Would you use scripts to take database out of archivelog mode?

Question

Background: we do not use RMAN for this production database and we do cold backups each night.
I've been asked to take one of our databases out of archivelog mode for a few days and then to put it back into archivelog mode.
Being somewhat neurotic about these things - I've only ever done it manually - so that I can make sure nothing goes wrong and to check that there was a last successful cold backup prior to the switch.
However, management have asked that it be done automatically by scripts after the backup.
Whilst I can create a script that will do this - the question I'd like to ask the forum is - have you and would you?

Answer 1

I do not agree with taking any Oracle database out of archivelog mode and I do not approve of management's decision and have told them why and all the risks they run.
This is not a test database - it IS a production database and other users do make changes daily.
What I did not reveal until now is that I have also been asked to take this database out of archivelog mode for 5 days and suspend all cold back-ups for the whole server (with 3 other instances on the server - also impacted) - and have warned them accordingly.
However, management have been swayed by someone who feels they know better - who claims to be a qualified Oracle DBA - I'm OCP but seems to have little sway since this person has convinced management that my 6 years experience and qualification is less than his application experience in GIS.
They want to run an upload which will take around 5 days to do - I'm somewhat miffed that an upload can take that length of time and question whether a GIS system should need such an upload for such a length of time.
Let's no go into the problem with tablespaces and space within the tablespaces - the upload will fail because they do not understand that if the tablespace fills up - the upload will fail - but despite asking - I've been told that there is no problem - there will be a problem - since I refuse to allow autoextend on this database - because of the potential to run out of disk space.
I've explained until I'm blue in the face - never mind that even if the upload is successful - the cold backup tape won't be able to backup the data because it'll be potentially over the limit.
I've even warned of my past experience x 3 - when a RAID controller failed - literally died in front of my eyes - I just happened to be standing next to one of the servers talking to someone - and the whole server was lost - but because we had the archivelogs - and the previous day's backup - we got it up and running after Dell replaced the RAID controller - within 1 hour.
That's the joy of Oracle - nothing is lost other than the last archivelog - depending on the time you set the switch.
I love this system - it's so recoverable - it such a joy to work with.
But then when I get told to suspend cold backups and then at the same time suspend archivelog mode - I go mad.
Don't get me started - I've gone down the usual path of why hire an OCP when you won't listen to what I say, etc.
Management are demanding I switch it automatically out of archivelog mode etc.
I can do it - but my question is still - have you guys?
And why did you do it?
All I can say is that I hope that the server has a RAID contoller failure and finally they listen to me - the chances of that are minimal - so management are saying - do it or else.
Oh and by the way - the GIS team have asked for this to be bi-monthly - automated.
And have said that:
"You can not rollback or recover the loading process using the log files because it involves a far more complex operation that would most likely not be successful."
Nice phrase that to an OCP.
I welcome your comments - as I really am in a hole here.
Thanks.