LOAD真的是这个大脑受损吗? [英] Is LOAD really this brain damaged?

查看:128
本文介绍了LOAD真的是这个大脑受损吗?的处理方法,对大家解决问题具有一定的参考价值,需要的朋友们下面随着小编来一起学习吧!

问题描述

DB2 WSUE LUW v8.2 FP4(又名v8.1 FP11)

RHEL AS 4

我从一个模式以IXF格式导出然后在另一台服务器上加载另一个模式



我正在使用的数据库建模工具喜欢将PK列放在

非常接近CREATE TABLE ddl,因此FROM模式中列的排序有时与TO模式中的排序不同。


我发现:

1.如果列的顺序不同,我*必须*使用

METHOD N(colx,coly,colz)子句负荷。所有列名

都是一样的 - 为什么不能把它弄清楚?


2.如果列的顺序是相同的我*不得*在LOAD上使用

METHOD N(colx,coly,colz)子句。如果我这样做,我会得到:

SQL3086N没有指定要加载到

数据库列中的源列0或者指定的源列不存在,

但是数据库列不可为空。


所以,如果我具体关于列的排序,那么我真的不需要

,LOAD不行吗?我不能太具体了吗?


我错过了什么吗? <抓头>


aj


PS - 在你问之前,我使用的是LOAD而不是IMPORT,因为我需要

IDENTITYOVERRIDE。

DB2 WSUE LUW v8.2 FP4 (aka v8.1 FP11)
RHEL AS 4

I am EXPORTing in IXF format from one schema and then LOADing into
another schema on another server.

The DB modeling tool I am using likes to put the PK columns at the
very top of the CREATE TABLE ddl, so the ordering of the columns in
the FROM schema is sometimes different from the ordering in the TO schema.

I have discovered that:
1. If the ordering of the columns is different, I *must* use the
METHOD N (colx,coly,colz) clause on the LOAD. All the column names
are the same - why can''t LOAD figure it out?

2. If the ordering of the columns is the same I *must not* use the
METHOD N (colx,coly,colz) clause on the LOAD. If I do, I''ll get:
SQL3086N There was no source column specified to be loaded into
database column "0" or the specified source column does not exist,
but the database column is not nullable.

So if I''m specific about the column ordering when I really don''t need
to be, the LOAD won''t work? I can''t be TOO specific??

Am I missing something? <scratching head>

aj

PS - Before you ask, I am using LOAD rather than IMPORT because I need
IDENTITYOVERRIDE.

推荐答案

aj写道:
DB2 WSUE LUW v8.2 FP4(又名v8.1 FP11)
RHEL AS 4

我从一个模式导出IXF格式,然后加载到另一个服务器上的另一个模式。

数据库我正在使用的建模工具喜欢将PK列放在CREATE TABLE ddl的最顶端,因此FROM模式中列的排序有时与TO模式中的排序不同。

我发现:
1.如果列的顺序不同,我*必须*使用
METHOD N(colx,coly,colz)子句加载。所有的列名都是一样的 - 为什么不能把它搞清楚?
我想他们想要保存而不是抱歉...

2.如果列的顺序相同,我*不得*使用
METHOD N(关于LOAD的colx,coly,colz)条款。如果我这样做,我会得到:
SQL3086N没有指定要加载到
数据库列0的源列。或者指定的源列不存在,
但数据库列不可为空。
DB2 WSUE LUW v8.2 FP4 (aka v8.1 FP11)
RHEL AS 4

I am EXPORTing in IXF format from one schema and then LOADing into
another schema on another server.

The DB modeling tool I am using likes to put the PK columns at the
very top of the CREATE TABLE ddl, so the ordering of the columns in
the FROM schema is sometimes different from the ordering in the TO schema.

I have discovered that:
1. If the ordering of the columns is different, I *must* use the
METHOD N (colx,coly,colz) clause on the LOAD. All the column names
are the same - why can''t LOAD figure it out? I guess they want to be rather save than sorry...

2. If the ordering of the columns is the same I *must not* use the
METHOD N (colx,coly,colz) clause on the LOAD. If I do, I''ll get:
SQL3086N There was no source column specified to be loaded into
database column "0" or the specified source column does not exist,
but the database column is not nullable.



看起来更像是缺少规范中的列。 />
目标列和来源的签名是什么?


我认为完整的复制方案可能在这里有序。

干杯

Serge


-

Serge Rielau

DB2解决方案开发

IBM多伦多实验室


It seems more like you''re missing a column in your specification.
What is the signature of the target column and the source?

I think a complete repro scenario might be in order here.
Cheers
Serge

--
Serge Rielau
DB2 Solutions Development
IBM Toronto Lab


感谢您的回复,Serge。


.... snip ..


Serge Rielau写道:
Thanks for the reply, Serge.

....snip..

Serge Rielau wrote:
2.如果列的顺序相同,我*不得*使用
LOAD上的方法N(colx,coly,colz)子句。如果我这样做,我会得到:
SQL3086N没有指定要加载到
数据库列0的源列。或者指定的源列不存在,
但数据库列不可为空。
2. If the ordering of the columns is the same I *must not* use the
METHOD N (colx,coly,colz) clause on the LOAD. If I do, I''ll get:
SQL3086N There was no source column specified to be loaded into
database column "0" or the specified source column does not exist,
but the database column is not nullable.


看起来更像是缺少规范中的列。
什么是目标栏和来源的签名?


It seems more like you''re missing a column in your specification.
What is the signature of the target column and the source?




我不这么认为。如果我只是在列排序匹配的LOAD上删除METHOD N(colx,coly,colz)

子句,它就可以完美地工作

** **。 />

aj



I don''t think so. If I simply remove the METHOD N (colx,coly,colz)
clause on the LOAD where the column ordering matches, it works
**perfectly**.

aj


aj写道:
感谢回复,Serge。

...剪辑..

Serge Rielau写道:
Thanks for the reply, Serge.

...snip..

Serge Rielau wrote:
2.如果列的顺序是相同的我*必须不要在LOAD上使用
METHOD N(colx,coly,colz)子句。如果我这样做,我会得到:
SQL3086N没有指定要加载到
数据库列0的源列。或者指定的源列不存在,
但数据库列不可为空。
2. If the ordering of the columns is the same I *must not* use the
METHOD N (colx,coly,colz) clause on the LOAD. If I do, I''ll get:
SQL3086N There was no source column specified to be loaded into
database column "0" or the specified source column does not exist,
but the database column is not nullable.


看起来更像是缺少规范中的列。
什么是目标栏和来源的签名?


It seems more like you''re missing a column in your specification.
What is the signature of the target column and the source?



我不这么认为。如果我只是在列顺序匹配的LOAD上删除METHOD N(colx,coly,colz)
子句,它就可以完美地工作**。

aj



I don''t think so. If I simply remove the METHOD N (colx,coly,colz)
clause on the LOAD where the column ordering matches, it works
**perfectly**.

aj



所以,难道你有4列而不是3列吗?


干杯

Serge

-

Serge Rielau

DB2解决方案开发

IBM多伦多实验室


So, could it be you have 4 columns instead of 3?

Cheers
Serge
--
Serge Rielau
DB2 Solutions Development
IBM Toronto Lab


这篇关于LOAD真的是这个大脑受损吗?的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!

查看全文
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆