Cloud Storage

Overview

Dialer.CX can automatically archive call recordings to any S3-compatible cloud storage provider. This offloads storage from your local server, reduces costs through compression, and provides durable long-term archival for compliance requirements.

How It Works

The archival process runs automatically as a background job:

1. Calls are recorded locally in high-quality WAV format
2. A background job runs every 5 minutes (managed by the scheduled jobs system)
3. Recordings are converted from WAV to compressed MP3, saving approximately 90% of storage
4. MP3 files are uploaded to your S3 bucket with the original filename preserved
5. Upload is verified with a checksum comparison to ensure file integrity
6. Local files are purged according to your retention policy after successful upload

Note

The conversion and upload process is non-blocking. Recordings remain available for playback throughout the process. Once uploaded, the system seamlessly serves recordings from either local storage or S3 depending on where the file currently resides.

Setup

Navigate to Admin > Settings > Recording Storage to configure cloud archival:

1. Enter your S3 credentials (Access Key ID and Secret Access Key)
2. Select or enter your bucket name (the system can auto-discover existing buckets)
3. Choose the region where your bucket is located
4. Click Test Connection to verify the credentials have proper access
5. Set the local retention policy — how many days to keep files locally after successful upload
6. Click Save to activate the archival process

Compatible Providers

Any S3-compatible storage provider works, including major cloud platforms, self-hosted solutions, and budget storage providers. If your provider offers an S3-compatible API endpoint, it will work with Dialer.CX.

For providers other than the default, enter the custom endpoint URL in the configuration form.

Credentials

• Access Key ID and Secret Key are stored encrypted in the database
• Credentials are only decrypted in memory when needed for upload operations
• Use a dedicated service account with minimal permissions (put, get, list on a single bucket)

Tip

Create a dedicated service account with only the permissions needed, scoped to a single bucket. Never use root or broad-access keys for this integration.

Monitoring

Check upload status at Admin > Scheduled Jobs:

• Last run time and result for the recording upload job
• Number of recordings uploaded per batch (typically processes all pending files every 5 minutes)
• Error details for any failed uploads, including the specific file and error message
• Queue depth showing how many recordings are waiting to be uploaded

Storage Savings

The WAV-to-MP3 conversion provides significant cost savings:

Format	Size (1-hour call)	Relative Cost
WAV (original)	~115 MB	Baseline
MP3 (archived)	~14 MB	~88% less

That is an 88% reduction in storage costs, and the savings scale linearly with call volume.

Caution

Do not manually delete files from the S3 bucket. The system tracks which recordings have been uploaded and where they are stored. Manual deletion will cause playback failures for those recordings. Use the platform’s retention settings to manage lifecycle.

Troubleshooting

Issue	Cause	Solution
Test connection fails	Invalid credentials or bucket	Verify Access Key, Secret Key, bucket name, and region
Uploads stuck at 0	Background job not running	Check Admin > Scheduled Jobs for the upload job status
Permission denied on upload	Insufficient permissions	Add write permission to the storage service account
Slow uploads	Large backlog or limited bandwidth	Uploads process in batches; the queue will clear over time