Skip to main content

​
add-cache-headers.sh

Adds cache control headers to Azure Blob Storage files that don’t have them set.

​
Description

This script scans all blobs in the specified container and adds the cache control header Cache-Control: public, max-age=3600, must-revalidate to files that don’t have it or have a different value. It skips files that already have the correct header set.

​
Prerequisites

  • Azure CLI installed and configured
  • jq for JSON processing
  • Authenticated Azure session (az login)
  • Write access to Azure Storage Account

​
Configuration

Edit these variables in the script: 
ACCOUNT_NAME="testdinostr"
CONTAINER_NAME="docs"
CACHE_CONTROL="public, max-age=3600, must-revalidate"

​
Usage

​
Basic Syntax

./scripts/add-cache-headers.sh [PARALLEL]

​
Parameters

PARALLEL (optional, default: 10)
  • Number of parallel operations to run simultaneously

​
Examples

​
Default Usage (10 Parallel Operations)

./scripts/add-cache-headers.sh

​
High Parallelism

./scripts/add-cache-headers.sh 20

​
Conservative Approach

./scripts/add-cache-headers.sh 5

​
Output

Adding cache headers to blobs
Account: testdinostr
Container: docs
Cache-Control: public, max-age=3600, must-revalidate
Parallel operations: 10

Fetching storage account key...
Storage key retrieved successfully

Fetching blob list...
Found 156 blobs

Processing blobs...

✓ Updated: docs/ai-insights/emerging-failures.webp
⊘ Skipped (already set): docs/ai-insights/key-metrics.webp
✗ Failed: docs/some-file.webp
...

✓ Complete!

Summary:
  Total blobs: 156
  Updated: 89
  Skipped: 65
  Failed: 2

​
Features

  • Smart Detection: Only updates blobs that need it
  • Parallel Processing: Updates multiple blobs simultaneously
  • Progress Tracking: Real-time status with symbols:
    • ✓ Updated successfully
    • ⊘ Skipped (already has correct header)
    • ✗ Failed to update
  • Summary Report: Shows counts at the end

​
Cache Control Header

The script sets: 
Cache-Control: public, max-age=3600, must-revalidate
Meaning:
  • public - Can be cached by browsers and CDNs
  • max-age=3600 - Cache for 1 hour (3600 seconds)
  • must-revalidate - Must check with server after expiration

​
Customizing Cache Duration

Edit the script to change cache duration: 
# 1 hour (default)
CACHE_CONTROL="public, max-age=3600, must-revalidate"

# 1 day
CACHE_CONTROL="public, max-age=86400, must-revalidate"

# 1 week
CACHE_CONTROL="public, max-age=604800, must-revalidate"

# 1 year
CACHE_CONTROL="public, max-age=31536000, must-revalidate"

​
Performance

Blob CountRecommended Parallel
< 1005-10
100-50010-15
500-100015-20
> 100020-30

​
Execution Time

Approximate times (depends on network):
  • 100 blobs: ~30 seconds (10 parallel)
  • 500 blobs: ~2 minutes (15 parallel)
  • 1000 blobs: ~3-4 minutes (20 parallel)

​
Error Handling

​
Failed to Fetch Storage Key

Failed to fetch storage account key
Solution: Run az login and verify access

​
Failed Updates

Individual blob update failures are tracked and reported in the summary. Common causes:
  • Blob is locked
  • Insufficient permissions
  • Blob doesn’t exist (deleted during operation)

​
Missing jq

jq: command not found
Solution: 
# macOS
brew install jq

# Linux
apt-get install jq

​
Verification

​
Check a Specific Blob

az storage blob show \
  --account-name testdinostr \
  --container-name docs \
  --name "docs/ai-insights/key-metrics.webp" \
  --query "properties.contentSettings.cacheControl"

​
List All Blobs Without Cache Headers

az storage blob list \
  --account-name testdinostr \
  --container-name docs \
  --query "[?properties.contentSettings.cacheControl==null].name"

​
Use Cases

​
Initial Setup

Run once to add cache headers to all existing blobs: 
./scripts/add-cache-headers.sh

​
After Bulk Upload

After uploading new files without cache headers: 
./scripts/add-cache-headers.sh

​
Periodic Maintenance

Run periodically to ensure all blobs have proper caching: 
# Add to cron job
0 2 * * 0 /path/to/scripts/add-cache-headers.sh

​
Best Practices

  1. Test First: Run on a test container before production
  2. Monitor Progress: Watch for failed updates
  3. Verify Results: Check a few blobs manually after completion
  4. Schedule Wisely: Run during low-traffic periods
  5. Backup First: Consider backing up blob metadata before bulk updates

​
Troubleshooting

​
All Updates Failing

  • Verify write permissions on storage account
  • Check if container has immutability policy
  • Ensure blobs aren’t locked

​
Slow Performance

  • Reduce parallel count
  • Check network latency to Azure region
  • Run from Azure VM in same region for best speed

​
Inconsistent Results

  • Some blobs may be updated by other processes
  • Re-run the script to catch any missed blobs