adop fails after cloning

[hidayat]

[applmgr6@SCCHODDBAP FMW_Home]$ /p06/appleam/fs2/FMW_Home/oracle_common/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="/p06/appleam/fs2/FMW_Home/oracle_common" ORACLE_HOME_NAME="/p06/appleam/fs2/EBSapps/10.1.2" CLUSTER_NODES="{}"
Starting Oracle Universal Installer...

Checking swap space: must be greater than 500 MB.   Actual 398713 MB    Passed
Preparing to launch Oracle Universal Installer from /tmp/OraInstall2016-02-07_01-48-57PM. Please wait ...[applmgr6@SCCHODDBAP FMW_Home]$ The inventory pointer is located at /etc/oraInst.loc
The inventory is located at /p06/appleam/oraInventory
'AttachHome' failed.

[hidayat]

Hi Erman,

I tried to attach the home but it failed.
do i have to re configure apps tier?

Thanks

[ErmanArslansOracleBlog]

First of all, your  ORACLE HOME Name is wrong.
Setting it like this : ORACLE_HOME_NAME="/p06/appleam/fs2/EBSapps/10.1.2" is weird.

Please read the documents carefully.

You can re-register any one of the 3 FMW application tier homes as follows. Important, you must run these for both file systems (fs1 and fs2)

<path to fs1>/FMW_Home/oracle_common/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="<path to Oracle Home>" ORACLE_HOME_NAME="Oracle Home Name" CLUSTER_NODES="{}"
<path to fs2>/FMW_Home/oracle_common/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="<path to Oracle Home>" ORACLE_HOME_NAME="Oracle Home Name" CLUSTER_NODES="{}"

e.g.
/u01/test/fs1/FMW_Home/oracle_common/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="/u01/test/fs1/FMW_Home/oracle_common" ORACLE_HOME_NAME="VIS_TOOLS__u03_VIS_fs1_EBSapps_10_1_2" CLUSTER_NODES="{}"
/u01/test/fs2/FMW_Home/oracle_common/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="/u01/test/fs1/FMW_Home/oracle_common" ORACLE_HOME_NAME="VIS_TOOLS__u03_VIS_fs2_EBSapps_10_1_2" CLUSTER_NODES="{}"

 

If you need to register the tools (10.1.2) home you can use these commands, again you will need to run this for both fs1 and fs2:

<path to fs1>/EBSapps/10.1.2/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="<path to the 10.1.2 Home>" ORACLE_HOME_NAME="<Oracle Home Name>"
<path to fs2>/EBSapps/10.1.2/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="<path to the 10.1.2 Home>" ORACLE_HOME_NAME="<Oracle Home Name>"

e.g.
/u01/test/fs1/EBSapps/10.1.2/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="/u01/VIS/fs1/EBSapps/10.1.2" ORACLE_HOME_NAME="VIS_TOOLS__u03_VIS_fs1_EBSapps_10_1_2"
/u01/test/fs2/EBSapps/10.1.2/oui/bin/runInstaller -silent -attachHome -invPtrLoc /etc/oraInst.loc ORACLE_HOME="/u01/VIS/fs2/EBSapps/10.1.2" ORACLE_HOME_NAME="VIS_TOOLS__u03_VIS_fs2_EBSapps_10_1_2"



Make sure each ORACLE_HOME_NAME is unique. Once you have completed the steps for all missing Oracle Homes confirm that the inventory.xml file now contains all missing homes. Log files for the attachment process can be found at oraInventory/logs

[ErmanArslansOracleBlog]

You dont have the second filesystem info in your inventory as you see, so your inventory does not include oracle home for fs2, so it is missing.
Follow this document:
R12.2 How To Re-attach Oracle Homes To The Central Inventory (Doc ID 1586607.1)

[ErmanArslansOracleBlog]

Here is a proper inventory file for you. You see? It contains the oracle homes from both fs1 and fs2.
So if your inventory does not contain the oracle homes from both filesystems, reattach them.
(not that: this inventory is taken from a clone system, which was running on a single server(app+db on single server, so thats why the inventory contains the db oracle home as well.)

<?xml version="1.0" standalone="yes" ?>


<INVENTORY>
<COMPOSITEHOME_LIST>
</COMPOSITEHOME_LIST>
<VERSION_INFO>
   <SAVED_WITH>10.1.0.6.0</SAVED_WITH>
   <MINIMUM_VER>2.1.0.6.0</MINIMUM_VER>
</VERSION_INFO>
<HOME_LIST>
<HOME NAME="CLONE_DB__u01_install_CLONE_11_2_0" LOC="/u01/install/CLONE/11.2.0" TYPE="O" IDX="1"/>
<HOME NAME="OH688521181" LOC="/u01/install/fs1/FMW_Home/oracle_common" TYPE="O" IDX="6">
   <REFHOMELIST>
      <REFHOME LOC="/u01/install/fs1/FMW_Home/webtier"/>
   </REFHOMELIST>
</HOME>
<HOME NAME="OH1720757241" LOC="/u01/install/fs1/FMW_Home/webtier" TYPE="O" IDX="7">
   <DEPHOMELIST>
      <DEPHOME LOC="/u01/install/fs1/FMW_Home/oracle_common"/>
   </DEPHOMELIST>
</HOME>
<HOME NAME="OH710303641" LOC="/u01/install/fs1/FMW_Home/Oracle_EBS-app1" TYPE="O" IDX="8"/>
<HOME NAME="CLONE_TOOLS__u01_install_fs1_EBSapps_10_1_2" LOC="/u01/install/fs1/EBSapps/10.1.2" TYPE="O" IDX="9"/>
<HOME NAME="OH621514428" LOC="/u01/install/fs2/FMW_Home/oracle_common" TYPE="O" IDX="2">
   <REFHOMELIST>
      <REFHOME LOC="/u01/install/fs2/FMW_Home/webtier"/>
   </REFHOMELIST>
</HOME>
<HOME NAME="OH1577137702" LOC="/u01/install/fs2/FMW_Home/webtier" TYPE="O" IDX="3">
   <DEPHOMELIST>
      <DEPHOME LOC="/u01/install/fs2/FMW_Home/oracle_common"/>
   </DEPHOMELIST>
</HOME>
<HOME NAME="OH741323448" LOC="/u01/install/fs2/FMW_Home/Oracle_EBS-app1" TYPE="O" IDX="4"/>
<HOME NAME="CLONE_TOOLS__u01_install_fs2_EBSapps_10_1_2" LOC="/u01/install/fs2/EBSapps/10.1.2" TYPE="O" IDX="5"/>
</HOME_LIST>
</INVENTORY>

[hidayat]

Hi erman,

how are you?
we have multinode environment.
 i ran adop prepare phase on primary node p1, we had some ts mismatch issues, we resolved the issues, prepare phase  completed successfully, then we ran abort phase on the master node p1 only,  it completed successfully as well. after that we ran cleanup phase on master node p1 , cleanup phase completed successfully. when we check the adop -status we are getting following warning message while running adop -status.

WARNING: One or more application tier nodes were abandoned during this patching cycle.
 You must restore or delete these nodes by following the instructions in
 My Oracle Support Knowledge Document 1677498.1

Everything seems to be working normal in apps and db, is this normal to see this because we only performed prepare action on node 1 only?

Please advise

[hidayat]

[applmgr@SCCHOPAP1 ~]$ adop -status

 Enter the APPS password:




 Current Patching Session ID: 30

 Node Name Node Type Phase Status Started Finished Elapsed
 --------------- --------------- --------------- --------------- ------------------------------ ------------------------------ ------------
 scchopap1 master PREPARE SESSION ABORTED 16-FEB-16 15:14:31 +03:00 16-FEB-16 15:22:42 +03:00 0:08:11
 APPLY SESSION ABORTED
 FINALIZE SESSION ABORTED
 CUTOVER SESSION ABORTED
 CLEANUP COMPLETED 17-FEB-16 12:02:55 +03:00 17-FEB-16 13:47:31 +03:00 1:44:36

 scchopap2 slave PREPARE NOT STARTED
 APPLY NOT STARTED
 FINALIZE NOT STARTED
 CUTOVER NOT STARTED
 CLEANUP NOT STARTED

 scchopdmz slave PREPARE NOT STARTED
 APPLY NOT STARTED
 FINALIZE NOT STARTED
 CUTOVER NOT STARTED
 CLEANUP NOT STARTED




 File System Synchronization Used in this Patching Cycle: Full



 INFORMATION: Patching cycle aborted, so fs_clone will run automatically on scchopap1 node in prepare phase of next patching cycle.



 WARNING: One or more application tier nodes were abandoned during this patching cycle.
 You must restore or delete these nodes by following the instructions in
 My Oracle Support Knowledge Document 1677498.1

 For more information, run ADOP Status Report by using -detail option


 Generating ADOP Status Report at location: /scclive/applmgr/SCC/fs_ne/EBSapps/log/status_20160217_134746/adzdshowstatus.out
 Please wait...
 Done...!

[ErmanArslansOracleBlog]

Send me the file: $ADOP_LOG_HOME/<latest_session_id>/adop_latest_timestamp.log