Re: Large SGA: sgabeg, ORA-04030,

From: <manager43_at_my-deja.com>
Date: Tue, 16 Nov 1999 09:29:16 GMT
Message-ID: <80r85c$k54$1@nnrp1.deja.com>

Hi,

Workaround -

Reduce kernel size via options linked in
See article Prob# 1014589.6 on the metalink HOW TO CALCULATE THE VALUE FOR SGABEG WHEN RELOCATING THE SGA ON SOLARIS
See article Prob# 1026731.6 on the metalink HOW TO VERIFY THE LOCATION OF THE SGABEG ADDRESS

Vladimir

In article <383024d1.2258704_at_news.earthlink.net>, andreyNSPAM_at_bookexchange.net (NetComrade) wrote:
> Hey all,
>
> Does anybody know why this could be coming up upon an index creation?
>
> ORA-04030: out of process memory when trying to allo
> cate 4112 bytes (callheap,sort area)
> ORA-07324: smpall: malloc error while allocating pga.
>
> Oracle 7.3, E4500, 10Gig Ram.
> 3 Oracle DB's running at 2.5, 1.5, 2 Gig each.
>
> Also it seems that when I decrease the sort_area_size from 8Meg to 4
> Meg that error disappears (we have this index creation in a batch job)
>
> Also, I have done something according to the
> SGA size, sgabeg attach address and Sun architectures 61896.1
> article on metalink.
>
> From paper:
> If you have successfully reduced the value of sgabeg, and subsequently
> receive
> errors:
> ORA-4030: out of process memory when trying to allocate %s bytes
> (%s,%s)
> or
> ORA-7324: smpall: malloc error while allocating pga.
> then it's very likely that you've lowered the start address into an
> area which
> the PGA needs to extend into. Raise sgabeg, and try again.
>
> Does anybody know how I should resize, play around with those
> parameters?
>
> Thank you.
> ---------------
> Andrey Dmitriev eFax: (978) 383-5892 Daytime: (917) 373-5417
> AOL: NetComrade ICQ: 11340726 remove NSPAM to email
>
>

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Tue Nov 16 1999 - 03:29:16 CST