Tags: compute, database, error, errors, microsoft, mysql, nodes, oracle, provider, reproduce, server, servers, sql, tcp, whileconnecting

TCP Provider, error: 0

On Database » Microsoft SQL Server

1,571 words with 0 Comments; publish: Tue, 03 Jun 2008 23:15:00 GMT; (25062.50, « »)

Hello,

I'm doing some testing to reproduce a set of errors we're getting while

connecting to one of our SQL Servers from nodes in a compute cluster.

I've worked my way up to using 24 nodes to connect, run a simple query,

write the results to disk (or, write any exception to disk). At 24 nodes, I'm

starting to get exceptions...as follows:

"A connection was successfully established with the server, but then an

error occurred during the login process. (provider: TCP Provider, error: 0 -

An existing connection was forcibly closed by the remote host.)"

and

"A connection was successfully established with the server, but then an

error occurred during the pre-login handshake. When connecting to SQL Server

2005, this failure may be caused by the fact that under the default settings

SQL Server does not allow remote connections. (provider: TCP Provider, error:

0 - An existing connection was forcibly closed by the remote host.)"

Is this caused by opening 24 connections virtually simultaneously? Or, does

it indicate a network problem? Or, is it unclear? What else should I look for?

As a temporary work-around, we're staggering each node by 1 second (in

production). That seems to eliminate the problem...indicating to me that the

server cannot handle so many connections all at once. If so, what is the

determining factor? Is this is TCP limitation or an SQL Server limitation?

Lots of questions...I'll accept whatever answers folks might have.

All Comments

Leave a comment...

  • 0 Comments