Showing posts with label requires. Show all posts
Showing posts with label requires. Show all posts

Tuesday, March 20, 2012

Asymmetric communication from ms-sql-m protocol from an SQL Cluster

We are installing an application that requires access to the ms-sql-m
protocol (UDP/1434) as well as the data port (TCP/1433). The SQL Server
we are using is part of an N+1 cluster. The issue is that when we try
to communicate to the node instance xxx.xxx.123.226 recieve the
ms-sql-m response from the physical device ip xxx.xxx.123.222 causing
an asymmetric IP communication and the response appears to be dropped
by the request as one might expect. This causing our installation to
fail.
Has anyone run into this issue before, no of a common misconfiguration
in clustering services that leads to this, or aware of any documented
bug?
Thanks in advance for you posts.
Actually, this is typical of MS cluster applications. The response comes
back from the underlying NIC address, not the cluster virtual address. It
ain't a bug, it's a feature. Or at least it has always operated this way
and could therefore be considered a standard.
Sorry this isn't the answer you were looking for, but it is the way the
system actually works.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
<ddnash@.gmail.com> wrote in message
news:1165524543.657106.172130@.79g2000cws.googlegro ups.com...
> We are installing an application that requires access to the ms-sql-m
> protocol (UDP/1434) as well as the data port (TCP/1433). The SQL Server
> we are using is part of an N+1 cluster. The issue is that when we try
> to communicate to the node instance xxx.xxx.123.226 recieve the
> ms-sql-m response from the physical device ip xxx.xxx.123.222 causing
> an asymmetric IP communication and the response appears to be dropped
> by the request as one might expect. This causing our installation to
> fail.
> Has anyone run into this issue before, no of a common misconfiguration
> in clustering services that leads to this, or aware of any documented
> bug?
>
> Thanks in advance for you posts.
>
|||I found the following article that does acknowledge the issue and
states that MS has chosen not to address it at this point, but there
are a couple of workarounds.
http://blogs.msdn.com/sql_protocols/archive/2006/02/27/539706.aspx
Thanks for the post.
Geoff N. Hiten wrote:[vbcol=seagreen]
> Actually, this is typical of MS cluster applications. The response comes
> back from the underlying NIC address, not the cluster virtual address. It
> ain't a bug, it's a feature. Or at least it has always operated this way
> and could therefore be considered a standard.
> Sorry this isn't the answer you were looking for, but it is the way the
> system actually works.
> --
> Geoff N. Hiten
> Senior Database Administrator
> Microsoft SQL Server MVP
>
> <ddnash@.gmail.com> wrote in message
> news:1165524543.657106.172130@.79g2000cws.googlegro ups.com...
sql

Asymmetric communication from ms-sql-m protocol from an SQL Cluster

We are installing an application that requires access to the ms-sql-m
protocol (UDP/1434) as well as the data port (TCP/1433). The SQL Server
we are using is part of an N+1 cluster. The issue is that when we try
to communicate to the node instance xxx.xxx.123.226 recieve the
ms-sql-m response from the physical device ip xxx.xxx.123.222 causing
an asymmetric IP communication and the response appears to be dropped
by the request as one might expect. This causing our installation to
fail.
Has anyone run into this issue before, no of a common misconfiguration
in clustering services that leads to this, or aware of any documented
bug?
Thanks in advance for you posts.Actually, this is typical of MS cluster applications. The response comes
back from the underlying NIC address, not the cluster virtual address. It
ain't a bug, it's a feature. :) Or at least it has always operated this way
and could therefore be considered a standard.
Sorry this isn't the answer you were looking for, but it is the way the
system actually works.
--
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
<ddnash@.gmail.com> wrote in message
news:1165524543.657106.172130@.79g2000cws.googlegroups.com...
> We are installing an application that requires access to the ms-sql-m
> protocol (UDP/1434) as well as the data port (TCP/1433). The SQL Server
> we are using is part of an N+1 cluster. The issue is that when we try
> to communicate to the node instance xxx.xxx.123.226 recieve the
> ms-sql-m response from the physical device ip xxx.xxx.123.222 causing
> an asymmetric IP communication and the response appears to be dropped
> by the request as one might expect. This causing our installation to
> fail.
> Has anyone run into this issue before, no of a common misconfiguration
> in clustering services that leads to this, or aware of any documented
> bug?
>
> Thanks in advance for you posts.
>|||I found the following article that does acknowledge the issue and
states that MS has chosen not to address it at this point, but there
are a couple of workarounds.
http://blogs.msdn.com/sql_protocols/archive/2006/02/27/539706.aspx
Thanks for the post.
Geoff N. Hiten wrote:
> Actually, this is typical of MS cluster applications. The response comes
> back from the underlying NIC address, not the cluster virtual address. It
> ain't a bug, it's a feature. :) Or at least it has always operated this way
> and could therefore be considered a standard.
> Sorry this isn't the answer you were looking for, but it is the way the
> system actually works.
> --
> Geoff N. Hiten
> Senior Database Administrator
> Microsoft SQL Server MVP
>
> <ddnash@.gmail.com> wrote in message
> news:1165524543.657106.172130@.79g2000cws.googlegroups.com...
> > We are installing an application that requires access to the ms-sql-m
> > protocol (UDP/1434) as well as the data port (TCP/1433). The SQL Server
> >
> > we are using is part of an N+1 cluster. The issue is that when we try
> > to communicate to the node instance xxx.xxx.123.226 recieve the
> > ms-sql-m response from the physical device ip xxx.xxx.123.222 causing
> > an asymmetric IP communication and the response appears to be dropped
> > by the request as one might expect. This causing our installation to
> > fail.
> >
> > Has anyone run into this issue before, no of a common misconfiguration
> > in clustering services that leads to this, or aware of any documented
> > bug?
> >
> >
> > Thanks in advance for you posts.
> >

Asymmetric communication from ms-sql-m protocol from an SQL Cluster

We are installing an application that requires access to the ms-sql-m
protocol (UDP/1434) as well as the data port (TCP/1433). The SQL Server
we are using is part of an N+1 cluster. The issue is that when we try
to communicate to the node instance xxx.xxx.123.226 recieve the
ms-sql-m response from the physical device ip xxx.xxx.123.222 causing
an asymmetric IP communication and the response appears to be dropped
by the request as one might expect. This causing our installation to
fail.
Has anyone run into this issue before, no of a common misconfiguration
in clustering services that leads to this, or aware of any documented
bug?
Thanks in advance for you posts.Actually, this is typical of MS cluster applications. The response comes
back from the underlying NIC address, not the cluster virtual address. It
ain't a bug, it's a feature. Or at least it has always operated this way
and could therefore be considered a standard.
Sorry this isn't the answer you were looking for, but it is the way the
system actually works.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
<ddnash@.gmail.com> wrote in message
news:1165524543.657106.172130@.79g2000cws.googlegroups.com...
> We are installing an application that requires access to the ms-sql-m
> protocol (UDP/1434) as well as the data port (TCP/1433). The SQL Server
> we are using is part of an N+1 cluster. The issue is that when we try
> to communicate to the node instance xxx.xxx.123.226 recieve the
> ms-sql-m response from the physical device ip xxx.xxx.123.222 causing
> an asymmetric IP communication and the response appears to be dropped
> by the request as one might expect. This causing our installation to
> fail.
> Has anyone run into this issue before, no of a common misconfiguration
> in clustering services that leads to this, or aware of any documented
> bug?
>
> Thanks in advance for you posts.
>|||I found the following article that does acknowledge the issue and
states that MS has chosen not to address it at this point, but there
are a couple of workarounds.
http://blogs.msdn.com/sql_protocols.../27/539706.aspx
Thanks for the post.
Geoff N. Hiten wrote:[vbcol=seagreen]
> Actually, this is typical of MS cluster applications. The response comes
> back from the underlying NIC address, not the cluster virtual address. It
> ain't a bug, it's a feature. Or at least it has always operated this w
ay
> and could therefore be considered a standard.
> Sorry this isn't the answer you were looking for, but it is the way the
> system actually works.
> --
> Geoff N. Hiten
> Senior Database Administrator
> Microsoft SQL Server MVP
>
> <ddnash@.gmail.com> wrote in message
> news:1165524543.657106.172130@.79g2000cws.googlegroups.com...

Monday, March 19, 2012

Assistance on implementing Data Mining

Hi,

I'm new to SQL Server and data mining, so please forgive my ignorance...

I'm working on a project which requires me to use the datamining provided by SQL Server 2005. I've a table for which i want to predict the values in a table (Encyclopedia)

The table contains the following fields:

Component

Major Attribute

Minor Attributes(which is basically a list of CSV for attributes in no particular order)

I want to predict the component if i enter the attributes ..... my questions:

1. Should i change the table structure in any way to assist in data mining?

2. What model would be preferrable?

3. If i'm using the model will it extend to the data added to the table automatically or do i have to update it regularily.

I need to submit the project by 20th... and i'm not even started. I tried a lot on my own..... but couldn't get anywhere without definitive assistance from anyone.

Please help

Thanks and Regards,

Sundeep Singh

You should probably use a nested table which contains multiple rows for each of the Minor Attributes associated with a Component - so your mining model might look something like this:

CREATE MINING MODEL CompPredict(
CaseID LONG KEY,
Component TEXT DISCRETE PREDICT,
MajorAttribute TEXT DISCRETE,
MinorAttributes TABLE(
MinorAttribute TEXT KEY
)
)
USING Microsoft_Decision_Trees

If you add data to the source database that you process your model from, you will need to reprocess the model either manually or using scheduled job or Integration Service package.

|||

Hi Raman

Thankx for ur answer..................

I converted the tables as you said

TableComponent(ComponentID, BodyPart,MajorAttribute,ComponentName)

TableMinor(ComponentID,MinorAttribute)

Used TableComponent as Case(ComponentID as Key), and TableMinor as Nested(MinorAttribute as key)

I tried predicting Component Name using BodyPart, MajorAttribute and Minor Attribute as Input.....

I tried it using Microsoft Decision trees and it generated only single node for it.........

What do you think can be the problem....................

Moreover i need to generate the association rules from the data but again it is generating no rules from the given data...........

Thanks

Sundeep Singh

|||How big is your data set? Have you tried tweaking the algorithm parameters (lower COMPLEXITY_PENALTY, MINIMUM_LEAF_CASES, MINIMUM_SUPPORT)?|||

Hi,

I tried to reduce the parameters and got some rules... but there is a problem. the association rules that are being generated use a single minorAtt at a time.

I mean that in the entire collection there is not one that uses two MinorAttributes to Predict the column.

And can you please help me with the query... assuming that i have a major symptom and some minor symptoms and i want to predict what Components they can be for along with the probability of correctness..

Thanks

Sundeep Singh

|||

Hi,

played with the algo params and it worked.. i don't know how as yet.. but i got some rules with more than one minor symptom.

But the problem with the query remains... somebody please help!!!!!!!! only four days left for submission....

Thanks

Sundeep Singh

|||

You can do a SELECT COUNT(*) FROM TableMinor and SELECT COUNT(DISTINCT ComponentID) FROM TableMinor. If these numbers are the same, than you have one minor/component, otherwise you do have multiple minors/component and everything's OK (except maybe your data).

If you don't have multiple minors/component, you don't need a nested table. From the initial description it seemed like there were.

|||

The number of minor are more than one, but the query that i asked for was to predict the component when i enter the minor and major attributes along with the probability of match...

Supposing that i have named my model as Encyclopedia.

Thanks

|||Try this to get the top 3 predictions for component:

SELECT FLATTENED
TopCount(PredictHistogram(Component), $AdjustedProbability, 3)
FROM [ComponentPredictModel]
NATURAL PREDICTION JOIN
(SELECT 'xxx' as MajorAttribute,
(SELECT ( SELECT 'x' AS Minor UNION
SELECT 'y' AS Minor )
AS [MinorAttributes])
) AS t

Thursday, March 8, 2012

Assigning DATEADD results to variable

I am working on a stored procedure that requires saving the results of
several DATEADD calculations to variables. The problem I am having is
illustrated by the following code:
----
declare @.LastOfRecMo datetime
declare @.FirstOfNextMo datetime
declare @.LastOfNextMo datetime
declare @.AprilFirst datetime
select @.AprilFirst = '4/1/2005'
select @.LastOfRecMo = '3/31/2005'
select @.FirstOfNextMo = DATEADD(d, 1, @.LastOfRecMo)
select @.LastOfNextMo = DATEADD(d, -1, DATEADD(m, 1, @.AprilFirst))
select LastOfRecMo = @.LastOfRecMo,
FirstOfNextMo = @.FirstOfNextMo,
FirstOfNextMo2 = DATEADD(d, 1, @.LastOfRecMo),
LastOfNextMo = @.LastOfNextMo
----
I would expect the @.FirstOfNextMo variable to be set to '4/1/2005', but as
you can see by running the above code, the result, returned as FirstOfNextMo
,
is always NULL, although FirstOfNextMo2 returns the correct date using the
identical calculation. Interestingly, @.LastOfNextMo works correctly with a
much more complex DATEADD calculation.
What's going on here?Sheldon Penner wrote:
> I am working on a stored procedure that requires saving the results of
> several DATEADD calculations to variables. The problem I am having is
> illustrated by the following code:
> ----
> declare @.LastOfRecMo datetime
> declare @.FirstOfNextMo datetime
> declare @.LastOfNextMo datetime
> declare @.AprilFirst datetime
> select @.AprilFirst = '4/1/2005'
> select @.LastOfRecMo = '3/31/2005'
> select @.FirstOfNextMo = DATEADD(d, 1, @.LastOfRecMo)
> select @.LastOfNextMo = DATEADD(d, -1, DATEADD(m, 1, @.AprilFirst))
> select LastOfRecMo = @.LastOfRecMo,
> FirstOfNextMo = @.FirstOfNextMo,
> FirstOfNextMo2 = DATEADD(d, 1, @.LastOfRecMo),
> LastOfNextMo = @.LastOfNextMo
> ----
> I would expect the @.FirstOfNextMo variable to be set to '4/1/2005',
> but as you can see by running the above code, the result, returned as
> FirstOfNextMo, is always NULL, although FirstOfNextMo2 returns the
> correct date using the identical calculation. Interestingly,
> @.LastOfNextMo works correctly with a much more complex DATEADD
> calculation.
> What's going on here?
First thing is the date format you are using is not portable. The only
portable formats are:
yyyymmdd
yyyy-mm-ddThh:mm:ss.mmm(no spaces)
But your code works fine for me with the bad date format. Try changing
the date format and see what you get.
David Gugick
Imceda Software
www.imceda.com|||I'm baffled! I worked on this problem for hours yesterday, restarted the SQ
L
server, rebooted the computer, and could not get the code to run properly.
This morning, after reading your post, I tried again and the problem has gon
e
away. Thank you for your response.
"David Gugick" wrote:

> Sheldon Penner wrote:
> First thing is the date format you are using is not portable. The only
> portable formats are:
> yyyymmdd
> yyyy-mm-ddThh:mm:ss.mmm(no spaces)
> But your code works fine for me with the bad date format. Try changing
> the date format and see what you get.
>
> --
> David Gugick
> Imceda Software
> www.imceda.com
>|||I thought the problem was solved, but it's not. DATEADD behaves erratically
when given a variable as an argument. The code I sent you yesterday
inexplicably runs correctly now, but the following example, closer to what I
am actually using, does not:
----
declare @.ReceiverDate datetime
declare @.FirstOfRecMo datetime
declare @.LastOfRecMo datetime
declare @.FirstOfNextMo datetime
declare @.LastOfNextMo datetime
-- Receiver Date is 3/27/2005
set @.ReceiverDate = '20050327'
-- First Day of Receiver Month
select @.FirstOfRecMo = cast(cast(Year(@.ReceiverDate) as char(4)) + right('0'
+ cast(Month(@.ReceiverDate) as varchar(2)), 2) + '01' as datetime)
select myVar = @.FirstOfRecMo,
calc = cast(cast(Year(@.ReceiverDate) as char(4)) + right('0' +
cast(Month(@.ReceiverDate) as varchar(2)), 2) + '01' as datetime)
-- First Day of Month after Receiver Month
select @.FirstOfNextMo = DATEADD(m, 1, @.FirstOfRecMo)
select myVar = @.FirstOfNextMo,
calc = DATEADD(m, 1, @.FirstOfRecMo)
-- First Day of Receiver Month
select @.LastOfRecMo = DATEADD(d, -1, @.FirstOfNextMo)
select myVar = @.LastOfRecMo,
calc = DATEADD(d, -1, @.FirstOfNextMo)
-- Last Day of Month after Receiver Month
select @.LastOfNextMo = DATEADD(d, -1, DATEADD(m, 1, @.FirstOfNextMo))
select myVar = @.LastOfNextMo,
calc = DATEADD(d, -1, DATEADD(m, 1, @.FirstOfNextMo))
----
I entered the Receiver Date in the format you recommended, although in the
actual application, the date is pulled from the database.
The first calculation correctly returns '3/1/2005' in both the myVar and
calc columns, indicating that '3/1/2005' was successfully assigned to the
variable @.FirstOfRecMo.
The second calculation, however, fails to assign the results of the DATEADD
function to @.FirstOfNextMo. It returns NULL in the myVar column, although
the calc column correctly returns '4/1/2005'.
The other two calculations return NULL in both columns.
Is the problem with my SQL Server, perhaps? Does the code run correctly on
your computer?
Your help is very much appreciated.
"David Gugick" wrote:

> Sheldon Penner wrote:
> First thing is the date format you are using is not portable. The only
> portable formats are:
> yyyymmdd
> yyyy-mm-ddThh:mm:ss.mmm(no spaces)
> But your code works fine for me with the bad date format. Try changing
> the date format and see what you get.
>
> --
> David Gugick
> Imceda Software
> www.imceda.com
>|||This is very creepy!
When I type the code into Query Analyzer, it produces the errors I described
in my previous messages. However, if I copy the code to this newsgroup, pos
t
it, then copy the code from the post and paste it into Query Analyzer, it
runs properly.
Evidently, SQL Server is messing with my code. Any thoughts?
"Sheldon Penner" wrote:
> I thought the problem was solved, but it's not. DATEADD behaves erratical
ly
> when given a variable as an argument. The code I sent you yesterday
> inexplicably runs correctly now, but the following example, closer to what
I
> am actually using, does not:
> ----
> declare @.ReceiverDate datetime
> declare @.FirstOfRecMo datetime
> declare @.LastOfRecMo datetime
> declare @.FirstOfNextMo datetime
> declare @.LastOfNextMo datetime
> -- Receiver Date is 3/27/2005
> set @.ReceiverDate = '20050327'
> -- First Day of Receiver Month
> select @.FirstOfRecMo = cast(cast(Year(@.ReceiverDate) as char(4)) + right('
0'
> + cast(Month(@.ReceiverDate) as varchar(2)), 2) + '01' as datetime)
> select myVar = @.FirstOfRecMo,
> calc = cast(cast(Year(@.ReceiverDate) as char(4)) + right('0' +
> cast(Month(@.ReceiverDate) as varchar(2)), 2) + '01' as datetime)
> -- First Day of Month after Receiver Month
> select @.FirstOfNextMo = DATEADD(m, 1, @.FirstOfRecMo)
> select myVar = @.FirstOfNextMo,
> calc = DATEADD(m, 1, @.FirstOfRecMo)
> -- First Day of Receiver Month
> select @.LastOfRecMo = DATEADD(d, -1, @.FirstOfNextMo)
> select myVar = @.LastOfRecMo,
> calc = DATEADD(d, -1, @.FirstOfNextMo)
> -- Last Day of Month after Receiver Month
> select @.LastOfNextMo = DATEADD(d, -1, DATEADD(m, 1, @.FirstOfNextMo))
> select myVar = @.LastOfNextMo,
> calc = DATEADD(d, -1, DATEADD(m, 1, @.FirstOfNextMo))
> ----
> I entered the Receiver Date in the format you recommended, although in the
> actual application, the date is pulled from the database.
> The first calculation correctly returns '3/1/2005' in both the myVar and
> calc columns, indicating that '3/1/2005' was successfully assigned to the
> variable @.FirstOfRecMo.
> The second calculation, however, fails to assign the results of the DATEAD
D
> function to @.FirstOfNextMo. It returns NULL in the myVar column, although
> the calc column correctly returns '4/1/2005'.
> The other two calculations return NULL in both columns.
> Is the problem with my SQL Server, perhaps? Does the code run correctly o
n
> your computer?
> Your help is very much appreciated.
> "David Gugick" wrote:
>