Run pre/post checks with the Cluster Verification Utility (cluvfy)
OracleRAC & GridCommand
# Full check after a Grid installation/change:
cluvfy stage -post crsinst -n all -verbose
# Before adding a node:
cluvfy stage -pre nodeadd -n <new_node> -verbose
# Component checks:
cluvfy comp nodecon -n all -verbose
cluvfy comp clocksync -n all -verbose
cluvfy comp ocr -n all -verbose
cluvfy comp asm -n all -verbose
# Report only failed checks:
cluvfy stage -post crsinst -n all -verbose | grep -iE "failed|warning"What it does
Validates network connectivity, clock synchronization, OCR, ASM, user equivalence and OS prerequisites on all nodes; reports problems with suggested fixes.
When to use it
Before and after Grid installation, node addition and upgrades; as a baseline validation for unexplained cluster problems.
Caution
Read-only, but it connects to all nodes over ssh; the grid user needs ssh equivalence. Do not run the -fixup scripts (root) without reviewing them; they change OS parameters.
Related
- Check RAC database, instance and service status with srvctl
- Stop / start a RAC database or a single instance with srvctl (rolling)
- Add a RAC service with preferred/available instances (including Application Continuity)
- Relocate a RAC service to another instance
- Check cluster resource status (crsctl stat res -t)
- Check cluster stack health, the node list and the Clusterware version