Re: Interpreting csscan results

  • From: "Don Seiler" <don@xxxxxxxxx>
  • To: oracle_l <oracle-l@xxxxxxxxxxxxx>
  • Date: Wed, 9 May 2007 11:12:25 -0500

I looked deeper into the specific exceptions.  All of the data
dictionary exceptions are just histograms on the application data
fields with the WestEuro characters, except for this oddball:

User  : SYS
Table : SOURCE$
Column: SOURCE
Type  : VARCHAR2(4000)
Number of Exceptions         : 1
Max Post Conversion Data Size: 4000

ROWID              Exception Type      Size Cell Data(first 30 bytes)
------------------ ------------------ ----- ------------------------------
AAAABHAABAAAHqVABL lossy conversion         end;
------------------ ------------------ ----- ------------------------------

Not sure why "end;" is a problem.

Don.

On 5/9/07, Don Seiler <don@xxxxxxxxx> wrote:
Morning all.  Our production database (10.2.0.2 on RHEL3) is USASCII7,
and we've recently had issues with western euro characters being used.
 I'm researching the idea of migrating to WE8ISO8859P1.

So last night I ran csscan with those from/to parameters, and I'm more
than a little confused.  My reading led me to believe that since
WE8ISO8859P1 is a complete superset of USASCII7, there should be no
issues.  Perhaps I'm just interpreting the results wrong.

The scan summary says:
Some character type data in the data dictionary are not convertible to
the new character set
Some character type application data are not convertible to the new
character set

Here is the summary for each data dictionary and app data:

[Data Dictionary Conversion Summary]

Datatype                    Changeless      Convertible
Truncation            Lossy
--------------------- ---------------- ----------------
---------------- ----------------
VARCHAR2                    17,344,573                0
0               30
CHAR                             1,216                0
0                0
LONG                           916,150                0
0                0
CLOB                         1,505,729                0
0                0
VARRAY                          17,408                0
0                0
--------------------- ---------------- ----------------
---------------- ----------------
Total                       19,785,076                0
0               30
Total in percentage            100.000%           0.000%
0.000%           0.000%

The data dictionary can not be safely migrated using the CSALTER script

[Application Data Conversion Summary]

Datatype                    Changeless      Convertible
Truncation            Lossy
--------------------- ---------------- ----------------
---------------- ----------------
VARCHAR2                49,180,912,298                0
0           11,819
CHAR                    10,295,777,604                0
0              736
LONG                             4,957                0
0                0
CLOB                                 1                0
0                0
VARRAY                               0                0
0                0
--------------------- ---------------- ----------------
---------------- ----------------
Total                   59,476,694,860                0
0           12,555
Total in percentage            100.000%           0.000%
0.000%           0.000%

The scan.err file lists those names with mangled WE characters with a
"lossy conversion" exception.  I'm wondering if this (in particular
the data dictionary warning) means my database will be horribly
corrupted if I try CSALTER, or if it just means I will have to correct
those particular fields afterward.

--
Don Seiler
oracle blog: http://ora.seiler.us
ultimate: http://www.mufc.us



--
Don Seiler
oracle blog: http://ora.seiler.us
ultimate: http://www.mufc.us
--
//www.freelists.org/webpage/oracle-l


Other related posts: