| Author |
Topic |
|
Laura Beranek
Starting Member
9 Posts |
Posted - 2007-11-23 : 15:30:54
|
| We have a 6 node cluster with 4 SQL Server named instances in the cluster. I am noticing a lot of warnings in the event log with the following information:Event Type: Warning Event Source: SQLBrowser Event Category: None Event ID: 3 Description: The configuration of the AdminConnection\TCP protocol in the SQL instance <instance name> is not valid. The event occurs for all four instances. Additional Information:* OS is Windows 2003 R2 Enterprise x64 Edition w/SP2 * SQL Server 2005 Enterprise Edition w/SP1, hot fix level 2221* All SQL services runs under the same domain user account I did some internet searches and found several others who have had this problem. However, I didn't see any solutions. There also seems to be a similar problem with SQL Express but that is not what I am using. Any thoughts are appreciated. |
|
|
KenW
Constraint Violating Yak Guru
391 Posts |
Posted - 2007-11-23 : 16:25:37
|
quote: Originally posted by Laura BeranekThe configuration of the AdminConnection\TCP protocol in the SQL instance <instance name> is not valid. /snipAny thoughts are appreciated.
According to the error message, there's a problem with the configuration of the TCP protocol on the SQL Server instance. Your post doesn't mention checking that configuration at all, or indicate what the current values are if you did in fact check them.Did you check the TCP Protocol configuration on any or all of the four instances? If not, do so. If you did, what are the values currently set for the configuration? |
 |
|
|
Laura Beranek
Starting Member
9 Posts |
Posted - 2007-11-26 : 10:21:00
|
| What should I be looking for in the TCP protocol configuration? On the Protocol tab, it says it's enabled, Keep Alive = 30000, Listen All = yes, No Delay = No. On the IP Address tab, it lists 6 IP addresses. All are active, all are not enabled, TCP Dynamic Ports is 0. For the one labled "IPAll", the TCP Dynamic Ports has a value.I'm not entirely sure how to interpret all of this as it relates to SQL Browser.Thanks for anay additional help! |
 |
|
|
jen
Master Smack Fu Yak Hacker
4110 Posts |
Posted - 2007-11-26 : 13:11:52
|
| this means all your nodes should have the same protocols enabled and disabled on them...also check the client configuration on sql server manager and be sure that the protocols are the same as the ones configured on the server side.--------------------keeping it simple... |
 |
|
|
Laura Beranek
Starting Member
9 Posts |
Posted - 2007-11-26 : 14:52:59
|
| All instances have the same protocols enabled and disabled on them. The client configuration on sql server manager matches as much as it can (the only thing that is really consistent is the keep alive value). For the client, the default port is 1433, it is enabled, keep alive is 30000 and keep alive interval is 1000. |
 |
|
|
jen
Master Smack Fu Yak Hacker
4110 Posts |
Posted - 2007-11-26 : 15:16:09
|
| not "much as it can"... it should be all nodes on the cluster must have the same protocols enabled and disabled. also you have to restart the sql browsers to commit the changesthe isalive and keepalive has nothing to do with the error... focus on the sql browser...sql browser is nothing to worry about unless you don't specify port numbers for your data connection filethe only purpose of sql browser is so you don't have to input a port number which means the instances on that machine can be detected and discovered...which you don't want to in any case for security purpose--------------------keeping it simple... |
 |
|
|
Laura Beranek
Starting Member
9 Posts |
Posted - 2007-11-26 : 16:47:51
|
| If everything matches for each instance, then why am I getting these warnings? Also, I haven't made any changes, so I shouldn't need to stop and start SQL Browser.My "much as it can" comment is referring to the client versus server protocols. The properties are not the same, so I wasn't sure what you wanted me to verify was the same.I will focus more on determining if we need SQL Browser enabled.Thanks! |
 |
|
|
Laura Beranek
Starting Member
9 Posts |
Posted - 2007-11-30 : 11:20:39
|
| In case anyone is interested, we determined what was causing this. When you install in a cluster, the application log gets flooded with these warnings and they don't stop until you stop and start the SQL Browser service. Microsoft acknowledges that this is a known issue. We are noticing it a lot right now because we are in the process of installing all of our instances in two different 6 node clusters.Thanks for all advice given. |
 |
|
|
|