000029158 - IMG V6.8.1 installation stalls during Oracle installation portion of install process

Document created by RSA Customer Support Employee on Jun 15, 2016Last modified by RSA Customer Support Employee on Apr 21, 2017
Version 2Show Document
  • View in full screen mode

Article Content

Article Number000029158
Applies ToAveksa Appliance
Jboss Application Server
Versions 6.8.1  (note that this process can be used for any version)
IssueProblem Summary;
After rebuilding an Aveksa appliance, and running the install.sh script as outlined in the Installation documentation for Jboss appliance, its noted that the installation seems to hang while installing Oracle.  No errors are noted in the /tmp/aveksa-install.log file, and the /tmp/aveksa/oracle.log file is not created.
The output on the terminal shows:

Installing Oracle ASM drivers... 
Red Hat Enterprise Linux Server release 5.8 (Tikanga) 
Red Hat EL 5 64-bit 
installing oracleasm OS rpms 
cleaning up directory /tmp/asmlib 
configuring oracleasm 
Installing Oracle...

Since there were no errors in the default log files, the next step to to try and obtain additional diagnostics from the install scripts.  The execution of install.sh was terminated with ctrl-c.
At this point, since the orignal installation was not sucessful and the install terminated, per the Installation documentation, /tmp/aveksa/staging/deploy/uninstall.sh was run. 
Next, to obtain additional installation information, a  -x  was added to the first line of the Oracle install script, /tmp/aveksa/staging/deploy/installOracle.sh, so that the script output would be more verbose and 
hopefully provide a clue to the underlying problem.
The first line was changed from  #!/bin/bash    to   #!/bin/bash -x   and install.sh was re-executed.
When the Oracle install starts, we see more verbose output, no errors are seen but we now see the specific line that the script is now stopping at seems to be the sntp command, as shown below:
+ chown oracle:oinstall /u01
+ chmod ugo+r /tmp/aveksa/staging/deploy/db.rsp /tmp/aveksa/staging/deploy/grid.rsp /tmp/aveksa/staging/deploy/install_WAS85.rsp
+ '[' -f /etc/profile.d/oracle.sh ']'
+ '[' -f /etc/profile.d/oracle.csh ']'
+ unset ORA_CRS_HOME
+ unset ORA_ASM_HOME
+ pgrep ntpd
+ '[' -f /etc/ntp.conf ']'
++ grep '^fudge' /etc/ntp.conf
++ wc -l
+ '[' 1 -eq 0 ']'
++ grep '^server' /etc/ntp.conf
++ wc -l
+ '[' 4 -eq 0 ']'
+ sntp -P no -r 0.pool.ntp.org


Oracle installations require that the ntpd server is running and correct.  Checking that status of the

server showed that the ntpd service was not running:

#service ntpd status

ntpd is stopped

The ntpd service was started using this command:

# service ntpd start

ntpd: Synchronizing with time server:                      [  OK  ]

Starting ntpd:                                             [  OK  ]

Once the ntpd service was started, the Oracle portion of the ACM installation completed without error.

The original Oracle install script, /tmp/aveksa/staging/deploy/installOracle.sh,  was restored to it's 

default state by removing the -x option on the first line.