• Article Type: General
• Product: Aleph
• Product Version: 18.01

Description:
These errors appeared in the ue_01 log:

Jul 31 12:42 abc01:
run_e_08.18849:Oracle error: update_cursor3 z01
run_e_01_word.18849:Oracle error: io_z07a_count and
run_e_01_z0102.18849:Oracle error: io_z07a_count

abc50:
run_e_11.18849: Oracle error: update_cursor z105

These errors are followed by
ORA-03135: connection lost contact and
ORA-03114: not connected to ORACLE.

Jul 14th: run_e_01.14420:Oracle error: update_cursor1 z07

On the Test server, run_e_01.11216 from Aug.12, 2009:
Oracle error: update_cursor1 z07
ORA-03135: connection lost contact
Oracle error: update_cursor1 z07
ORA-03114: not connected to ORACLE

Resolution:
The fact that these Oracle update cursor and io errors are found in connection with:

ORA-03135: connection lost contact and
ORA-03114: not connected to ORACLE

indicates that either Oracle came down or this is a transient event having to do with the the network connection between the two servers in the TWO_TASK environment rather than a problem with ALEPH or Oracle itself.

This is not Aleph- related, but rather "hardware/local communication"-related.

Assuming that Oracle did not actually come down at this time, your staff should check to see if the times of these messages corresponded to any particular network problems.

Note: You can do util a/17/14 for a particular table to confirm that its Oracle indexes are OK.

"update_cursor3 z01" refers to z01_id3
"update_cursor z105" refers to z105_id
"update_cursor1 z07" refers to z07_id1

• Article last edited: 10/8/2013