I have a user who is Business owner of a database that is transactional.
At times some of here staff will be the leadblocker.
She has requested that she be allowed the abiity to KILL spids that are blocking her remaining users. All have blessed this process and I havve been tasked to create a way for this to be accomplished.
Is this as simple as allowing this user a permission or is it more in depth.Only a system administrator can kill processes.
My suggestion would be to create a login specifically for this purpose (kills by a business owner), and write a small application that actually uses that login to do the kill.
You have a minor problem working out how to authenticate the user, but I wouldn't think that would be too complicated, especially if you use Windows Authentication.
-PatP|||My suggestion would be to find out why the processess are being blocked and actually fix the problem...but that's just me.|||I have to agree with Pat and Blindman here on the right way to approach this. You may end up with this user killing backup jobs, data updates, and other stuff that can be a real pain to both troubleshoot, and fix.
That said, you can accomplish this by adding the user to the processadmin server role. Good luck.|||Thank you all for your suggestions.
I will try to implement. I would attempt to fix the problem however I am just keeping the lights on, on this server. No longer in my sphere of control.|||Thank you all for your suggestions.
I will try to implement. I would attempt to fix the problem however I am just keeping the lights on, on this server. No longer in my sphere of control.Pehaps, given the requirements of this task, that is for the best from your point of view ;)
Showing posts with label requested. Show all posts
Showing posts with label requested. Show all posts
Thursday, February 16, 2012
Sunday, February 12, 2012
All Required Data Is Not Immediately Available - What to do?
What are some acceptable ways to deal with "required" data that is
unavailable at the time a row is added to a table?
Example: Lab tests are requested for a patient. Some results come in right
away and others come days later. All are eventually required in order for
the patient's lab records to be completed. It is unacceptable to the
customer to wait until ALL the values are available; they want to enter the
values AS they become available.
Obviously we cannot place a "NOT NULL" constraint on the columns in order to
allow for required data coming in late. I'd prefer to avoid having
application-level logic take care of this; but I currently don't see a way
around that.
Suggestions?
Thanks!"Fred Mertz" <A@.B.COM> wrote in message
news:%23DOJXeumGHA.856@.TK2MSFTNGP03.phx.gbl...
> What are some acceptable ways to deal with "required" data that is
> unavailable at the time a row is added to a table?
> Example: Lab tests are requested for a patient. Some results come in right
> away and others come days later. All are eventually required in order for
> the patient's lab records to be completed. It is unacceptable to the
> customer to wait until ALL the values are available; they want to enter
> the values AS they become available.
> Obviously we cannot place a "NOT NULL" constraint on the columns in order
> to allow for required data coming in late. I'd prefer to avoid having
> application-level logic take care of this; but I currently don't see a way
> around that.
> Suggestions?
> Thanks!
>
Like this for example:
CREATE TABLE Patients (PatientID INT NOT NULL PRIMARY KEY, PatientLastName
VARCHAR(35) NOT NULL /* ... etc */);
CREATE TABLE PatientTestResults (PatientID INT NOT NULL PRIMARY KEY
REFERENCES Patients (PatientID), Result VARCHAR(20) NOT NULL);
David Portas, SQL Server MVP
Whenever possible please post enough code to reproduce your problem.
Including CREATE TABLE and INSERT statements usually helps.
State what version of SQL Server you are using and specify the content
of any error messages.
SQL Server Books Online:
http://msdn2.microsoft.com/library/ms130214(en-US,SQL.90).aspx
--|||So it sounds as if you want to enforce data validation before a status
can be changed or checked. If the process of changing status is manual
(i.e, a technician clicks a button that "signs off" on the data), you
could use a trigger to verify that data is complete before allowing
that status to be changed.
If the process is automatic (you just want to check if the status is
complete), you could run a check process in your stored procedure to
look for NULL tuples and report that status is Incomplete.
Tossing some ideas around.
Stu
Fred Mertz wrote:
> What are some acceptable ways to deal with "required" data that is
> unavailable at the time a row is added to a table?
> Example: Lab tests are requested for a patient. Some results come in right
> away and others come days later. All are eventually required in order for
> the patient's lab records to be completed. It is unacceptable to the
> customer to wait until ALL the values are available; they want to enter th
e
> values AS they become available.
> Obviously we cannot place a "NOT NULL" constraint on the columns in order
to
> allow for required data coming in late. I'd prefer to avoid having
> application-level logic take care of this; but I currently don't see a way
> around that.
> Suggestions?
> Thanks!|||RE:
<< Tossing some ideas around >>
Just what I was looking for. Your ideas are helpful.|||Just a guess here -REMOVE the NOT NULL constraint?
CREATE a 'holding' table that accepts the incomplete data, and move it to
the 'real' table when complete.
Alternatively, set a default value. If char/varchar datatype default to
'N/A' or whatever makes sense. If numeric datatype, and if zero has
significance, then default to a predetermined 'magic number' that signals
the data is still missing, e.g., 999.
Arnie Rowland, YACE*
"To be successful, your heart must accompany your knowledge."
*Yet Another certification Exam
"Fred Mertz" <A@.B.COM> wrote in message
news:%23DOJXeumGHA.856@.TK2MSFTNGP03.phx.gbl...
> What are some acceptable ways to deal with "required" data that is
> unavailable at the time a row is added to a table?
> Example: Lab tests are requested for a patient. Some results come in right
> away and others come days later. All are eventually required in order for
> the patient's lab records to be completed. It is unacceptable to the
> customer to wait until ALL the values are available; they want to enter
> the values AS they become available.
> Obviously we cannot place a "NOT NULL" constraint on the columns in order
> to allow for required data coming in late. I'd prefer to avoid having
> application-level logic take care of this; but I currently don't see a way
> around that.
> Suggestions?
> Thanks!
>
unavailable at the time a row is added to a table?
Example: Lab tests are requested for a patient. Some results come in right
away and others come days later. All are eventually required in order for
the patient's lab records to be completed. It is unacceptable to the
customer to wait until ALL the values are available; they want to enter the
values AS they become available.
Obviously we cannot place a "NOT NULL" constraint on the columns in order to
allow for required data coming in late. I'd prefer to avoid having
application-level logic take care of this; but I currently don't see a way
around that.
Suggestions?
Thanks!"Fred Mertz" <A@.B.COM> wrote in message
news:%23DOJXeumGHA.856@.TK2MSFTNGP03.phx.gbl...
> What are some acceptable ways to deal with "required" data that is
> unavailable at the time a row is added to a table?
> Example: Lab tests are requested for a patient. Some results come in right
> away and others come days later. All are eventually required in order for
> the patient's lab records to be completed. It is unacceptable to the
> customer to wait until ALL the values are available; they want to enter
> the values AS they become available.
> Obviously we cannot place a "NOT NULL" constraint on the columns in order
> to allow for required data coming in late. I'd prefer to avoid having
> application-level logic take care of this; but I currently don't see a way
> around that.
> Suggestions?
> Thanks!
>
Like this for example:
CREATE TABLE Patients (PatientID INT NOT NULL PRIMARY KEY, PatientLastName
VARCHAR(35) NOT NULL /* ... etc */);
CREATE TABLE PatientTestResults (PatientID INT NOT NULL PRIMARY KEY
REFERENCES Patients (PatientID), Result VARCHAR(20) NOT NULL);
David Portas, SQL Server MVP
Whenever possible please post enough code to reproduce your problem.
Including CREATE TABLE and INSERT statements usually helps.
State what version of SQL Server you are using and specify the content
of any error messages.
SQL Server Books Online:
http://msdn2.microsoft.com/library/ms130214(en-US,SQL.90).aspx
--|||So it sounds as if you want to enforce data validation before a status
can be changed or checked. If the process of changing status is manual
(i.e, a technician clicks a button that "signs off" on the data), you
could use a trigger to verify that data is complete before allowing
that status to be changed.
If the process is automatic (you just want to check if the status is
complete), you could run a check process in your stored procedure to
look for NULL tuples and report that status is Incomplete.
Tossing some ideas around.
Stu
Fred Mertz wrote:
> What are some acceptable ways to deal with "required" data that is
> unavailable at the time a row is added to a table?
> Example: Lab tests are requested for a patient. Some results come in right
> away and others come days later. All are eventually required in order for
> the patient's lab records to be completed. It is unacceptable to the
> customer to wait until ALL the values are available; they want to enter th
e
> values AS they become available.
> Obviously we cannot place a "NOT NULL" constraint on the columns in order
to
> allow for required data coming in late. I'd prefer to avoid having
> application-level logic take care of this; but I currently don't see a way
> around that.
> Suggestions?
> Thanks!|||RE:
<< Tossing some ideas around >>
Just what I was looking for. Your ideas are helpful.|||Just a guess here -REMOVE the NOT NULL constraint?
CREATE a 'holding' table that accepts the incomplete data, and move it to
the 'real' table when complete.
Alternatively, set a default value. If char/varchar datatype default to
'N/A' or whatever makes sense. If numeric datatype, and if zero has
significance, then default to a predetermined 'magic number' that signals
the data is still missing, e.g., 999.
Arnie Rowland, YACE*
"To be successful, your heart must accompany your knowledge."
*Yet Another certification Exam
"Fred Mertz" <A@.B.COM> wrote in message
news:%23DOJXeumGHA.856@.TK2MSFTNGP03.phx.gbl...
> What are some acceptable ways to deal with "required" data that is
> unavailable at the time a row is added to a table?
> Example: Lab tests are requested for a patient. Some results come in right
> away and others come days later. All are eventually required in order for
> the patient's lab records to be completed. It is unacceptable to the
> customer to wait until ALL the values are available; they want to enter
> the values AS they become available.
> Obviously we cannot place a "NOT NULL" constraint on the columns in order
> to allow for required data coming in late. I'd prefer to avoid having
> application-level logic take care of this; but I currently don't see a way
> around that.
> Suggestions?
> Thanks!
>
Labels:
acceptable,
available,
database,
immediately,
isunavailable,
lab,
microsoft,
mysql,
oracle,
requested,
required,
row,
server,
sql,
tableexample,
time
Subscribe to:
Posts (Atom)