Re: Taking Binary backup while instance is running

  • From: Subodh Deshpande <deshpande.subodh@xxxxxxxxx>
  • To: ora-apps-dba@xxxxxxxxxxxxx
  • Date: Mon, 13 Dec 2010 16:03:18 +0530

does the environment variable points to proper oracle home..
it sounds like you also had changed oracle home, if so then is the new
tnsnames ora points to correct databases with new home(check new
listener.ora) and did you checked the connectivity from any client with a
query not just with tnsping

from where these ETL jobs are fired are they dba jobs..
after patching did you comipled all the objects and further are there any
database links being used, are these links valid

On 13 December 2010 13:20, Shastry(DBA) <shastry17@xxxxxxxxx> wrote:

> Hi All,
>    I recently had a task on CPU Patching July 10 on my production instance
> (runs with HP-UX 64 bit PA-RISC). We had less time to apply the patch on
> production as it took more time on Dev and Stage. We took an online backup
> of binary and renamed it as 9208_BAK and exported all the required
> parameters like LD_LIBRARY_PATH, ORACLE_HOME,  PATH, usr/bin/cc etc to point
> to the backed-up oracle home on the same server. The opatch applied
> successfully on new oracle home and kept it ready for catcpu.sql to run
> during the downtime window. Post moving the oracle home back to 9208,
> applied @catcpu.sql successfully. Observations found nothing on alert log
> and patch logs. Now some of the application ETL jobs facing TNS -host lost
> contact post CPU patch(No changes done at apps side). Please help in
> understanding where exactly issue lies when am able to tnsping from external
> host and connect to DB without any issue. what might have created issues in
> terms of libraries, oracle binary files?
> Thanks,
> Anant


Other related posts: