Microsoft KB Archive/93554

{|
 * width="100%"|

The information in this article applies to:
 * Microsoft SQL Server version 4.2 for OS/2

BUG# OS/2: 1558 (4.2)

SYMPTOMS
An insert with union into a float (not null) column will produce the error:

Updated or inserted row is bigger than maximum size (10 bytes) allowed for the table. (msg 511, level 16, state 2)&quot; This only happens if the float value is generated from a division involving two float values. For example:

insert into ck_tab select 1.0/1.25 union select b     from testbug The table ck_tab is defined as: create table ck_tab(a float NOT NULL) The table testbug is defined as: create table testbug(b float NOT NULL) Furthermore, the problem does not occur if the float column is defined to accept NULLs.

CAUSE
SQL server does not correctly handle the maximum row size of float column in an insert into statement with union.

WORKAROUND
Change the table definition from NOT NULL to NULL.

STATUS
Microsoft has confirmed this to be a problem in SQL Server version 4.2 for OS/2. We are researching this problem and will post new information here in the Microsoft Knowledge Base as it becomes available.
 * }

-

Last reviewed: April 28, 1997

© 1998 Microsoft Corporation. All rights reserved. Terms of Use.