FNDSM_Queue

[syedahmed]

hi Emran,

need your kind help in this after i clone 12.2.7 facing issue with concurrent managers not getting start
open SR also not getting proper response longtime

oracle provide the following step to excute to regist FNSDSM but this also getting failed can you please advice on this any alternate way to achieve this

cd $FND_TOP/patch/115/sql

SQL> @afdcm037.sql;
declare
*
ERROR at line 1:
ORA-01400: cannot insert NULL into
("APPLSYS"."FND_CONCURRENT_QUEUES"."CONCURRENT_QUEUE_NAME")
ORA-06512: at "APPS.FND_MANAGER", line 541
ORA-06512: at "APPS.FND_CP_FNDSM", line 228
ORA-06512: at line 32
ORA-06512: at line 32

[ErmanArslansOracleBlog]

Probably a wrong/missing record in FND_NODEs.

Did you try ;

stopping app services , running FND conc clone setup clean, and then running autoconfig on db tier and then on apps tier, and starting the apps services again.

Not to forget -> Verify the records before and after ->

  Verify before/after:
  SELECT node_name, status, support_cp FROM fnd_nodes;
  SELECT concurrent_queue_name, target_node FROM fnd_concurrent_queues
  WHERE concurrent_queue_name LIKE 'FNDSM%';

[syedahmed]

Hi Emran,

tried this acctionplan many times but same results

  SELECT node_name, status, support_cp FROM fnd_nodes;
3 rows return with proper hostname for db and application as expected

  SELECT concurrent_queue_name, target_node FROM fnd_concurrent_queues
  WHERE concurrent_queue_name LIKE 'FNDSM%';

NULL no rows

[ErmanArslansOracleBlog]

Check support_cp on FND_NODES. IS it set for the correct node? (If SUPPORT_CP='N' on apps node → that's why no FNDSM rows created)

 Also verify context file:
  grep s_isConc $CONTEXT_FILE
  grep s_cp_twotask $CONTEXT_FILE
  s_isConc must be YES. If NO → edit context, rerun autoconfig.


If you find anything to correct there, correct it and then rerun afdcm037 with hostname param. (match FND_NODES.NODE_NAME)

 If still fails, share:
  - SELECT node_name, support_cp FROM fnd_nodes; output
  - Exact hostname passed to script
  - s_isConc value from context

[syedahmed]

 Also verify context file:
  grep s_isConc $CONTEXT_FILE
<TIER_NODE oa_var="s_isConc">YES</TIER_NODE>
<TIER_NODEDEV oa_var="s_isConcDev">YES</TIER_NODEDEV>

  grep s_cp_twotask $CONTEXT_FILE
  s_isConc must be YES. If NO → edit context, rerun autoconfig.

<CP_TWOTASK oa_var="s_cp_twotask">NEWPDB</CP_TWOTASK>


If you find anything to correct there, correct it and then rerun afdcm037 with hostname param. (match FND_NODES.NODE_NAME)

 If still fails, share:
  - SELECT node_name, support_cp FROM fnd_nodes;
SQL> SELECT node_name, support_cp FROM fnd_nodes;

NODE_NAME                      S
------------------------------ -
dbtest                         N
APPTEST                        Y
AUTHENTICATION                 N

Note: FND_NODES table  >Column name (NODE_NAME#1) showing NULL still

[ErmanArslansOracleBlog]

so you have null record in node name column of the FND_NODES table.
How is it created there? Via autoconfig?
why don't you delete that row and retry?

Bytheway why is it there? you repopulated fnd_nodes via autoconfig run(s) right? So probably, autoconfig gets a null hostname from somewhere, most probably from context file. Is there a null hostname in the context file? Check that as well.
grep -iE "s_dbhost|s_webhost|s_hostname|s_servername|s_apps_jdbc" $CONTEXT_FILE ... something like  (></) means null /empty
Also ensure your hostname in /etc/hosts is resolvable.

[syedahmed]

so you have null record in node name column of the FND_NODES table.  YES

How is it created there? Via autoconfig? YES

why don't you delete that row and retry? which row eactly ? & i  tried many time using  FND_CONC_CLONE.SETUP_CLEAN


Bytheway why is it there? you repopulated fnd_nodes via autoconfig run(s) right? YES

So probably, autoconfig gets a null hostname from somewhere, most probably from context file. Is there a null hostname in the context file?

Check that as well.
grep -iE "s_dbhost|s_webhost|s_hostname|s_servername|s_apps_jdbc" $CONTEXT_FILE ... something like  (></) means null /empty
Also ensure your hostname in /etc/hosts is resolvable.


          <dbhost oa_var="s_dbhost">sdbtest</dbhost>
         <hostname oa_var="s_webhost">saptest</hostname>
         <jdbc_url oa_var="s_apps_jdbc_connect_descriptor">jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS_LIST=(LOAD_BALANCE=YES)(FAILOVER=YES)(ADDRESS=(PROTOCOL=tcp)(HOST=sdbtest.domain)(PORT=1552)))(CONNECT_DATA=(SERVICE_NAME=ebs_NEWPDB)))</jdbc_url>
         <jdbc_connect_alias oa_var="s_apps_jdbc_connect_alias">NEWPDB_BALANCE</jdbc_connect_alias>
         <patch_jdbc_url oa_var="s_apps_jdbc_patch_connect_descriptor">jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=sbtest.domain)(PORT=1552))(CONNECT_DATA=(SERVICE_NAME=NEWPDB_ebs_patch)(INSTANCE_NAME=DEV)))</patch_jdbc_url>
      <host oa_var="s_hostname">saptest</host>

[ErmanArslansOracleBlog]

Ensure you have no name mismatch..  (for instance FND_NODES -> APPTEST, context - >saptest)
patch_jdbc -> sbtest (seems like typo!)

 Do broader grep — you may be missing an empty var:
  grep -oE 's_[a-z_]*host[a-z_]*="[^"]*"' $CONTEXT_FILE | sort -u
  grep -iE 'oa_var="s_[^"]*"[^>]*></' $CONTEXT_FILE
  Second one finds ANY empty oa_var. Share that output.

 Do broader grep — you may be missing an empty var:
  grep -oE 's_[a-z_]*host[a-z_]*="[^"]*"' $CONTEXT_FILE | sort -u
  grep -iE 'oa_var="s_[^"]*"[^>]*></' $CONTEXT_FILE
  Second one finds ANY empty oa_var. Share that output.

[syedahmed]

Hi,

yes i edited & typo

i checked there is no empty oa_vars

i want to tell you 1 issue encountered during application clone the following during adcfgclone.pl appsTier dualfs

i delete the following Index owned by APPLSYS then restart clone got complted sucesfully everything works just have issue in concurrent managers specially registering the FNDSM

FYI

DROP INDEX APPLSYS."FND~NODES~U1";  >>.I drop this INDEX

CREATE UNIQUE INDEX APPLSYS."FND~NODES~U1" ON APPLSYS.FND_NODES
(NODE_NAME#1)
LOGGING
TABLESPACE APPS_TS_TX_IDX
PCTFREE    10
INITRANS   11
MAXTRANS   255
STORAGE    (
            INITIAL          128K
            NEXT             128K
            MINEXTENTS       1
            MAXEXTENTS       UNLIMITED
            PCTINCREASE      0
            BUFFER_POOL      DEFAULT
           );

Unique constraint error (00001) is OK if key already exists

Creating the DBC file...
java.sql.SQLIntegrityConstraintViolationException: ORA-00001: ?? ?????? ????? ?????? (APPLSYS.FND~NODES~U1)
ORA-06512: ??? "APPS.FND_CONCURRENT", line 1653
ORA-06512: ??? "APPS.FND_APP_SERVER_PKG", line 386
ORA-06512: ??? line 1
- Database error modifying the server

ADD call failed with exit code 1

Updating Server Security Authentication
ignoring @DEVPDB specification, using DBC file
java.sql.SQLRecoverableException: IO Error: Invalid number format for port number
Database connection to jdbc:oracle:thin:@host_name:port_number:database failed
Updating Server Security Authentication failed with exit code 1
Restoring DBC file from backed up location /u01/NEWAPP/fs2/inst/apps/DEVPDB_spptest/appltmp/TXK/DEVPDB_Wed_Jul_15_23_22_2026.dbc

adgendbc.sh ended at Wed Jul 15 23:22:28 +03 2026

adgendbc.sh exiting with status 1
ERRORCODE = 1 ERRORCODE_END


adcvmlog.xml renamed to /u01/NEWAPP/fs2/inst/apps/NEWPDB_spptest/admin/log/07191442/adcvmlog.xml.07191445


WARNING: [CVM Error Report]
The following report lists errors encountered during CVM Phase
      <filename>  <return code where appropriate>
  /u01/NEWAPP/fs2/EBSapps/appl/ad/12.0.0/bin/adgentns.pl  2
vi /u01/NEWAPP/fs2/inst/apps/NEWPDB_spptest/admin/log/07191442/adconfig.log

 No. of scripts failed in CVM phase: 1

[ErmanArslansOracleBlog]

Dropping FND~NODES~U1 removed the safety net → NULL/duplicate rows now allowed.

First;

you should cleanup the null rows and duplicate rows and keep only valid rows in fnd_nodes. and then recreate that unique standard index. (Do NOT skip index recreation — that's why NULL keeps coming back.)

Then you should check the hostnames. You've had three hostnames across sessions: saptest, APPTEST, spptest. Pick the real one(s)

Fix $CONTEXT_FILE if any variant is wrong. Also DEVPDB vs NEWPDB appears in your logs — service name mismatch.

aAso you got "Invalid number format for port number" . DBC file has literal placeholder text. Likely s_apps_jdbc_connect_descriptor malformed, or FND_NODES.SERVER_ADDRESS contains junk. Check:
  SELECT node_name, server_address FROM fnd_nodes;

So basically you should;

drop bads rows and recreate the index, fix hostname/service in the xontext, run autoconfig on apps tier and run afdcm037.sql with correct hostname.

[syedahmed]

Unable to recreate this INdex using following script

CREATE UNIQUE INDEX APPLSYS."FND~NODES~U1" ON APPLSYS.FND_NODES
(NODE_NAME#1)
LOGGING
TABLESPACE APPS_TS_TX_IDX
PCTFREE    10
INITRANS   11
MAXTRANS   255
STORAGE    (
            INITIAL          128K
            NEXT             128K
            MINEXTENTS       1
            MAXEXTENTS       UNLIMITED
            PCTINCREASE      0
            BUFFER_POOL      DEFAULT
           );

[ErmanArslansOracleBlog]

You don't share the error you got !

probably due to duplicates on NODE_NAME#1 (which is UPPER(NODE_NAME)).

find them and delete them, and verify the remaining one and then recreate the index.

 Share the exact ORA error if still fails.!

[syedahmed]

getting this Error

[Error] Execution (1: 55): ORA-01452: cannot CREATE UNIQUE INDEX; duplicate keys found

[ErmanArslansOracleBlog]

Okay in my last update, I already told you what to do.

[syedahmed]

the index which i drop was Edition based Index is thing it was with APPLSYS."FND~NODES~U1"

The Actual Index is  APPLSYS.FND_NODES_U1

DROP INDEX APPLSYS.FND_NODES_U1

CREATE UNIQUE INDEX APPLSYS.FND_NODES_U1 ON APPLSYS.FND_NODES
(NODE_NAME)
LOGGING
TABLESPACE APPS_TS_TX_IDX
PCTFREE    10
INITRANS   11
MAXTRANS   255
STORAGE    (
            INITIAL          128K
            NEXT             128K
            MINEXTENTS       1
            MAXEXTENTS       UNLIMITED
            PCTINCREASE      0
            BUFFER_POOL      DEFAULT
           );

i tried to recreate the following Index which i droped but same error due to the actual Index exist

[ErmanArslansOracleBlog]

Hmm..   The ~ index enforces uniqueness in editioning layer — that's why NULLs kept surviving during autoconfig.
That's EBR (Edition-Based Redefinition). You can't CREATE INDEX manually on the ~ name. It must be recreated as an editioning object via "ad_zd".

AD_ZD_TABLE.UPGRADE..