Read about updates, corrections, and clarifications for Adaptive Server Reference Manual: Procedures.
sp_addobjectdef
The Usage section of the sp_addobjectdef reference page includes this bullet item, which is incomplete:
- Use sp_addobjectdef before issuing any create table or create existing table commands.
Replace with:
- Use sp_addobjectdef before issuing any create table or create existing table commands. However, if a remote table exists, you need not use sp_addobjectdef before executing create proxy_table.
sp_bindexeclass
The description of this stored procedure reads: "Associates an execution class with a client application, login, or stored
procedure."
Update the description, with "Associates an execution class with a client application, login, stored procedure, or default execution class."
Syntax updates
The definitions of the parameters have been updated:
-
object_name – the description should now read, "is the name of the client application, login, or stored procedure to be associated with the execution class,
classname. If object_type is df, it should be null."
-
object_type – the description should now read, "identifies the type of object_name. Use ap for application, lg for login, pr for stored procedure, or df for
user-defined default execution class."
-
scope – the description should now read, "is the name of a client application or login, or it can be null for ap, lg, or df objects. It is the name of the
stored procedure owner (user name) for objects. When the object with object_name interacts with the application
or login, classname attributes apply for the scope you set."
Example
Add this to the Example section:
This statement assigns 'CLASS1' attributes to all tasks that are running with default execution attributes:
sp_bindexeclass NULL, 'DF', NULL, 'CLASS1'
Usage updates
Add this new bullet item:
- When binding an execution class to a default execution class, all tasks running with default execution attributes run with attributes of the new class.
In addition, these two bullet items have been modified, and should now read:
-
sp_bindexeclass associates an execution class with a client application, login, or stored procedure. It can also associate
an execution class to the default execution class. Use sp_addexeclass to create execution classes.
- Due to precedence and scoping rules, the execution class being bound may or may not have been in effect for the object_name. The object automatically binds itself to another execution class, depending on other
binding specifications, precedence, and scoping rules. If no other binding is applicable, the object binds to the
user-defined default execution class. If a user-defined default execution class is not specified, then the object binds
to the system-defined default execution class EC2.
sp_cacheconfig
The Data Cache Memory subsection in the Usage section
of the reference page for
sp_cacheconfig includes
a bullet item that states, in part:
- A data
cache requires a small percentage of overhead for structures that manage
the cache. All cache overhead is taken from the default data cache.
Since you can configure the default data cache size separately
from total memory, the cache overhead is taken from free memory,
not the default data cache. For this reason, the bullet should read:
- A data cache requires a small percentage of overhead
for structures that manage the cache.
All cache overhead
is taken from free memory.
sp_downgrade
The reference page for sp_downgrade should specify that you can execute this stored procedure only in the master database.
sp_dropexeclass
A bullet item in the Usage section reads as follows:
-
classname must not be bound to any client application, login, or stored
procedure. Unbind the execution class first, using sp_unbindexeclass, then
drop the execution class, using sp_dropexeclass.
Replace with:
-
classname must not be bound to any client application, login, stored procedure, or default execution class. Unbind
the execution class first, using sp_unbindexeclass, then drop the execution class, using sp_dropexeclass.
sp_lmconfig
There is a correction to the
sp_lmconfig stored procedure. The syntax in the book appears as:
sp_lmconfig
[ 'edition' [, edition_type ]]
[ , 'license type' [ , license_type_name ]]
[ , 'smtp host' [ , smtp_host_name ]]
[ , 'smtp port' [ , smtp_port_number ]]
[ , 'email sender' [ , sender_email_address ]]
[ , 'email recipients' [ , email_recipients ]]]
[ , 'email severity' [ , email_severity ]]
Replace with:
sp_lmconfig
[
[ 'edition' [, edition_type ]] |
[ 'license type' [, license_type_name ]] |
[ 'smtp host' [, smtp_host_name ]] |
[ 'smtp port' [, smtp_port_number ]] |
[ 'email sender' [, sender_email_address ]] |
[ 'email recipients' [, email_recipients ]] |
[ 'email severity' [, email_severity ]]
]
The sp_lmconfig procedure, when executed without parameters, also displays the server name from where the license
is checked out. Example #1 in the reference page should display the following, with the addition of the column "Server Name."
1> sp_lmconfig
2> go
Parameter Name Config Value
----------------- ------------
edition EE
license type CP
smtp host null
email recipients null
email severity null
smtp port null
email sender null
License Name Version Quantity Status Expiry Date Server Name
--------------- ---------- -------- ---------- ------------------- --------------------
ASE_HA 2010.03314 2 expirable Apr 1 2010 12:00AM cuprum
ASE_ASM null 0 not used null null
ASE_EJB null 0 not used null null
ASE_EFTS null 0 not used null null
ASE_DIRS null 0 not used null null
ASE_XRAY null 0 not used null null
ASE_ENCRYPTION null 0 not used null null
ASE_CORE 2010.03314 2 expirable Apr 1 2010 12:00AM cuprum
ASE_PARTITIONS null 0 not used null null
ASE_RLAC null 0 not used null null
ASE_MESSAGING_TIBJMS null 0 not used null null
ASE_MESSAGING_IBMMQ null 0 not used null null
ASE_MESSAGING_EASJMS null 0 not used null null
Property Name Property Value
------------- --------------
PE EE
LT CP
ME null
MC null
MS null
MM null
CP 0
AS A
(return status = 0)
sp_locklogin
Examples 2 and 4 use all, which sp_locklogin does not recognize . A note in the Usage section correctly states:
Note: A value of NULL for a login means all logins."
Example 2
locks all logins except those with the sa_role. The erroneous syntax is:
sp_locklogin "all", "lock", sa_role
Replace with:
sp_locklogin NULL, "lock", sa_role
Example 4
locks all login accounts that have not authenticated within the past 60 days. The erroneous syntax is:
sp_locklogin 'all', 'lock', NULL, 60
Replace with:
sp_locklogin NULL, 'lock', NULL, 60
sp_reportstats
In the reference page for sp_reportstats,
disregard the following bullet, which no longer applies:
-
sp_reportstats does
not report statistics for any process with a system user ID (suid)
of 0 or 1. This includes deadlock detection, checkpoint, housekeeper,
network, auditing, mirror handlers, and all users with sa_role.
sp_showcontrolinfo
The definitions of the parameters have been updated:
-
object_name – the description should now read, "is the name of the application, login, stored procedure, or engine group. Do not specify an object_name if you
specify ps or DF as the object_type. If you do not specify an object_name (or specify an object_name of null),
sp_showcontrolinfo displays information about all object names."
-
object_type – the description should now read, "is ap for application, lg for login, pr for stored procedure, eg for engine group, ps for process, or df for
user-defined default execution class. If you do not specify an object_type or specify an object_type of null,
sp_showcontrolinfo displays information about all types."
Usage updates
Add these two bullets:
- If object_type is df, sp_showcontrolinfo shows information about the user-defined default execution class any.
- If object_type is df, object_name and spid should be null.
sp_unbindexeclass
The description of this stored procedure reads: "Removes the execution class attribute previously associated with an client
application, login, or stored procedure for the specified scope."
Update the description with: "Removes the execution class attribute previously associated with a client application, login, stored procedure, or
default execution class for the specified scope."
Syntax updates
The definitions of the parameters have been updated:
-
object_name – the description should now read, "is the name of the application, login, or stored procedure for which to remove the association to the execution
class. If object_type is df, object_name should be null."
-
object_type – the description should now read, "identifies the type of object_name as ap, lg, pr, or df for application, login, stored procedure, or default
execution class."
-
scope – the description should now read, "is the application name or the login name for which the unbinding applies for an application or login. For stored procedures, scope is the user name of the stored procedure owner. scope is null for object type df."
Usage updates
Add this bullet item:
- When unbinding from user-defined default execution class, all tasks running with user-defined default execution class attributes run with attributes of the system-defined default execution class EC2.
The following bullet item appears in the Usage section:
- "Due to precedence and scoping rules, the execution class being unbound
may or may not have been in effect for the object called object_name. The
object automatically binds itself to another execution class, depending on
other binding specifications and precedence and scoping rules. If no other
binding is applicable, the object binds to the default execution class, EC2."
The text has been modified to read:
- "Due to precedence and scoping rules, the execution class being unbound may or may not have been in effect for the
object_name. The object automatically binds itself to another execution class, depending on other
binding specifications, and precedence and scoping rules. If no other binding is applicable, the object binds to
the user-defined default execution class. If there is no user-defined default execution class, the object binds to class EC2."