How to Resolve Common Oracle RAC Installation Errors: A Step‑by‑Step Guide

This article discusses problems encountered during the entire RAC installation process and how they were solved.

Error 1: Shared disk configuration

When the shared disk contains data, after installing ClusterWare, running root.sh fails with "Failed to upgrade Oracle Cluster Registry configuration". The fix is to clear the OCR and voting disks using dd: dd if=/dev/zero of=/dev/rdsk/c2t0d2s3 bs=1073741824 count=1 The of parameter specifies the target device, and bs sets the block size. After clearing, rerun root.sh and the error disappears.

Error 2: s0 partition usage

Oracle does not use the s0 partition by default; specifying it for OCR or voting disk also triggers the same upgrade error. The solution is to use a different partition instead of s0.

Error 3: VIP cannot find public interface

Oracle treats private IP ranges (192.168.*, 10.*, 172.16‑31.*) as private, so it cannot automatically bind VIPs. The cluvfy tool reports: ERROR: Could not find a suitable set of interfaces for VIPs. After ClusterWare installation, the second node’s root.sh shows:

The given interface(s), "ce0" is not public. Public interfaces should be used to configure virtual IPs.

Fix: run VIPCA manually as root to configure a public interface, referencing Oracle Doc ID Note:316583.1 and Bug 4437727.

Error 4: Missing default gateway

If /etc/defaultrouter is not set, VIPCA fails during the “Starting VIP application resource” step, producing CRS‑1006 and CRS‑0215 errors. Adding a correct default route resolves the issue.

Error 5: Private ID binding

During ClusterWare installation, the private ID concept can cause “Could not find a suitable set of interfaces for VIPs.” The workaround is to avoid manual VIP binding. Instead of:

ifconfig eth0:1 plumb

ifconfig eth0:1 172.25.198.224 netmask 255.255.0.0 broadcast 172.25.255.255 up

remove the manual configuration and let Oracle bind the VIP automatically.

Error 6: racg_install makefile failure

Compilation of racg_install fails because env_has.mk is missing. The file exists under $ORACLE_HOME/crs/crs/lib but not under $ORACLE_HOME/crs/lib, causing a conflict between CRS home and ORACLE_HOME. Setting ORACLE_HOME to /data/oracle/product/10.2/database (while keeping ClusterWare under /data/oracle/product/10.2/crs) resolves the error.

Error 7: ASM disk reuse after re‑building

When the storage device is not reformatted, residual ASM metadata makes previously assigned disks unusable for a new ASM instance. Clearing the OCR and voting disks with the dd command (as in Error 1) and choosing a new ASM disk group name avoids ORA‑15032 and ORA‑15063 errors.

Error 8: Database fails to start after patch 5117016

Node 2 shows multiple errors such as ORA‑01078, ORA‑01565, ORA‑17503, ORA‑03113, and ORA‑07445, with a core dump. Oracle Note 390591.1 provides a detailed solution.

Overall, the article lists numerous pitfalls that can trip up newcomers to Oracle RAC installation, emphasizing the need for thorough preparation, systematic analysis, and patience.