Client sqlplus SELECT + SPOOL knocks down server network

From: Carlos <miotromailcarlos_at_netscape.net>
Date: Mon, 10 Mar 2008 04:51:00 -0700 (PDT)
Message-ID: <54108cce-5755-4f15-9fbf-c809245e63c1@d62g2000hsf.googlegroups.com>

Hello all:

I'm facing a strange problem on a 10.2.0.3 Oracle on Oracle Enterprise Linux 5 server.

This morning someone was issuing a SELECT from a windows client via ODBC and suddenly all the server communications broke down: even no ping from or to that server worked.

Tried stop&start network but no avail. The server appears to be dead at communications until reboot.

So I tried to repeat the SELECT from sqlplus on a windows client to a OS file with SPOOL and all the communications on the server are gone once again. Ping only answers to 127.0.0.1 or its own IP address, and Linux is telling me that the network card is OK. Stop&start network didn't fix the problem either.

Client sqlnet.log shows:

Fatal NI connect error 12170.

  VERSION INFORMATION:

	TNS for 32-bit Windows: Version 10.2.0.1.0 - Production
	Windows NT TCP/IP NT Protocol Adapter for 32-bit Windows: Version

10.2.0.1.0 - Production
  Time: 10-MAR-2008 12:11:59
  Tracing not turned on.
  Tns error struct:
    ns main err code: 12535
    TNS-12535: Message 12535 not found; No message file for product=NETWORK, facility=TNS

    ns secondary err code: 12560
    nt main err code: 505
    TNS-00505: Message 505 not found; No message file for product=NETWORK, facility=TNS

    nt secondary err code: 60
    nt OS err code: 0
  Client address: <unknown>

The select is a two column (varchar2(9), char(1)) select of 11 million rows aprox.

Could anybody tell me where to search next? (I didn't find anything at metalink).

TIA. Cheers.

Carlos. Received on Mon Mar 10 2008 - 06:51:00 CDT