Andy,
Thanks for your response, and yes, this collocgroup is the first one
displayed while running your query.
Thought it was a bug too, but wanted to make it sure first !
Will you relay this to development team, or shall I open a PMR ?
Cheers.
Arnaud
************************************************************************
******
Panalpina Management Ltd., Basle, Switzerland, CIT Department
Viadukstrasse 42, P.O. Box 4002 Basel/CH
Phone: +41 (61) 226 11 11, FAX: +41 (61) 226 17 01
Direct: +41 (61) 226 19 78
e-mail: arnaud.brion AT panalpina DOT com
************************************************************************
******
-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
Andrew Raibeck
Sent: Friday, 26 August, 2005 13:08
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: SQL query problem on collocgroup table
Well, well, well, this looks like a bona fide bug to me. Is AIX_0_6 the
first collocation group name in the output from:
select collocgroup_name from collocgroup
If so, this would be consistent with my findings (though I cannot say I
have looked at this exhaustively): your script works only when the
collocation group name is the first one in the list.
As a test, try running this:
select node_name, collocgroup_name from collocgroup where
collocgroup_name in ('AIX_INACTIVE')
that should work, though I realize it isn't what you really want.
Andy Raibeck
IBM Software Group
Tivoli Storage Manager Client Development Internal Notes e-mail: Andrew
Raibeck/Tucson/IBM@IBMUS Internet e-mail: storman AT us.ibm DOT com
The only dumb question is the one that goes unasked.
The command line is your friend.
"Good enough" is the enemy of excellence.
"ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> wrote on 2005-08-26
02:12:31:
> Hi all,
>
> I'm facing some strange SQL problem while querying COLLOCGROUP table,
> and can't find out what is happening here ...
> I've built a script (q_collocgroup) which looks like :
>
> select collocgroup_name from collocgroup where node_name=upper('$1')
> select node_name, collocgroup_name from collocgroup where
> collocgroup_name in (select collocgroup_name from collocgroup where
> node_name=upper('$1'))
>
> For some nodes, the scripts works perfectly :
>
> run q_collocgroup asbrs000
>
> COLLOCGROUP_NAME
> ------------------
> AIX_0_6
>
> NODE_NAME COLLOCGROUP_NAME
> ------------------ ------------------
> ASBRS000 AIX_0_6
> PACRS070 AIX_0_6
> PACSP030 AIX_0_6
>
>
> For other ones, the second query fails :
>
> run q_collocgroup old_pacrs111
>
> COLLOCGROUP_NAME
> ------------------
> AIX_INACTIVE
> ANR2034E SELECT: No match found using this criteria.
> ANR1462I RUN: Command script Q_COLLOCGROUP completed successfully.
>
> I tried an equivalent query :
> (SELECT COLLOCGROUP.NODE_NAME, COLLOCGROUP.COLLOCGROUP_NAME FROM
> COLLOCGROUP WHERE (((COLLOCGROUP.COLLOCGROUP_NAME)=(SELECT
> COLLOCGROUP.COLLOCGROUP_NAME FROM COLLOCGROUP WHERE
> (((COLLOCGROUP.NODE_NAME)=[enter node_name]))))); using MS Access and
> ODBC driver, with the same problem
>
> Does someone have an idea of what is happening here ?
> Thanks in advance !
> Cheers.
>
> Arnaud
>
> **********************************************************************
> **
> ******
> Panalpina Management Ltd., Basle, Switzerland, CIT Department
> Viadukstrasse 42, P.O. Box 4002 Basel/CH
> Phone: +41 (61) 226 11 11, FAX: +41 (61) 226 17 01
> Direct: +41 (61) 226 19 78
> e-mail: arnaud.brion AT panalpina DOT com
> **********************************************************************
> **
> ******
|