ERROR: clssnmDiskPMT: 1 of 1 voting disks unavailable (0/0/1)

Issue seems to be in the voting disk permission on restart the database.
you need to make changes permanent to avoid such error in future by making entries in /etc/rc.local file and tested by reboot the system.

Following are the steps for fixing the issue:

1. Check with crsctl command that cluster ready service is working.

#crsctl check crs

Failure 1 contacting CSS daemon
Cannot communicate with CRS
Cannot communicate with EVM

2. Checked the voting disk permission and raw devices permission.

# Check ownership and permissions of voting disks
ls -l /dev/asm/*   # Example path if ASM devices are used
ls -l /dev/raw/*   # Example path if raw devices are used

# Change ownership to grid user and oinstall group
chown grid:oinstall /dev/asm/*



# Example for raw device permissions
chown grid:oinstall /dev/raw/raw1
chmod 660 /dev/raw/raw1
# Check voting disk configuration
crsctl query css votedisk

# Check ASM disk discovery
asmcmd lsdsk

3. Modified the raw devices permission as below:

# Set read/write permissions for owner and group
chmod 660 /dev/asm/*

To make change persistent you need to make entries of each raw devices in /etc/rc.local file as follows:

# Example entries in /etc/rc.local
chown grid:oinstall /dev/raw/raw1
chmod 660 /dev/raw/raw1

chown grid:oinstall /dev/raw/raw2
chmod 660 /dev/raw/raw2

chown grid:oinstall /dev/raw/raw3
chmod 660 /dev/raw/raw3

4. Stop the crs service with following command:

# Run as root
crsctl stop crs

# Run as root
crsctl stop cluster -all

5. Start the crs Service with following command:

# Run as root
crsctl start crs

# Run as root
crsctl start cluster -all

6. Check the status again with crsctl command:

# Run as root
crsctl check crs

# check cluster resources
# Run as grid/oracle user
crsctl stat res -t

# Check cluster status for all nodes
# Run as root
crsctl check cluster -all