Re: How big is an SCN ? (was I/O activity during HOT backups)

Oweson, An SCN is at least 48 bits long. Thus they can be allocated at a rate of 16,384 SCN’s per second for over 534 years without running out of them. At this rate, an Oracle database will run out of SCN’s in June, 2522 AD (using 31 day months for time stamps).

Hope this is what you are looking for.

Scott
> Hi Gang,
>
> I was giving the 8i Backup & Recovery course last week for Oracle (great way
> of getting all the latest training ...), and I explained to the students how
> the hot backups worked. I said, as has been stated, that when you put the
> tablespace (or datafile) into hot back-up mode, the SCN in the header is
> frozen.
>
> On of my students then asked what is the maximum size of an SCN - just how
> big can it get? I popped off to the WWS (WorldWide Support) desk, and asked
> them to investigate. The answer I got back was 'bigger than you'll ever
> need'. I wasn't happy with that, and wanted to know (in bytes) what the
> size of the SCN was.
>
> I was then told that there are different SCNs in the Header (of the file)
> and in each of the Data Blocks, and these SCNsare of different sizes!
> Apparently the 'Header' SCN is made up of two parts, each of which seemed to
> consist of 8 hexidecimal digits. The 'Block' SCNs only had the 8
> hexidecimal digits.
>
> Can anyone confirm this / shed more light on this matter?
>
>
> Secondly, according to the course notes, you can do 'hot' backups using
> RMan, which will only log changed blocks - resulting in smaller and faster
> backups. I saw that some of the RMan procedures are now integrated into the
> Oracle kernel (with 8i, not 8 - which means they can be accessed with the
> instance started but not even mounted). Are these the routines used? How
> does Oracle (and RMan) keep track of changed blocks, and what happens if the
> instance crashes while the RMan has an incremental backup running?
>
> There were no explantations in the notes at all! We had to assume (as an
> object of faith) that the instance would recover fine - this is VERY
> different if the instance crashes doing a 'normal' hot backup! I would sleep
> better (and would be able to recommend using RMan with good concience) if I
> understood how it worked ...
>
>
> Regards
> Oweson
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Oweson Flynn
> Senior Oracle DBA
> Email: OracleDBA_at_flysaa.com
> Tel: 978-9826
> Cell: 082-600-7-006
> Fax: 978-1568
>
>
>
>
> --
> Author: Oracle DBA - Oweson Flynn
> INET: OracleDBA_at_flysaa.com
>
> Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> San Diego, California -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from). You may
> also send the HELP command for other information (like subscribing).
Received on Wed Jun 21 2000 - 21:50:22 CDT