# Security

> How dlogify redacts, isolates, retains and delivers the log data you send.

How dlogify handles your data.

## Redaction before storage

dlogify masks personal data such as email addresses, tokens and card numbers in messages and attributes before a sample, ticket, summary or stored log line is written.

## Retention by plan

Error samples, tickets, INFO summaries and the history of redacted log lines are deleted after 7, 30 or 90 days depending on the plan.

## Log history you control

Redacted log lines are stored on Cloudflare R2 in the United States so you can replay what led to a ticket. They reach you only through the API, and an admin can turn the history off per project.

## Tenant isolation

Every tenant-owned table is protected by PostgreSQL row-level security. Each service connects with its own role and can only reach the data it needs.

## API keys

Keys are shown once and stored as hashes. A key can only ingest; it cannot read data or manage the account.

## Signed webhooks

Every webhook delivery is signed with HMAC-SHA256 using a per-channel secret, which is stored encrypted and shown once.

## Webhook destinations

Webhook URLs must use https and resolve only to public addresses, so a channel cannot reach internal networks.

## Subprocessors

AI classification and ticket writing go through OpenRouter to the configured models. Email is delivered by Resend.

Source: https://dlogify.com/security
