ASM Command-Line Utility (ASMCMD) – Complete Guide for Oracle DBAs

Introduction

Oracle ASM Command-Line Utility (ASMCMD) is a powerful and essential tool for Oracle DBAs and Grid Infrastructure administrators working with Automatic Storage Management (ASM).

ASMCMD provides a UNIX-like interface to manage ASM files and directories, making daily storage operations easier, faster, and more intuitive—especially in production environments where precision and safety matter.

What is ASMCMD?

ASMCMD is a command-line utility used to view, navigate, and manage files and directories inside ASM disk groups.

With ASMCMD, you can:

• Browse ASM disk groups
• Check space usage
• Monitor ASM clients
• View I/O statistics
• Manage aliases and directories

Before Oracle 10g Release 2, DBAs had to manage ASM mostly using OEM or SQL queries connected to the ASM instance. ASMCMD simplified this by offering filesystem-style commands similar to Linux.

Oracle ASM Background Processes – Explained for DBAs

Oracle Automatic Storage Management (ASM) relies on multiple background processes to manage disk groups, metadata, rebalancing, synchronization, and communication with Oracle Clusterware. Understanding these processes is crucial when troubleshooting ASM alerts, rebalance operations, or storage-related performance issues.

ARBx – ASM Rebalance Slaves

The ARBx processes (where x is a numeric identifier) are slave processes responsible for executing disk rebalance operations. These processes redistribute extents across ASM disks whenever disks are added, dropped, or rebalanced.

TIP: A high number of active ARBx processes usually indicates an ongoing rebalance. Always monitor rebalance impact during business hours.

ASMB – ASM Bridge

ASMB acts as the communication bridge between ASM and Cluster Synchronization Services (CSS).
It provides:

• Disk and diskgroup information to CSS
• Heartbeat functionality
• ASM statistics updates

This process is critical for cluster-wide ASM coordination.

CKPT – Checkpoint Process

In ASM (especially in RAC), the CKPT process manages cross-instance coordination and metadata checkpoints, ensuring consistency across nodes.

DBWR – Database Writer (ASM)

The DBWR process in ASM writes dirty ASM metadata buffers from the ASM buffer cache to disk.
It does not write user data files, only ASM metadata structures.

GMON – Group Monitor

GMON handles disk group–level operations, including:

• Disk offline and drop activities
• Disk group compatibility updates
• Disk state transitions

This process plays a major role during disk failures and planned maintenance.

KATE – ASM Temporary Errands (Konductor)

KATE is responsible for onlining ASM disks.
It is started only when an offline disk is brought back online and runs within the ASM instance.

TIP: If disk onlining hangs, check KATE activity and ASM alert logs.

LGWR – ASM Log Writer

The ASM LGWR process maintains the ASM Active Change Directory (ACD) and flushes metadata change records to disk.
It ensures durability and recoverability of ASM metadata.

MARK – Mark Allocation Unit for Resync

The MARK process coordinates staleness tracking and resynchronization metadata when disks go offline in redundancy-enabled disk groups (NORMAL/HIGH).

This process runs in the RDBMS instance, not the ASM instance.

PING – Network Latency Monitor

The PING process measures network latency between cluster nodes.
Its functionality is similar to PING in RDBMS instances and supports cluster health monitoring.

PMON – Process Monitor

PMON manages:

• ASM background process cleanup
• Detection and recovery of failed processes

It ensures ASM instance stability.

PSP0 – Process Spawner

PSP0 is responsible for spawning and managing Oracle background processes within the ASM instance.

PZ9x – Parallel Query Slaves

The PZ9x processes are parallel execution slaves used when querying GV$ views to fetch instance-wide information efficiently.

RBAL – Rebalance Coordinator

RBAL:

• Discovers all ASM disks
• Opens device files
• Coordinates rebalance activities
• Spawns ARBx slave processes

RBAL is always active during rebalance operations.

SMON – System Monitor

In ASM, SMON also acts as a liaison with Cluster Synchronization Services (CSS).
It helps with:

• Node membership tracking
• Instance recovery
• Disk group consistency checks

VKTM – Virtual Keeper of Time

VKTM maintains the high-resolution system timer, ensuring precise time-based operations in ASM, similar to its function in RDBMS instances.

ASMCMD Modes of Operation

1. Interactive Mode

• Shell-like environment
• Best for live administration and troubleshooting
• Similar to SQL*Plus interactive usage

2. Non-Interactive Mode

• Single command execution
• Ideal for scripts and automation

Common ASMCMD Commands (File & Directory Management)

Disk Group Management Commands

Running ASMCMD (Interactive Mode)

List Disk Groups

Checking ASM Directory & Files

ASM Disks Header and Mount Status

There are two important columns under V$ASM_DISK which are MOUNT_STATUS and HEADER_STATUS. These two columns are required when you add / remove ASM disks.

ASM Disk MOUNT_STATUS

• MISSING – Disk is known to be part of the ASM disk group, but no disk in the storage
• CLOSED – Disk is present in the storage system but is not being accessed by ASM
• OPENED – Disk is present in the storage system and is being accessed by ASM
• CACHED – Disk is present in the storage system, and is part of a disk group being accessed by the ASM instance. This is the normal state for disks in an ASM.

ASM Disk HEADER_STATUS

• UNKNOWN – ASM disk header has not been read
• CANDIDATE – Can be used
• INCOMPATIBLE – Version number in the disk header is not compatible with the ASM version
• PROVISIONED – Disk is not part of a disk group and may be added to a disk group
• MEMBER – Already member of a diskgroup
• FORMER – Once used, can be re-used
• CONFLICT – ASM disk was not mounted due to a conflict

Disk Group Information

ASM Clients Connected

Disk Usage in ASM

Discover Mounted & Dismounted Disk Groups

ASM Disk Details

ASMCMD Version

ASM I/O Statistics (Highly Useful in Production)

ASM SPFILE Location

Non-Interactive ASMCMD Usage

Adding Disks to a Disk Group

ALTER DISKGROUP DATA_PROD ADD DISK ‘/devices/DATA1′,’/device/DATA2’;

Dropping Disks and Disk Groups

ALTER DISKGROUP DATA_PROD DROP DISK DATA1;

DROP DISKGROUP DATA_PROD INCLUDING CONTENTS;

Note:

• DROP DISKGROUP statements requires the instance to be in MOUNT state.

Rebalance Disk Group

• ASM rebalance a disk group automatically, whenever we add or remove disks form disk group.
• Disk groups can be rebalanced manually using the REBALANCE clause of the ALTER DISKGROUP statement.
• If the POWER clause is omitted the ASM_POWER_LIMIT parameter value is used.