Metalink - alternate HTML site

From: <Philip>
Date: Wed, 17 Feb 2010 20:36:36 -0600
Message-ID: <2010217213628phdye_at_acm.org>



Try:

  https://supporthtml.oracle.com/

I've never even gotten the new site one to get past login.

Enjoy,

Philip

Mladen Gogala wrote:

Metalink disaster
14-Nov-09

This thing with the Metalink is turning into an unmitigated disaster of biblical proportions. I tried to log into Metalink because I want to open a TAR, but I cannot get in. Here is my problem:

SQL> create cluster print_clu(active char(1)) 2 tablespace imi
3 single table hashkeys 3;

Cluster created.

Elapsed: 00:00:00.12
SQL> create table print_tmp
2 cluster print_clu(active)
3 as select * from print where rownum<0; create table print_tmp
*
ERROR at line 1:
ORA-03001: unimplemented feature

The version is 10.2.0.4 on Linux, 64 bit. I want to know what exactly is the "unimplemented feature" and why is it not possible to populate the cluster? Of course, ML being out, I cannot really ask anybody. Are we going to get some discounts for this support disaster? Who is the idiot who has insisted on replacing the old Metalink and why exactly was it done? Will the guilty be fired? This shows a serious lack of judgment and arrogance on the part of the Oracle Corp. This is not just a joke, we are paying significant amounts of money for support!

--

http://mgogala.freehostia.com

Previous Posts In This Thread:

On Saturday, November 14, 2009 12:57 AM
Mladen Gogala wrote:

Metalink disaster
This thing with the Metalink is turning into an unmitigated disaster of biblical proportions. I tried to log into Metalink because I want to open a TAR, but I cannot get in. Here is my problem:

SQL> create cluster print_clu(active char(1)) 2 tablespace imi
3 single table hashkeys 3;

Cluster created.

Elapsed: 00:00:00.12
SQL> create table print_tmp
2 cluster print_clu(active)
3 as select * from print where rownum<0; create table print_tmp
*
ERROR at line 1:
ORA-03001: unimplemented feature

The version is 10.2.0.4 on Linux, 64 bit. I want to know what exactly is the "unimplemented feature" and why is it not possible to populate the cluster? Of course, ML being out, I cannot really ask anybody. Are we going to get some discounts for this support disaster? Who is the idiot who has insisted on replacing the old Metalink and why exactly was it done? Will the guilty be fired? This shows a serious lack of judgment and arrogance on the part of the Oracle Corp. This is not just a joke, we are paying significant amounts of money for support!

--

http://mgogala.freehostia.com

On Saturday, November 14, 2009 5:09 AM
Matthias Hoys wrote:

I wonder how they tested the new system - not? I wonder how they tested the new system - not? Based on the old Metalink visitor statistics, they should have a good guess at the number of logins and concurrent transactions to expect, no? After all, there are currently tons of tools on the market to stress and load-test web applications before they got released into production, no? Or maybe the database backend of MOS is not Oracle anymore? Maybe it is MySQL??

Matthias

On Saturday, November 14, 2009 6:31 AM
Noons wrote:

Mladen Gogala wrote,on my timestamp of 14/11/2009 4:57 PM:The problem is you Mladen Gogala wrote,on my timestamp of 14/11/2009 4:57 PM:

The problem is you are trying to create a TAR. Apparently, to "get out in front" they now create only a SR...

(1K apologies: could not resist)

Does it work without CTAS?
As in: create table followed by insert ... select...?

On Saturday, November 14, 2009 6:48 AM
MBPP wrote:

enJust a guess now, can it be because of some reserved word from thetable en

Just a guess now, can it be because of some reserved word from the table "print"?

On Saturday, November 14, 2009 7:01 AM
MBPP wrote:

openAnother idea, does the table "print" has any BLOB or CLOB column? open

Another idea, does the table "print" has any BLOB or CLOB column?

On Saturday, November 14, 2009 10:33 AM
Charles Hooper wrote:

Mladen,You might want to check the column definitions for the table PRINT. Mladen,

You might want to check the column definitions for the table PRINT. While I have little experience working with clusters, I think that I was able to reproduce the problem that you are experiencing.

Here is a test case I created:
CREATE CLUSTER PRINT_CLU(ACTIVE CHAR(1)) TABLESPACE USER_DATA
SINGLE TABLE HASHKEYS 3; CREATE TABLE PRINT AS
SELECT
*
FROM
DBA_OBJECTS; ALTER TABLE PRINT RENAME COLUMN OBJECT_TYPE TO ACTIVE; CREATE TABLE PRINT_TMP
CLUSTER PRINT_CLU(ACTIVE)
AS SELECT * FROM PRINT WHERE ROWNUM<0;

ERROR at line 2:
ORA-01753: column definition incompatible with clustered column definition

/* OK, not the error, so eliminates that as a possibility */

DROP TABLE PRINT; CREATE TABLE PRINT AS
SELECT
*
FROM
DBA_OBJECTS; ALTER TABLE PRINT ADD (ACTIVE CHAR(1)); CREATE TABLE PRINT_TMP
CLUSTER PRINT_CLU(ACTIVE)
AS SELECT * FROM PRINT WHERE ROWNUM<0;

Table created.

/* OK, that worked */

DROP TABLE PRINT_TMP; ALTER TABLE PRINT MODIFY ACTIVE VARCHAR2(1); CREATE TABLE PRINT_TMP
CLUSTER PRINT_CLU(ACTIVE)
AS SELECT * FROM PRINT WHERE ROWNUM<0;

ERROR at line 2:
ORA-01753: column definition incompatible with clustered column definition

/* OK, column definition must be the same, different error, so eliminates that as a possibility */

ALTER TABLE PRINT MODIFY ACTIVE CHAR(1); ALTER TABLE PRINT ADD LONG_RAW LONG RAW; CREATE TABLE PRINT_TMP
CLUSTER PRINT_CLU(ACTIVE)
AS SELECT * FROM PRINT WHERE ROWNUM<0;

ERROR at line 3:

On Saturday, November 14, 2009 11:05 AM
Charles Hooper wrote:

(snip)(snip)Just a follow up.  Oracle 11.1.0.
(snip)
(snip)

Just a follow up. Oracle 11.1.0.7 also produces the same error message when the PRINT table contained a BLOB column. A search of the documentation found this:
http://download.oracle.com/docs/cd/B28359_01/server.111/b28286/statements_7= 002.htm
"Restrictions on Cluster Tables
Cluster tables are subject to the following restrictions: =95Object tables and tables containing LOB columns or columns of the Any* Oracle-supplied types cannot be part of a cluster. =95You cannot specify the parallel_clause or CACHE or NOCACHE for a table that is part of a cluster.
=95You cannot specify CLUSTER with either ROWDEPENDENCIES or NOROWDEPENDENCIES unless the cluster has been created with the same ROWDEPENDENCIES or NOROWDEPENDENCIES setting."

I guess that is something else that may be checked, but I am not sure why a more appropriate error message is not returned as was the case when a LONG RAW column was included in the PRINT table.

Charles Hooper
IT Manager/Oracle DBA
K&M Machine-Fabricating, Inc.

On Saturday, November 14, 2009 11:21 AM
hpuxrac wrote:

snipIt started working for me yesterday consistently. snip

It started working for me yesterday consistently. Was able to login a bunch of times yesterday starting late morning. No failures. I was able to login last night ( checking access from home ) and again this morning I am able to get in.

To me anyhow it appears as if MOS may be "somewhat fixed". What problems are you having when you try to log in?

On Sunday, November 15, 2009 10:49 AM
Mladen Gogala wrote:

Nope, I tried that, too.--http://mgogala.freehostia.com Nope, I tried that, too.

--

http://mgogala.freehostia.com

On Sunday, November 15, 2009 10:50 AM
Mladen Gogala wrote:

Yes, that is the problem.--http://mgogala.freehostia.com Yes, that is the problem.

--

http://mgogala.freehostia.com

On Sunday, November 15, 2009 12:50 PM
Mladen Gogala wrote:

The problem is that tables containing CLOB columns cannot be clustered. The problem is that tables containing CLOB columns cannot be clustered. That limitation is present even in the release 11.2. BTW, the new and shiny "securefile storage" is not as efficient as one would hope. I have an expdp file 28GB in size (actually, 7 4GB slices) and I dedicated 45GB for the CLOB column. De-duplication was turned on and compression set to "high". It failed with the error "cannot extend tablespace". Bummer. It seems that the new feature is not worth the money Oracle is asking for it.

--

http://mgogala.freehostia.com

On Sunday, November 15, 2009 12:56 PM
Mladen Gogala wrote:

Me too. Bookmarks are back but my old patch searches are gone. Me too. Bookmarks are back but my old patch searches are gone. This was not what I would call a seamless and smooth upgrade. If one of my upgrades was like this one, there would be a heck to pay, pay cuts and pink slips, too.

--

http://mgogala.freehostia.com

Submitted via EggHeadCafe - Software Developer Portal of Choice Sending SMTP email from within BizTalk Orchestration http://www.eggheadcafe.com/tutorials/aspnet/9dd0f346-baf9-4674-a50f-1716445b26bc/sending-smtp-email-from-w.aspx Received on Wed Feb 17 2010 - 20:36:36 CST

Original text of this message