Identification section "no process tcp EP" part of "Oracle DB on Unix" pattern creating/updating only one Oracle Instance CI despite multiple Oracle SIDs are running - Known Error

Identification section "no process tcp EP" part of "Oracle DB on Unix" pattern creating/updating only one Oracle Instance CI despite multiple Oracle SIDs are running Description During Oracle DB Instance discovery, when a ora_pmon_SID process is detected on UNIX like OS, discovery pattern " Oracle DB on Unix " is triggered. Due to permission issues during Process Detection phase while execution of the following commands:

ls -l /proc/PID/exe ls -l /proc/PID/cwd cat /proc/PID/environ | tr "\000" "\n" Command result: Sorry, user USER > is not allowed to execute COMMAND > as root on HOST >

The first identification section " Identification for Oracle DB entry point type(s) - no instance " is not triggered, causing fallback to the identification section " no process tcp EP ", where the issue is observed. When multiple Oracle Instances are running on a single host, not all of them are getting discovered. This is happening due to the logic in extension section, where the shell command is taking all the LOCAL oracle processes, running on this particular host, but always assigning the first SID (Service ID).

Steps to Reproduce

1. Trigger IP based discovery against UNIX like OS host where ora_pmon process is running.

2. Ensure discovery pattern " Oracle DB On Unix " is triggered.

3. Ensure identification section " no process tcp EP " is triggered.

4. Observe only 1 Oracle DB Instance is created/updated in CI class Oracle Instance [cmdb_ci_db_ora_instance], despite multiple ora_pmon processes are running on the host.

Workaround The problematic behaviour is already fixed in the context of PRB1937379 with release target December 2025 as part of plugin " Visibility Content " [ sn_pattern_design ] version 6.29.0

Changes upon discovery pattern " Oracle DB on Unix ", identification section " no process tcp EP "