HDISV Day 2 Operations Guide
Overview
Day 2 operations cover the ongoing management and maintenance of your HDISV instance after initial deployment. This guide focuses on tasks you can perform from z/OS and through the ZLPN portal.
Table of Contents
- Managing Your Instance
- z/OS Operations
- Backup and Recovery
- Monitoring and Maintenance
- Best Practices
Managing Your Instance
Starting and Stopping z/OS
Stop z/OS (MSI Mode)
From ZLPN Portal:
- Log into ZLPN portal
- Navigate to your instance
- Click "Stop" button
- Wait for status to show "Stopped"
From z/OS Console:
# Graceful shutdown command
/V XCF,LTE1,OFFLINE
Important: Always use graceful shutdown to prevent data corruption.
Start z/OS (MSI Mode)
From ZLPN Portal:
- Log into ZLPN portal
- Navigate to your instance
- Click "Start" button
- Wait 5-10 minutes for z/OS to complete startup
Verify z/OS is Ready:
# Test SSH connection
ssh zosadmn@<your-ip>
# Test 3270 connection
# Connect via your 3270 emulator to <your-ip>:2023
LAI Mode Operations
For LAI mode instances, you have direct access to the Linux layer and can manage the z/OS service manually. Contact support or refer to internal documentation for Linux layer operations.
Recreating an Instance (MSI Mode Only)
Recreating refreshes your z/OS system while preserving your user data.
When to Recreate
- Update z/OS - Get latest maintenance and fixes
- Change Middleware - Add or remove DB2, CICS, IMS, MQ
- Switch z/OS Version - Move to different z/OS release
- Recover from Issues - Fresh start while keeping data
What Gets Preserved
Preserved:
- USER01 volume (your datasets)
- USMS01 volume (SMS-managed datasets)
- Your IP address
- SSH keys
Refreshed:
- z/OS system files
- System configuration
- Middleware installations
Pre-Recreate Checklist
1. Backup Critical Data
# From TSO/ISPF
# Export important datasets using:
# - ISPF 3.4 (DSLIST) - Download via IND$FILE
# - FTP from your workstation
# - SFTP/SCP via SSH
2. Document Configuration
- Note any custom JCL procedures
- Save RACF user definitions (if any)
- Document middleware configurations
3. Verify Backup
- Test that you can access backed up files
- Ensure all critical data is saved
Recreate Process
1. Initiate Recreate
- Log into ZLPN portal
- Navigate to your instance
- Click "Recreate" button
- Select new options if desired:
- z/OS version
- Middleware selection
2. Wait for Completion
- Process typically takes 30-60 minutes
- You'll receive notification when complete
- Instance status will show "Active"
3. Verify After Recreate
# SSH to z/OS
ssh zosadmn@<your-ip>
# Check your datasets are present
# From TSO
LISTDS
# From ISPF 3.4
# Enter: ZOSADMN.**
# Verify your datasets are listed
4. Restore Configurations
- Reapply any custom JCL
- Reconfigure middleware if needed
- Test your applications
Post-Recreate Tasks
1. Set New Password
# SSH to z/OS
ssh zosadmn@<your-ip>
# Set your password
password_change.py -u ZOSADMN -p 'YourNewPassword123!'
2. Verify Middleware
# From TSO
# Check DB2 (if selected)
DSN
# Check CICS (if selected)
# Connect via 3270 to CICS region
# Check IMS (if selected)
# Verify IMS is active
3. Test Applications
- Run test jobs
- Verify dataset access
- Check application functionality
z/OS Operations
Dataset Management
Backing Up Datasets
Method 1: FTP from Workstation
# From your workstation
ftp <your-ip>
# Login as zosadmn
# Use your password
# Download dataset
get 'ZOSADMN.MY.DATASET' mydataset.txt
# Download multiple datasets
mget 'ZOSADMN.BACKUP.*'
Method 2: SFTP/SCP
# From your workstation
sftp zosadmn@<your-ip>
# Download dataset
get //'ZOSADMN.MY.DATASET' mydataset.txt
Method 3: DFSMSdss (from z/OS)
# From TSO
# Use ISPF 3.4 to export datasets
# Or submit DFSMSdss JCL
See Dataset Migration Guide for detailed instructions.
Restoring Datasets
From FTP:
# From your workstation
ftp <your-ip>
# Upload dataset
put mydataset.txt 'ZOSADMN.MY.DATASET'
From SFTP:
sftp zosadmn@<your-ip>
# Upload dataset
put mydataset.txt //'ZOSADMN.MY.DATASET'
Cleaning Up Old Datasets
From ISPF 3.4:
- Enter dataset pattern:
ZOSADMN.** - Type
Dnext to datasets to delete - Press Enter to confirm
From TSO:
# Delete single dataset
DELETE 'ZOSADMN.OLD.DATASET'
# Delete generation data group
DELETE 'ZOSADMN.MY.GDG' GDG
Job Management
Monitoring Jobs
From SDSF:
- Enter
SDSFfrom TSO - Type
STfor status display - View active jobs
- Type
?next to job for output
From USS (SSH):
# View job output
# Use SDSF or TSO commands
Canceling Jobs
From SDSF:
- Enter
SDSF - Type
STfor status - Type
Cnext to job to cancel
From TSO:
# Cancel job by name
CANCEL jobname
System Monitoring
Check System Status
From TSO:
# Display active tasks
D A,ALL
# Display system resources
D M=CPU
# Display storage
D ASM
Monitor Storage Usage
From ISPF 3.4:
- Enter:
ZOSADMN.** - Review space used by datasets
- Look for datasets using excessive space
From TSO:
# List catalog entries
LISTCAT ENTRIES('USER01') ALL
# Check volume space
LISTCAT ENTRIES('USMS01') ALL
See Managing Storage for detailed storage management.
Backup and Recovery
Backup Strategy
What to Backup
Critical Data:
- Application source code
- JCL procedures
- Configuration files
- Production datasets
- Test data
How Often:
- Daily: Active development datasets
- Weekly: All user datasets
- Before Recreate: Everything important
Backup Methods
1. Automated FTP Script
#!/bin/bash
# backup-datasets.sh
HOST="<your-ip>"
USER="zosadmn"
ftp -n $HOST <<EOF
user $USER
binary
cd //
mget 'ZOSADMN.PROD.*'
bye
EOF
2. Manual Export
- Use ISPF 3.4 to download datasets
- Use IND$FILE for file transfer
- Use SFTP for USS files
3. Dataset Migration
- See Dataset Migration Guide
- Use DFSMSdss for large datasets
Recovery Procedures
Restore from Backup
1. Upload Datasets
# From your workstation
ftp <your-ip>
# Login
put mydataset.txt 'ZOSADMN.MY.DATASET'
2. Verify Restoration
# From ISPF 3.4
# Check dataset exists and has correct data
3. Test Applications
- Run test jobs
- Verify data integrity
- Check application functionality
Disaster Recovery
If Instance is Lost:
- Create new instance through ZLPN portal
- Restore datasets from backups
- Reconfigure applications
- Test thoroughly
If Data is Corrupted:
- Stop affected applications
- Restore from last good backup
- Verify data integrity
- Resume operations
Monitoring and Maintenance
Regular Maintenance Tasks
Daily Tasks
1. Check System Status
- Verify z/OS is running
- Check for error messages
- Monitor job completion
2. Review Storage
- Check space usage
- Clean up old datasets
- Monitor growth trends
3. Backup Active Data
- Export changed datasets
- Save new source code
- Update backup inventory
Weekly Tasks
1. Full Backup
- Backup all user datasets
- Export JCL procedures
- Save configuration files
2. Review Performance
- Check job run times
- Monitor resource usage
- Identify bottlenecks
3. Clean Up
- Delete old job output
- Remove temporary datasets
- Archive old data
Monthly Tasks
1. Review Instance Size
- Assess if current size is adequate
- Consider upgrade if needed
- Plan for growth
2. Update Documentation
- Document configuration changes
- Update procedures
- Review backup strategy
3. Test Recovery
- Practice restore procedures
- Verify backups are valid
- Update recovery plans
Performance Monitoring
Check Resource Usage
From z/OS:
# Display CPU usage
D M=CPU
# Display storage usage
D ASM
# Display I/O activity
D U,DASD
Optimize Performance
1. Review Job Priorities
- Adjust job classes
- Optimize JCL
- Balance workload
2. Manage Storage
- Compress datasets
- Delete unnecessary files
- Request additional storage if needed
3. Consider Upgrade
- If consistently slow, consider larger instance
- Contact support to discuss options