Adding Nodes

[latifa]

Hi Erman,

When adding nodes to the master one (with : perl adcfgclone.pl appsTier dualfs), I have the following WARNING message and always the same (in the second part of cloning I mean while cloning the patch fs):

Running:
perl adaddnode.pl -appsuser=APPS -promptmsg=hide

WARNING: Whole add node operation is completed, only setting of APPL_TOP_ID has failed.
Please check the error log in adaddnode_<timeStamp> and try to run it manually if required.

I check that log file, it says:
ERROR at line 1:
ORA-00001: unique constraint (APPLSYS.AD_APPL_TOPS_U2) violated
ORA-06512: at line 16
Error configuring snapshot information for the source node...cannot proceed with setting the appl top name for the node.

I did something on some servers: I deleted the node in AD_APPL_TOPS and rerun perl adaddnode.pl -appsuser=APPS. The error message changed as following:

ERROR at line 1:
ORA-01403: no data found
ORA-06512: at line 159
Error configuring snapshot information for the targer node...cannot proceed with setting the appl top name for the node.

What do you think about that error ? how can I avoid it ?

Thank you in adavance for your reply.

Regards,
Latifa

[ErmanArslansOracleBlog]

Hi Latifa

check this note -> Adaddnode.Pl Fails With Ora-00001: Unique Constraint (Applsys.Ad_appl_tops_u2) Violated (Doc ID 1435834.1)

This is for EBS 11i, but I think the logic of the error is the same.
Do the checks documented in document above.

Consider the solution documented there.

Ping me if you have any questions.

[latifa]

Hi Erman,

Thank you for your reply.

I have two questions:

1. What do they mean by :

%%START_OF_TOKEN%%
APPL_TOP_NAME
%%END_OF_TOKEN%%

%%START_OF_VALUE%%
<old_node>
%%END_OF_VALUE%%

in my file I have for example :

## Start of Configuration Record
  %%START_OF_TOKEN%%
        APPL_TOP_NAME
  %%END_OF_TOKEN%%

  %%START_OF_VALUE%%
        udkfinp07
  %%END_OF_VALUE%%
## End of Configuration Record

---> udkfinp07 is the node I want to add and it was previously there.
--> question:  Shull I put the master server name there ? or leave it like that ?


2. As deleted all nodes from AD_APPL_TOPS table, I don't have the old APPL_TOP_IDs. Dates in that snapshots tables are old:

appl_top_ids.txt


I think I have to reapply the note id you sent me on all nodes even the master ? Before, I have done following:
- delete node from ad_appl_tops.
- rerun addnode
- run autconf

Regards,
Latifa

[ErmanArslansOracleBlog]

I think your adconfig.txt file is okay.

Please do the following;

1. Backup ad_appl_tops table

2. Remove the incorrectly inserted target atName from AD_APPL_TOPS table:
SQL> delete from AD_APPL_TOPS where NAME='<new_node>';    (I think this is udkfinp07 for your case)
SQL> commit;

3. Run adaddnode.pl

What is the result?

[latifa]

Hi Erman,

as result (as before):

declare
*
ERROR at line 1:
ORA-01403: no data found
ORA-06512: at line 159


Error configuring snapshot information for the targer node...cannot proceed with setting the appl top name for the node.

Regards
latifa

[latifa]

however I run autoconf on both fs and it's OK. The node is registred in fnd_nodes, ad_appl_tops.

[ErmanArslansOracleBlog]

cool..
So the first action plan actually does the job , right?
I mean the one documented in Oracle Support note 1435834.1.. (it includes autoconfig)

[latifa]

I guess yes. Sorry I didn't check fnd_nodes and ad_appl_tops before running autoconfig.

I saw only the error message after running adaddnode.pl and than run autoconf :(

So than as the node is registred in fnd_nodes and ad_appl_tops, can we consider that the error message has no meaning while adding the node with adaddnode.pl ?

[ErmanArslansOracleBlog]

Hi Latifa,


If you can do the following verification and if it is ok, then you are ok..

--Verify that new_node was added as a new node and has an associated snapshot. The following queries should return:


SQL> select APPL_TOP_ID, NAME from ad_appl_tops;
APPL_TOP_ID NAME
----------- --------------------------------------------------
...
<Newid> <new_node>

SQL> select APPL_TOP_ID, SNAPSHOT_ID, SNAPSHOT_NAME from ad_snapshots;
APPL_TOP_ID SNAPSHOT_ID SNAPSHOT_NAME
----------- ----------- ----------------------
...
<Newid> <NewSnap_id> CURRENT_VIEW

[latifa]

Hi Erman,

These are the results of the sql queries:

SQL> select APPL_TOP_ID, SNAPSHOT_ID, SNAPSHOT_NAME, CREATION_DATE from ad_snapshots ;

       2233        7175 CURRENT_VIEW         26-JAN-17
       1233        6175 CURRENT_VIEW         16-DEC-16
        209        6176 GLOBAL_VIEW          16-DEC-16
       1233        5173 CURRENT_VIEW         30-NOV-12
        209        5174 GLOBAL_VIEW          30-NOV-12
        231        5175 *PRESEEDED*12.2.0.0  08-DEC-16

6 rows selected.

SQL> select APPL_TOP_ID, NAME, CREATION_DATE from ad_appl_tops ;
        210 appsun01                                           02-JAN-07
        230 appsun13                                           02-JAN-07
        231 *PRESEEDED*                                        02-JAN-07
        232 appsun02                                           23-JAN-08
       2242 udkfinp04                                          10-APR-17
       3246 udkfinp08                                          27-APR-17
       3243 udkfinp06                                          25-APR-17
       1232 appsun03                                           02-APR-09
        209 GLOBAL                                             26-JUL-06
        169 *PRESEEDED*                                        29-OCT-04
        189 ap981sun                                           15-JUL-06
       2244 udkfinp05                                          11-APR-17
       3245 udkfinp07                                          27-APR-17
       2241 udkfinp03                                          04-APR-17

14 rows selected.

It seems it's not OK :(

What to do now ?

Regards,
Latifa

[ErmanArslansOracleBlog]

You have the new node added .. It can be seen in ad_appl_tops table.
But, you don't have the row for that node's appl_top_id in your ad_snapshots table.
Your ad_snapshots table is not aligned with your ad_appl_tops table.. The appl_top ids are not there in the ad_snapshots table , in general...

Did you cloned this nodes without any problems?
I m asking this because, before running adaddnode.pl, you should have been renamed new nodes in your ad_appl_tops tables, but their ID should be same as the old_node(your clone source).

Do the action plan documented in "Adaddnode.Pl Fails With Ora-00001: Unique Constraint (Applsys.Ad_appl_tops_u2) Violated (Doc ID 1435834.1)"
Step by step...

Attention -> on step 3 "Rename the APPL_TOP_NAME in $APPL_TOP/admin/adconfig.txt to match the value in the old node system"
So, you will use the old node name here..

[latifa]

Hi Erman,

I applied the Doc. ID but still have problem with adding the node to the ad_snapshots table.

please find the logs. It looks some errors but not clear for me:

adconfig.log
adaddnode_Fri_May_12_10_51_52_2017.adaddnode_Fri_May_12_10_51_52_2017

Thanks in advance for your advice.

Regards,
Latifa

[latifa]

just one detail: I supposed tha my old_node is the primary server: udkfinp03. I put that node_name in the $APPL_TOP/admin/adconfig.txt.

[ErmanArslansOracleBlog]

Check this note -> adaddnode.pl Ends With ORA-01403: No Data Found And ORA-06512: at line 159 (Doc ID 455130.1)

[ErmanArslansOracleBlog]

Also send me the following files for analysis;

/opt/application/V12/fs2/EBSapps/appl/ad/12.0.0/patch/115/sql/adadmrat.sql

/opt/application/V12/fs2/inst/apps/DBOFINP1_udkfinp08/temp/adadmat_sql.spool

[latifa]

Here you are:

adadmrat.zip

[latifa]

I ran adadmin on the 08 node, result: failed with following error:

Error in addgbc():
Multiple Sets of Books is not compatible with MultiOrg


AD Administration error:
Error getting configuration information in adadmin_main()


I have done adadmin on the master server; same error.

Regards,
Latifa

[ErmanArslansOracleBlog]

The root cause should be the following sql..
But update should not cause any ORA-01403 errors.
I suspect there may be another thing related with the ad_appl_tops table.. A trigger maybe.

    update ad_appl_tops
    set name = l_new_appltop_name
    where applications_system_name = l_new_apps_system_name
    and name = l_old_appltop_name;

Again;

Before running adaddnode.pl the AD_APPL_TOPS table should look like you cloned <old_node> to <new_node>, i.e. <old_node> should have been renamed to <new_node>, keeping the same ID as <old_node>.

INFO:

/opt/application/V12/fs2/EBSapps/10.1.2/bin/sqlplus -s /nolog @/opt/application/V12/fs2/EBSapps/appl/ad/12.0.0/patch/115/sql/adadmrat.sql APPS ******* DBOFINP1 DBOFINP1 udkfinp08 udkfinp03 /opt/application/V12/fs2/inst/apps/DBOFINP1_udkfinp08/temp/adadmat_sql.spool

ARGUMENT NUMBERS:

  l_new_apps_system_name  := '&4';
  l_old_appltop_name  := '&5';

According to the arguments ->

your l_old_appltop_name= udkfinp08
your l_new_apps_system_name= DBOFINP1


Send me the following sql's output;

1)select * from ad_appl_tops; (in a good formatted way)
2)check the triggers on ad_appl_tops and ad_timestamps tables (not editioning views). Send me their scripts, (If there are any)

[ErmanArslansOracleBlog]

Bytheway, there are no errors  in the "/opt/application/V12/fs2/inst/apps/DBOFINP1_udkfinp08/temp/adadmat_sql.spool" file that you sent to me?

??Why is there such a difference? the outputs in the spool file and the output in the log are also different. did you send me the correct spool file? (spool file produced when you run adaddnode.pl, which encountered ORA-1403 error)

[latifa]

You are right. The one containing the error message is the follwoing:

adadmdat_sql.spool

there is a difference in the name

et this one is the spool :

adaddnode_Fri_May_12_10_51_52_2017.adaddnode_Fri_May_12_10_51_52_2017