We have a primary and a secondary server setup. THe primary replicates to the secondary but not vice-versa lately. I have the following in the FillDB.log repeatedly. Is there an SQL command I can to “release” this at all"?
A replication lock request for QUESTIONRESULTS (Shared) timed out
Hi HBK, could you help me understand what is going on better?
The FillDB log on the primary server contains information on pulling data from the secondary BES Server while the secondary FillDB log contains information on pulling data from the primary BES Server. Does the replication lock message appears in the primary or secondary server filldb log file?
Do BES Clients report into both servers? How many to each?
Could you post or email me both the FillDB logs?
It is normal to see some replication lock time outs. The server will simply try again later and hopefully complete. I’m just not sure if you get the replication time out on every attempt or not.
Ok, got the logs. There is problems with both servers pulling from each other. The primary pulling data from secondary frequently times out on QUESTIONRESULTS replication. The Secondary pulling from the primary has that problem too but it also has these:
Fri, 10 Jul 2009 21:40:09 -0400 – Replication failed for server ‘SERVER’: Database Error:
Microsoft
ODBC SQL Server Driver
TDS buffer length too large (S1000:35456)
Both of the problems are symptoms of the network being too slow transferring data between the servers. Here are things you can look into to try to help.
Ensure that SQL is configured to use TCP/IP network connections, as apposed to Named Pipes, as these will be fastest over the network. See SQL Network Configuration tools to do this. Also, ensure your DSNs are configured to use TCP/IP correctly as well.
Please let us know if any of these appear to help or not.