If Target mid-tier and Source mid-tier are on the same server, EBS clone script adcfgclone.pl may fail with port conflict error.
Prots used for an EBS instance is listed in file $INST_TOP/admin/out/portpool.lst. Depending on which port has conflict on the server, the cloning error message will be different.
- If the port for httpd is busy and used by another environment on the server, the cloning script will stop on Target instance with errors:
ERROR: Failed to configure the target system,
please check the logfile in : $RUN_BASE/inst/apps/$CONTEXT_NAME/admin/log/clone
CLONE-26003 Error in validating listen host and port.
CLONE-26176 In config group httpd.conf , the value of "Listen" config property was xxxx. xxxx was not free.
Log message shows the problem is from file httpd.conf. When I checked the port number and compare it in file $CONTEXT_FILE and in file $FMW_HOME/webtier/instances/EBS_web_<SID>_OHS1/config/OHS/EBS_web_<SID>/httpd.conf.
$ grep Listen httpd.conf
$ grep s_http_listen_parameter $CONTEXT_FILE
I saw they are different in Source instance. Even they are not the same, Apache still works fine on Source instance and httpd.worker processes occupy the port number in httpd.conf. After the clone script copied file httpd.conf to Target instance, Apache failed to start because OHS was running on the same port as the other instance
Apparently httpd.conf of Source instance was edited by picking up randomly a port number. httpd.conf can be edited and updated manually as AutoConfig in R12.2 does not update it. For more details on modifying port values for OHS, see Doc ID 1905593.1 (Managing Configuration of Oracle HTTP Server and Web Application Services in Oracle E-Business Suite Release 12.2).
- While cloning script uses a temporary port but it was not available, the error could be
CLONE-20372 Server port validation failed.
The fix could be just shutdown apps services of Source instance to let the cloning complete. See Doc ID 2002613.1
- Doc ID 2437111.1 gives a fix on port for s_ohs_adminport.
On Linux server, use command to find if a port is used or not:
$ netstat -tuanp | grep 6230
(Not all processes could be identified, non-owned process info
will not be shown, you would have to be root to see it all.)
tcp 0 0 167.69.109.82:6230 0.0.0.0:* LISTEN
Then the process owner (or root) can use below line to find what process is using the port:
$ lsof -i :6230
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
opmn 31876 ebsdev 9u IPv4 42886593 0t0 TCP server3d.domain.com:6230 (LISTEN)
After adcfgclone.pl failed, you have to remove/rename folders BEFORE re-run clone script adcfgclone.pl
- Remove two new folders under $RUN_BASE (or only folder $RUN_BASE/FMW_Home, depending on the failure stage) to avoid error "Exiting cloning as FMW Home already exists".
- Also remove the folder defined by inventory_loc in file /etc/oraInst.loc to avoid error "Oracle Homes are already registered in the inventory". If the inventory is just for one environment, do not try to detach a home because it may give error:
$ ./runInstaller -detachhome $FMW_HOMEe/oracle_common
Starting Oracle Universal Installer...
Checking swap space: must be greater than 500 MB. Actual 13519 MB Passed
The inventory pointer is located at /etc/oraInst.loc
The inventory is located at ... ...
The operation failed as it was called without name of the Oracle Home being attached.
- Since the script failed before the completion, there is not need to remove info in the database.
