From: "Richard Sutherland" <rvsutherland@yahoo.com>
Newsgroups: comp.databases.oracle.server
Subject: Re: Oracle 8.1.6.1 import
Date: Mon, 1 Jan 2001 18:09:29 -0500
Organization: Posted via Supernews, http://www.supernews.com
Message-ID: <t52373br2un962@corp.supernews.com>
References: <92oipv$vei$1@nnrp1.deja.com> <t50os96mdish93@corp.supernews.com> <t512bt44q0fv06@corp.supernews.com> <92qg2d$82n$1@nnrp1.deja.com>
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 5.00.2615.200
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
X-Complaints-To: newsabuse@supernews.com
Lines: 74


I can't use Direct Path export on Linux, either.  Try taking the 'direct =
y' out.  As I recall, there is a bug mentioned on Oracle's web page for
Linux.

Richard

<mshields0350@my-deja.com> wrote in message
news:92qg2d$82n$1@nnrp1.deja.com...
> In article <t512bt44q0fv06@corp.supernews.com>,
>   "Richard Sutherland" <rsutherland@In-Touch.net> wrote:
> > Actually, you need 65535 -- 65536 doesn't work
> > > Are both databases on the same OS/version?  What are they?
> > >
>
> I tried setting the recordlength with no change in results.  Here is
> the error message:
>
> >>>>imp sys/xxxxx file=io_data.dmp ignore=y full=y recordlength=65535
>
> Import: Release 8.1.6.1.0 - Production on Mon Jan 1 11:40:17 2001
>
> (c) Copyright 1999 Oracle Corporation.  All rights reserved.
>
> Connected to: Oracle8i Enterprise Edition Release 8.1.6.1.0 - Production
> With the Partitioning option
> JServer Release 8.1.6.0.0 - Production
>
> Export file created by EXPORT:V08.01.06 via direct path
> import done in US7ASCII character set and US7ASCII NCHAR character set
> IMP-00046: using FILESIZE value from export file of 2147483648
> . importing SYS's objects into SYS
> . importing DBO's objects into DBO
> . . importing table                    "ERROR_LOG"        108 rows
> imported
> . . importing table                       "IO_AGE"          2 rows
> imported
> . . importing table                  "IO_ANALYSIS"Segmentation fault
> (core dumped)
>
> The OS is Redhat Linux 6.2 - kernel 2.2.16-4.  The hardware and
> operating systems are identical.  The export was created with:
> TABLES=( about 40 tables )
> ROWS=y
> INDEXES=n
> CONSTRAINTS=n
> STATISTICS=none
> GRANTS=y
> COMPRESS=y
> DIRECT=y
>
> On the same database, I have already imported a 90 Million row
> partitioned table with no problems.
>
> If I do the import with INDEXFILE=x, the file 'x' is truncated at:
> REM  ... 2 rows
> REM  CREATE TABLE "DBO"."IO_ANALYSIS" ("ID_ANALYSIS" NUMBER(8, 0) NOT
> NULL
> REM  ENABLE, "ID_CLIENT" NUMBER(8, 0) NOT NULL ENABLE, "ID_UNIT_DEF"
> REM  NUMBER(8, 0) NOT NULL ENABLE, "ID_REC" NUMBER(8, 0) NOT NULL
> ENABLE,
>
> Which almost makes me think that the seg fault is happening during
> table creation, EVEN though the table gets created in the DB correctly
> before the seg fault.
>
> > > Richard Sutherland
> > > rvsutherland@yahoo.com
> > >
>
>
> Sent via Deja.com
> http://www.deja.com/