RE: expdp - cell single block physical read/ read request

From: Mark W. Farnham <mwf_at_rsiz.com>
Date: Thu, 16 Dec 2021 09:57:02 -0500
Message-ID: <6f8e01d7f28d$2f80ce00$8e826a00$_at_rsiz.com>



I believe you raise another valid reason which might be common. Tim Gorman also mentioned small tables.  

(I didn’t see that after my first export failed, since I made it a habit that IF I needed an export I would arrange for quiescence and single row scan the table before I exported it.)  

I nearly always used “hot backup” for the whole database once we got to 6.0.36 and sqlplus COPY for individual things, but took special care pre-emptively if an export was truly required.  

IF you do only need part, a sqlplus COPY to the same machine was a neat magic trick as far as network speed, giving you a nice clean thing to export as long as you had the room.  

Repeat after me: “More Disk”  

mwf  

From: Andrew Kerber [mailto:andrew.kerber_at_gmail.com] Sent: Thursday, December 16, 2021 9:39 AM To: Mark W. Farnham
Cc: Jack van Zanen; ORACLE-L
Subject: Re: expdp - cell single block physical read/ read request  

Doesnt that mean its reading the undo tablespace for transaction consistency?  

On Thu, Dec 16, 2021 at 8:37 AM Mark W. Farnham <mwf_at_rsiz.com> wrote:

If memory serves this can happen with migrated rows, multi-block rows, and/or multi-piece rows. Or at least some of those reasons.  

Oracle is going to want the contents of your row to be contiguous in your export file, so it needs all the pieces. I can’t remember what happens with in-line and out of line binary object columns.  

mwf  

From: oracle-l-bounce_at_freelists.org [mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Jack van Zanen Sent: Wednesday, December 15, 2021 9:59 PM To: oracle-l_at_freelists.org
Subject: expdp - cell single block physical read/ read request  

Hi    

Oracle 12.2.0.1 Exadata  

I am running a datapump and two of the workers are spending a lot of time on
- cell single block physical read

This does not make sense to me as it is full expdp I would expected a full table scan type of wait, not single block  

I checked mos but my search did not find anything related to my version  

Does anyone know of any reason why this might happen?    

Jack van Zanen  

  <https://docs.google.com/uc?id=0BwovDucFT1fXaEREVHNWRWZyNjg&export=download&resourcekey=0-139CmIg7avLkAQJI-NUXxg>



This e-mail and any attachments may contain confidential material for the sole use of the intended recipient. If you are not the intended recipient, please be aware that any disclosure, copying, distribution or use of this e-mail or any attachment is prohibited. If you have received this e-mail in error, please contact the sender and delete all copies. Thank you for your cooperation
-- 

Andrew W. Kerber

'If at first you dont succeed, dont take up skydiving.'





--
http://www.freelists.org/webpage/oracle-l
Received on Thu Dec 16 2021 - 15:57:02 CET

Original text of this message