Other Oracle Background Processes (Beyond LGWR, DBWn, SMON, PMON, ARCn)

Besides the core processes you already met, an Oracle instance may start additional background processes depending on features enabled (Scheduler, AQ, RAC, Flashback, ASM, etc.). This guide groups the most common ones, clarifies legacy names, and gives quick checks you can run.

Modern process groups you may see

• Memory & Monitoring: MMAN (Memory Manager), MMON / MMNL (AWR snapshots, advisories), DIAG (diagnosability).
• Listener & Time: LREG (listener registration), VKTM (high-res time keeper).
• Scheduler: CJQ0 (job coordinator) and Jnnn (job slaves) for DBMS_SCHEDULER.
• Advanced Queuing: QMNC (queue monitor coordinator) and Qnnn workers.
• Flashback & Recovery: RVWR (Flashback Log writer), FBDA (Flashback Data Archive), CTWR (Block Change Tracking writer).
• Shared Server (MTS): Dnnn (dispatchers), Snnn (shared servers) when configured.
• RAC (cluster): LMON (Global Monitor), LMD0 (Lock Manager Daemon), LMSn (Global Cache Service); replace legacy LCKn.
• ASM (if +ASM present): RBAL, ARBn, ASMB for ASM disk group management.
• Distributed Txn: RECO (resolve in-doubt distributed transactions).

Legacy → modern name mapping (cleanups)

• LCKn (OPS era inter-instance locks) → superseded by LMON/LMD0/LMSn in RAC.
• SNPN/SNPO (job queue for DBMS_JOB) → use CJQ0 + Jnnn with DBMS_SCHEDULER.
• QMNO → QMNC (AQ coordinator) + Qnnn workers.
• Enterprise Manager (OEM) → prefer Enterprise Manager Cloud Control for monitoring/automation; not a DB background process but the recommended management plane.

Quick identification queries

-- All visible background processes
SELECT name, paddr, description
FROM   v$bgprocess
WHERE  paddr <> HEXTORAW('00');

-- Shared Server & Dispatchers (when MTS is enabled)
SELECT name, status FROM v$shared_server;
SELECT name, status, protocol, (busy/(busy+idle)) util
FROM   v$dispatcher;

-- Scheduler health (modern replacement for DBMS_JOB)
SELECT job_name, state, last_start_date, run_count
FROM   dba_scheduler_jobs
ORDER  BY state, job_name;

-- Advanced Queuing monitors
SELECT * FROM v$queue;      -- queues
SELECT * FROM v$aq;         -- AQ metrics (version dependent)

When you’ll actually see them

• CJQ0/Jnnn: any time DBMS_SCHEDULER jobs exist; CJQ0 spawns Jnnn as needed.
• Dnnn/Snnn: only if you configured Shared Server; otherwise you’ll see dedicated servers per session.
• LMON/LMD0/LMSn: only in RAC; absent in single-instance databases.
• RVWR/FBDA/CTWR: appear when Flashback Database, Flashback Data Archive, or Block Change Tracking are enabled.
• ASM family (ASMB/RBAL/ARBn): on hosts running a separate +ASM instance that your DB uses.

Modernizing older guidance in this lesson

• DBMS_JOB → DBMS_SCHEDULER: migrate legacy jobs for calendars, windows, resource control, and event-based scheduling.
• OEM → Cloud Control: use EM Cloud Control for fleet monitoring, jobs, RMAN, and compliance rather than per-host utilities.
• LCKn: treat as historical. RAC uses GES/GCS (LMON/LMD0/LMSn) for inter-instance locking and cache fusion.

Drop-in examples

-- Create a modern Scheduler job (replacement for DBMS_JOB)
BEGIN
  DBMS_SCHEDULER.create_job(
    job_name        => 'daily_gather_stats',
    job_type        => 'PLSQL_BLOCK',
    job_action      => 'BEGIN DBMS_STATS.gather_database_stats; END;',
    start_date      => SYSTIMESTAMP,
    repeat_interval => 'FREQ=DAILY;BYHOUR=2;BYMINUTE=0;BYSECOND=0',
    enabled         => TRUE,
    comments        => 'Nightly stats'
  );
END;
/

-- Shared Server quick check
SHOW PARAMETER shared_servers;

Usefulness & critique

• The original page listed a few legacy names (e.g., LCKO, SNPO, QMNO) and OEM steps; these are updated here with RAC/GES, Scheduler, and AQ coordinators used today.
• Added practical queries keyed to V$ views and Scheduler so DBAs can verify what’s running instead of relying on OS guesses.
• Separated “when you’ll see them” to prevent confusion from processes that only exist under specific features.

Other Processes - Quiz