Re: ARCn
From: Nuno Souto <dbvision_at_iinet.net.au>
Date: Fri, 02 Aug 2013 20:40:40 +1000
Message-ID: <51FB8CA8.2080209_at_iinet.net.au>
On 2/08/2013 6:28 AM, Mark W. Farnham wrote:
>> reading member one, continue with member one of the next group
>> unless we end up switching due to corruption, at which time we
>> switch to member 2 and stick with that series unless we are forced
>> to switch again."
Date: Fri, 02 Aug 2013 20:40:40 +1000
Message-ID: <51FB8CA8.2080209_at_iinet.net.au>
On 2/08/2013 6:28 AM, Mark W. Farnham wrote:
>> one it reads? I personally would program this to say "start by
>> reading member one, continue with member one of the next group
>> unless we end up switching due to corruption, at which time we
>> switch to member 2 and stick with that series unless we are forced
>> to switch again."
I did some experimentation with this a few years ago - 9ir2, IIRC.
In a nutshell, I did setup 4 groups with 2 members each, "a" and "b".
redo01a,redo02b,redo03a,redo04b in file system /01,
then
redo01b, redo02a, redo03b, redo04a in file system /02.
Then watched where the reads were going as they got archived.
It was always the first created - which in my case was the "a" members.
Not sure if that is valid nowdays, outside of 9i. Can't see why not?
Perhaps someone can repeat the test with the release in question?
-- Cheers Nuno Souto dbvision_at_iinet.net.au -- http://www.freelists.org/webpage/oracle-lReceived on Fri Aug 02 2013 - 12:40:40 CEST