Re: get 3113/7445 when analyzing the redo log on solaris/oracle8172

I see some Oracle 9i stuff in your email, but you say that you use Oracle 8.1.7. Are you mixing different versions of Oracle products?

Best regards,

Ognjen Antonic - Ogo

"chao_ping" <member_at_dbforums.com> wrote in message news:1778310.1031235051_at_dbforums.com...
>
> I am trying to analyze the redo log in oracle 8172 64bit on
> solaris.Whatever i tried , it always core dump after fetching about 60k
> records from the v$logmnr_contents.
> I tried to analyze the logfile on 8172(64bit), 901 64bit on solairs, but
> in vain.
> Here is the example output:
>
> 04-SEP-02 update "EACHPAY"."UL_USERLOGIN
> 04-SEP-02 update "ACCOUNT"."USERSTATISIN
> 04-SEP-02 update "BIDDER"."SHOPCATALOG"
> ERROR:
> ORA-03113: end-of-file on communication channel
> 66600 rows selected.
>
> and in the background trace file, there is alert like:
> Errors in file
> /export/home/oracle9i/admin/test9i/udump/test9i_ora_4534.trc:
> ORA-07445: exception encountered: core dump [FFFFFFFF7F790794]
> [SIGSEGV] [Address not mapped to object] [0x0000000D8] [] []
> Count(*)from v$logmnr_contents also give out such error.
> I tried to enlarge the shared_pool, shared_pool_reserved to
> SQL> show parameter shared_pool
>
> NAME TYPE VALUE
> ------------------------------------ -----------
> ------------------------------
> shared_pool_reserved_size big integer 20000000
> shared_pool_size big integer 234881024
> Still get that error.
> Is there someone also ever hit similiar error ?
>
> --
> Posted via http://dbforums.com
>
Received on Mon Sep 23 2002 - 23:34:19 CEST