var sidebar_align = 'right';
var content_container_margin = parseInt('290px');
var sidebar_width = parseInt('270px');
Unanswered: New Install on Solaris *
I am attempting to install Oracle 8.1.7 on a Soalris 8 system (Ultra-5, 512 MB RAM, 20 GB disk).
I made sure that Solaris was patched up as required for Oracle.
The install goes fine. However, it is when I attempt to start the Oracle Sever that I have a problem.
I run svrmgrl to start Oracle, and this is what happens:
Oracle Server Manager Release 184.108.40.206.0 - Production
Copyright (c) 1997, 1999, Oracle Corporation. All Rights Reserved.
Oracle8i Enterprise Edition Release 220.127.116.11.0 - Production
With the Partitioning option
JServer Release 18.104.22.168.0 - Production
SVRMGR> connect internal
ORA-03113: end-of-file on communication channel
After that, I quit svrmgrl and check for the presence of oracle processes:
Server Manager complete.
bash-2.03$ ps -ef | grep oracle
oracle 9856 1 0 14:09:30 ? 0:00 /usr/openwin/bin/xterm -display 22.214.171.124:0.0
oracle 9895 9858 0 15:01:55 pts/2 0:00 grep oracle
oracle 8016 1 0 Feb 12 ? 0:00 /opt/oracle/product/8.1.7/bin/tnslsnr LISTENER -inherit
oracle 9858 9856 0 14:09:30 pts/2 0:00 bash
As you can see, only the listener is running, and I cannot connect to Oracle with sqlplus.
I am at my wits end. I have learned through some of the forums that the ORA-01331 error is a shared memory problem of some kind, but I am not sure how it relates to my problem, and what I would have to do to fix it.
Any help in this area would be awesome.
Although I come from WinNT/2000 environment but still if it helps.. Just check out whether the environmental variables are in shape in ur .profile file.
Secondly issue the command Shutdown abort then try to issue startup command.
Hope it helps
There is no evidence to suggest this is a shared memory problem.
The ORA-3113 is a generic error - it means communication has been lost with the server process - we don't know why.
Check the alert.log, USER_DUMP_DEST for any trace files, and CORE_DUMP_DEST for core files.
If there are no other errors in the alert log and no trace files/core dumps (I'm sure there must be) then truss the sqlplus session.
If you don't know how to use truss, let me know.
My 1st suspicion is that /etc/system does not have the correct values.
Please post the contents of your /etc/system file.