e@.bellsouth.net> wrote in message
> news:VjA_d.35353$6g7.31842@.bignews1.bellsouth.net. ..
>
|||Can you respond back with the version of your publisher and subscriber(s)?
Philip Vaughn
This posting is provided "as is" with no warranties and confers no rights.
"Fast Eddie" <e@.bellsouth.net> wrote in message
news:X5K_d.38774$c72.11001@.bignews3.bellsouth.net. ..
> There is no filtering, all users get all data. We are using SQLServer
> 2000. We confirmed today that the issue is the data lost between the
> creation of the snapshots. When a new snapshot is created, data that is
> the new snapshot but not in the snapshot downloaded will not syncronize if
> a sync was not performed immediately after the snapshot is loaded. Any
> settings that would cause this behavior?
>
> lip Vaughn" <philipv@.microsoft.com> wrote in message
> news:OSAMikALFHA.3552@.TK2MSFTNGP12.phx.gbl...
>

Merge replication and spids left behind

We are using Merge replication with clients from remote offices (SQL
2000, sp3). Recently, I have had a problem with users who are
replicating, and they shut down their laptops. The connection never
dies, and I end up with major blocking issues related to the
"orphaned" spid. The tables that are blocked are used to filter data
on each client. Since the orphaned spid is blocking, backups will run
forever, and have to be killed, and a SQL management job that
inserts/updates data in these tables has to be killed.

If I kill the spid, it shows a rollback at 0% and the status never
changes. The user has disconnected, and there is really nothing to
roll back. How can I get rid of this spid with out restarting SQL
server, or rebooting my server?

Any help would be greatly appreciated.

Thanks,

Amy Mamarshall@.rhtc.net (Amy M) wrote in message news:<119d3885.0408101552.7fe7cd72@.posting.google.com>...
> We are using Merge replication with clients from remote offices (SQL
> 2000, sp3). Recently, I have had a problem with users who are
> replicating, and they shut down their laptops. The connection never
> dies, and I end up with major blocking issues related to the
> "orphaned" spid. The tables that are blocked are used to filter data
> on each client. Since the orphaned spid is blocking, backups will run
> forever, and have to be killed, and a SQL management job that
> inserts/updates data in these tables has to be killed.
> If I kill the spid, it shows a rollback at 0% and the status never
> changes. The user has disconnected, and there is really nothing to
> roll back. How can I get rid of this spid with out restarting SQL
> server, or rebooting my server?
> Any help would be greatly appreciated.
> Thanks,
> Amy M

This KB article might be useful:

http://support.microsoft.com/defaul...kb;en-us;818552

If this doesn't help, you may want to post in
microsoft.public.sqlserver.replication, as your problem seems to be
quite specific.

Simon

Merge Replication and MSDE 2000

Hello..I am going to implement Merge Replication with MSDE 2000.
please let me know how many users MSDE 2000 can burden the workload. It
will degrade the preformance if more than 5 concurrent user will start
work on it concurrently.
What is the procedure in order to implement merge Replication on MSDE
2000.
Pls help me as soon as possible
Thanks in Advance,
Jayesh Patel
(jayesh@.infoworldpune.com)
its 8 simultaneous workloads. So if you are using connection pooling this
could mean many more users the 8, but it depends on what sort of work they
are doing. Work which takes a long time might mean you get throttling with 8
simultaneous connections. Work which completes in ms and work which takes
advantages of connection pooling could mean many more than 8 simultaneous
users.
To implement replication use any of the tutorials on the web, and run
profiler to capture the commands. Then apply these commands using osql.
Hilary Cotter
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com
"Jayesh" <dba.jayeshpatel@.gmail.com> wrote in message
news:1125318990.896170.122400@.g47g2000cwa.googlegr oups.com...
> Hello..I am going to implement Merge Replication with MSDE 2000.
> please let me know how many users MSDE 2000 can burden the workload. It
> will degrade the preformance if more than 5 concurrent user will start
> work on it concurrently.
> What is the procedure in order to implement merge Replication on MSDE
> 2000.
> Pls help me as soon as possible
> Thanks in Advance,
> Jayesh Patel
> (jayesh@.infoworldpune.com)
>
|||Thanks a lot for reply.
Can you guide me how to implement replication between SQL server and
MSDE ?
We have to setup Merge Replication between three continents.
1) asia
2)america
3) Europe
These 3 continents will be act as publisher and suscriber.
(by merging them)
But within one continent , there might be many more countries.
Country contains variuos city. City level database would be running on
MSDE.
Is it possible to replicate data this city level data with country as
well continent level.
Thanks in Advance,
Jayesh Patel

Merge Replication Agent Process is blocikgn other users

Hi,
I have Merge replication on SQL 2000 since last 6 months.
Recently I have notice that process is blocing other users and than timeouts on its own. Query Timeout is 600.
This started recently and I'm trying to track this down. Any clue or suggestion to this problem I have.
Thanks
Nilay
Nilay,
to find out more information about the cause of your blocking issues you
might like to use the MS scripts:
http://support.microsoft.com/default...;en-us;271509. If it is
simply a matter of having a lot of data modifications being synchronized
then you need to optimize the merge agent. You could:
increase -DownloadGenerationsPerBatch
make sure -MetadataRetentionCleanup is 1 and even run
sp_mergemetadataretentioncleanup manually
run the merge agent more frequently, so as to not accumulate changes
decrease the -PollingInterval if you are running continuously
increase -QueryTimeOut
Alternatively, if you accept a high latency, you might run the merge agent
out of hours.
HTH,
Paul Ibison
|||Thanks This helps