Re: Why can't an RMAN job be edited in 10G dbcontrol?

From: GS <>
Date: Wed, 04 Feb 2009 21:38:03 GMT
Message-ID: <%Unil.9831$Db2.1735_at_edtnps83>

joel garry wrote:
> On Feb 4, 8:04 am, GS <> wrote:
>> I set up a nightly backup job on a new database I just moved over to a
>> Windows 2008 server, and I was sure I told it to keep 3 days worth of
>> backups then delete obsolete backups. The problem is that it is not
>> deleting  the older backups, I have to do it manually, and I can't get
>> into the job to see or change what it is doing. Dbcontrol just tells me
>> that "Edit is not supported for this job type, only general information
>> about the job can be updated".
>> So now it looks like I have to either just create a new job and delete
>> this one, or just write the RMAN scripts and put them in scheduled tasks
>> so I can at least tweak or change them later if I want to.
>> I guess this is more of a rant than a question, I have about 1/2 my DB's
>> on 10G now and I am NOT getting used to the dbcontrol gui, in fact I
>> think it is a bloated crappy gui interface.  The 9i enterprise manager
>> with a management server IMHO was/is a far better tool. I suppose I
>> should give grid control another go, but after fighting with it for 2
>> days straight the last go around (and still not working) I am a tad
>> soured on it. IMO a tool should be simple to implement, not be more
>> complex than what you are trying to administer with it. </rant>

> There are different ways to specify things, show all in rman to show
> us which you are using.
> It works for me, in fact, it was a bit too aggressive until I figured
> out the redundancy can include 1 for a backup and another for the
> archivelog backup in the same script. Then it all stopped working for
> no reason at all. So I ranted like you (except I just used cron and
> scripts in 9i). Now it works again, the reconfigure worked this time
> (as opposed to last time, when it made things worse).
> I quickly learned to keep a text file of any scripts I'm running. The
> space reclamation stuff is simply bizarre. Yesterday it told me a
> table was 64G, with 62G reclaimable. In a 46G tablespace. That's 98%
> full. Oh wait, it's 93% full. And why does it ignore my changes to
> the warning of percentage full? Sometimes? And java 500 errors when
> I try to get a tablespace map. Sometimes. Then not other times.
> You'd think by they'd get it at least close to being right.
> But maybe the stupid architecture is so asynchronous it just splits
> its guts into parallel worlds. And any actual technical question on
> the oracle forum is just chirping crickets. It's not even worth it
> going to metalink until things are a smoking pile.
> jg
> --
> is bogus.

Thanks JG and CH

I was so caught up trying to do it via the GUI I forgot the obvious. I checked in RMAN prompt and my retention policy was the 3 days. I don't know why it isnt deleting them though, I'll change it to two days and see if it works tonight.

One thing I forgot to mention was that I also wanted to tweak the script so that it didn't delete the archives after backing them up, I want to keep them around for a day or so so I can use them to bring clones up to date when I build clones from a user-managed online backup. I'll just do this from the command line though, I need to brush up a bit on the RMAN syntax anyway.

btw, my opinion on dbcontrol is the same, I still think it stinks.. Received on Wed Feb 04 2009 - 15:38:03 CST

Original text of this message