Re: Oracle errors with Solaris and Raid Manager

From: Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk>
Date: Wed, 10 Feb 1999 19:51:19 -0000
Message-ID: <918676274.9500.0.nnrp-06.9e984b29@news.demon.co.uk>

Are you running with a 16K block size ? It may be a coincidence, but the example you quote multiplied by 16K gives 2GB - are the files exactly 2Gb or larger than 2Gb ?

There are still Oracle products that cannot cope with files larger than 2gb even on the Unix platforms that can.

Jonathan Lewis
Yet another Oracle-related web site: www.jlcomp.demon.co.uk

Olaf wrote in message
<240377D26A155C8B.07CD1B51D3E108CB.8D59B498B2589EAF_at_library-proxy.airnews.ne t>...
>I am running Solaris 2.6 and Raid Manager 6.1.1 along with Oracle 8.05.00.
>We are not using raw devices. From the UNIX side of things, everything
>looks fine. The controller check out OK under Raid Manager, and fsck turns
>up nothing wrong on any of the filesystems.
>
>However, we get the following types of errors when we some kinds of
>dataloads:
>
>
>SQL> analyze table monthly_bookings compute statistics;
>analyze table monthly_bookings compute statistics
>*
>ERROR at line 1:
>ORA-01115: IO error reading block from file 20 (block # 131072)
>ORA-01110: data file 20: '/u08/oradata/cosi/sist_idx_02_d.dbf'
>ORA-27072: skgfdisp: I/O error
>SVR4 Error: 34: Result too large
>Additional information: 131071
Received on Wed Feb 10 1999 - 13:51:19 CST