Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: ora-1652 on *undo* tablespace?

Re: ora-1652 on *undo* tablespace?

From: Chuck <skilover_nospam_at_bluebottle.com>
Date: Thu, 08 Mar 2007 16:18:54 GMT
Message-ID: <OHWHh.2027$3i.437@trnddc01>


Chuck wrote:

> Anurag Varma wrote:

>> This is strange. So according to you the active extents showed up as
>> 1M
>> and oracle returns an out of space error?
>> Is this a RAC instance?
>> How is undo tablespace created? LMT or DMT?
>> Are the datafiles in undo tablespace set to autoextend?
>> What was the transaction running? A big insert/update/delete?
>>
>> It just seems like a bug that you find active undo segments using
>> only 1M of space in a 7G tablespace and get an out of space error?
>>
>> Try checking metalink for bugs related to AUM in 9i.
>>
>> Anurag
>>
> 
> 
> Not RAC - single instance. I am thinking it's a bug too. Especially
> after seeing in the alert log that the ora-1555 that appeared a while
> later says the query duration was over 37 years.

Mea culpa. That was a different ora-1555 error. The next one after the ora-1652 says that the query ran for only 10 seconds. I find that very hard to believe given the retention period configured (3h) and the amount of unexpired undo extents at the time (2.5g). I think I am probably hitting one of those v9.2 undo bugs that someone else mentioned. Received on Thu Mar 08 2007 - 10:18:54 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US