Merge Replication works then fails, works then fails, and so on

I have a merge replication environment with one central
publisher/distributor and 35 clients. The merge uses continuous
replication. At the moment, 31 of the 35 subscriptions have error this
error:
The process could not check existence of generation at the 'Subscriber'.
(Source: Merge Replication Provider (Agent); Error number: -2147201003)
General network error. Check your network documentation.
(Source: SERVER002 (Data source); Error number: 11)
Occasionally some will start working again; and then later fail again. I'm
sure that all permissions must be correct because they work 70 to 80 % of
the time. I don't understand why so many fail so frequently though, and
then why they start working again. Also, there are 2 that NEVER fail with
this message.
Any ideas? I do have a separate schedule to restart all the replication
jobs every hour just in case they are not running - but that is a band-aid.
Thanks,
Rob Kraft
general network error almost always means that the connection has been
broken - its a network failure problem not a problem with SQL Server or
replication.
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
"Rob Kraft" <robkraft@.msn.com> wrote in message
news:%23Q6xWBJ7FHA.2012@.TK2MSFTNGP14.phx.gbl...
>I have a merge replication environment with one central
> publisher/distributor and 35 clients. The merge uses continuous
> replication. At the moment, 31 of the 35 subscriptions have error this
> error:
> The process could not check existence of generation at the 'Subscriber'.
> (Source: Merge Replication Provider (Agent); Error number: -2147201003)
> ----
> --
> General network error. Check your network documentation.
> (Source: SERVER002 (Data source); Error number: 11)
> ----
> --
> Occasionally some will start working again; and then later fail again.
> I'm
> sure that all permissions must be correct because they work 70 to 80 % of
> the time. I don't understand why so many fail so frequently though, and
> then why they start working again. Also, there are 2 that NEVER fail with
> this message.
> Any ideas? I do have a separate schedule to restart all the replication
> jobs every hour just in case they are not running - but that is a
> band-aid.
> Thanks,
> Rob Kraft
>
sql