Files
coder/docs/admin/setup/data-retention.md
T
Nick Vigilante c84aa564ba docs: normalize code-fence languages for Shiki compatibility (#27161)
Normalizes non-standard code-fence language tags across `docs/**` so a
strict highlighter (Shiki, used by Fumadocs) won't fail the build on an
unrecognized language, and unifies redundant synonym tags onto one
canonical form per language. The current renderer (Speed-Highlight)
detects the language from the code content, not the fence label, so this
drift wasn't visible until now.

## Changes

- `hcl` -> `tf` (199 fences, including indented ones nested in
numbered/bulleted lists). Shiki ships `hcl` and `terraform` as two
distinct grammars (not aliases); every `hcl`-tagged fence in `docs/**`
is actually Terraform resource/data/provider syntax, so the more
specific `terraform` grammar is correct for all of them. `tf` is Shiki's
own alias for that grammar, and it's also what GitHub's own markdown
renderer resolves to the same HCL/Terraform highlighting.
- `pwsh`/`powershell` -> `ps1`. Both `ps` and `ps1` are registered
PowerShell aliases in Shiki, but on GitHub's renderer only `.ps1` is a
registered file extension (`.ps` isn't), so `ps1` renders identically to
`powershell` there today while bare `ps` would silently lose
highlighting.
- `env` -> `dotenv` (a dedicated Shiki grammar for `KEY=VALUE` files)
- `text`/`output`/`none`/`url` -> `txt`. Same built-in plain-text
fallback either way, just shorter.
- `Dockerfile` -> `dockerfile` (lowercase)
- `bash`/`shell` -> `sh` (732 fences). Shiki and GitHub both alias all
three to a single shell grammar; this was already the style guide's
stated preference, just not enforced across the existing corpus until
now.
- `markdown` -> `md` (4 fences). Alias of the same grammar in both Shiki
and GitHub.
- `jsonc` -> `json` (1 fence). The block has no comments or trailing
commas, so it doesn't need the comments-capable grammar.
- `ts` -> `tsx` (2 fences, `docs/about/contributing/frontend.md`).
Verified the actual content tokenizes identically under both grammars,
and a sibling block in the same file already needs `tsx` for real JSX,
so unifying to one tag is safe for this file. Documented a caveat: `tsx`
mis-tokenizes the legacy angle-bracket type-assertion syntax
(`<Type>value`), which is invalid in real `.tsx` files anyway, so use
`value as Type` instead.
- `yml` -> `yaml` (1 fence)
- Updated `docs/.style/style-guide/formatting.md` to document all
canonical tags

`promql` (2 fences) and `caddyfile` (2 fences) are left as-is. Shiki
doesn't bundle a grammar for either, so they need a custom grammar
registration when the site adopts Shiki, rather than degrading to `txt`.
Tracked as follow-up work under DOCS-118 and
[DOCS-544](https://linear.app/codercom/issue/DOCS-544/vendor-a-local-promql-grammar-for-shiki-syntax-highlighting)
(promql).

Does not touch `offlinedocs/`.

Linear:
[DOCS-476](https://linear.app/codercom/issue/DOCS-476/normalize-docs-code-fence-languages-de-risk-shikifumadocs)

<details>
<summary>How the fence tags were verified</summary>

Each tag was tested against a real `shiki@latest` highlighter instance
(`codeToHtml`/`codeToTokens`) and cross-checked against GitHub's
`@wooorm/starry-night` grammar sources (the renderer that actually
displays these `.md` files today, in repo browsing and PR diffs), since
that's what determines whether brevity is safe before Shiki adoption:

```text
FAIL  env        -- Language `env` is not included in this bundle.
FAIL  Dockerfile -- Language `Dockerfile` is not included in this bundle.
FAIL  promql     -- Language `promql` is not included in this bundle.
FAIL  caddyfile  -- Language `caddyfile` is not included in this bundle.
FAIL  pwsh       -- Language `pwsh` is not included in this bundle.
FAIL  output     -- Language `output` is not included in this bundle.
```

`hcl` doesn't error in Shiki, since it's a real grammar, but that's
exactly the trap: it was silently rendering every fence with the generic
HCL grammar instead of the Terraform-specific one. Every `hcl`-tagged
fence in `docs/**` was manually checked against `origin/main` and is
genuinely Terraform content.

For `ts`/`tsx`, tokenizing the actual doc content confirmed identical
output under both grammars; a synthetic test with the legacy
angle-bracket cast syntax confirmed `tsx` degrades on that specific
construct, which the style guide now calls out.

The first normalization pass only matched fence tags at column 0
(`^```tag$`), missing tags indented inside numbered/bulleted lists. A
follow-up pass caught the remaining occurrences at any indentation
level.

</details>


---

*This PR description and the underlying changes were prepared with Coder
Agents assistance.*
2026-07-15 14:07:09 -04:00

7.8 KiB

Data Retention

Coder supports configurable retention policies that automatically purge old Audit Logs, Connection Logs, Workspace Agent Logs, API keys, and AI Gateway records. These policies help manage database growth by removing records older than a specified duration.

Overview

Large deployments can accumulate significant amounts of data over time. Retention policies help you:

  • Reduce database size: Automatically remove old records to free disk space.
  • Improve performance: Smaller tables mean faster queries and backups.
  • Meet compliance requirements: Configure retention periods that align with your organization's data retention policies.

Note

Retention policies are disabled by default (set to 0) to preserve existing behavior. The exceptions are API keys and workspace agent logs, which default to 7 days.

Configuration

You can configure retention policies using CLI flags, environment variables, or a YAML configuration file.

Settings

Setting CLI Flag Environment Variable Default Description
Audit Logs --audit-logs-retention CODER_AUDIT_LOGS_RETENTION 0 (disabled) How long to retain Audit Log entries
Connection Logs --connection-logs-retention CODER_CONNECTION_LOGS_RETENTION 0 (disabled) How long to retain Connection Logs
API Keys --api-keys-retention CODER_API_KEYS_RETENTION 7d How long to retain expired API keys
Workspace Agent Logs --workspace-agent-logs-retention CODER_WORKSPACE_AGENT_LOGS_RETENTION 7d How long to retain workspace agent logs
AI Gateway --ai-gateway-retention CODER_AI_GATEWAY_RETENTION 60d How long to retain AI Gateway records

Note

AI Gateway retention is configured separately from other retention settings. See AI Gateway Setup for detailed configuration options.

Duration Format

Retention durations support days (d) and weeks (w) in addition to standard Go duration units (h, m, s):

  • 7d - 7 days
  • 2w - 2 weeks
  • 30d - 30 days
  • 90d - 90 days
  • 365d - 1 year

CLI Example

coder server \
  --audit-logs-retention=365d \
  --connection-logs-retention=90d \
  --api-keys-retention=7d \
  --workspace-agent-logs-retention=7d \
  --ai-gateway-retention=60d

Environment Variables Example

export CODER_AUDIT_LOGS_RETENTION=365d
export CODER_CONNECTION_LOGS_RETENTION=90d
export CODER_API_KEYS_RETENTION=7d
export CODER_WORKSPACE_AGENT_LOGS_RETENTION=7d
export CODER_AI_GATEWAY_RETENTION=60d

YAML Configuration Example

retention:
  audit_logs: 365d
  connection_logs: 90d
  api_keys: 7d
  workspace_agent_logs: 7d

ai_gateway:
  retention: 60d

How Retention Works

Background Purge Process

Coder runs a background process that periodically deletes old records. The purge process:

  1. Runs approximately every 10 minutes.
  2. Processes records in batches to avoid database lock contention.
  3. Deletes records older than the configured retention period.
  4. Logs the number of deleted records for monitoring.

Effective Retention

Each retention setting controls its data type independently:

  • When set to a non-zero duration, records older than that duration are deleted.
  • When set to 0, retention is disabled and data is kept indefinitely.

API Keys Special Behavior

API key retention only affects expired keys. A key is deleted only when:

  1. The key has expired (past its expires_at timestamp).
  2. The key has been expired for longer than the retention period.

Setting --api-keys-retention=7d deletes keys that expired more than 7 days ago. Active keys are never deleted by the retention policy.

Keeping expired keys for a short period allows Coder to return a more helpful error message when users attempt to use an expired key.

Workspace Agent Logs Behavior

Workspace agent logs are deleted based on when the agent last connected, not the age of the logs themselves. Logs from the latest build of each workspace are always retained regardless of when the agent last connected. This ensures you can always debug issues with active workspaces.

For non-latest builds, logs are deleted if the agent hasn't connected within the retention period. Setting --workspace-agent-logs-retention=7d deletes logs for agents that haven't connected in 7 days (excluding those from the latest build).

AI Gateway Data Behavior

AI Gateway retention applies to interception records and all related data, including token usage, prompts, and tool invocations. The default of 60 days provides a reasonable balance between storage costs and the ability to analyze usage patterns.

For details on what data is retained, see the AI Gateway Data Retention documentation.

Best Practices

For most deployments, we recommend:

retention:
  audit_logs: 365d
  connection_logs: 90d
  api_keys: 7d
  workspace_agent_logs: 7d

ai_gateway:
  retention: 60d

Compliance Considerations

Warning

Audit Logs provide critical security and compliance information. Purging Audit Logs may impact your organization's ability to investigate security incidents or meet compliance requirements. Consult your security and compliance teams before configuring Audit Log retention.

Common compliance frameworks have varying retention requirements:

  • SOC 2: Typically requires 1 year of audit logs.
  • HIPAA: Requires 6 years for certain records.
  • PCI DSS: Requires 1 year of audit logs, with 3 months immediately available.
  • GDPR: Requires data minimization but does not specify maximum retention.

External Log Aggregation

If you use an external log aggregation system (Splunk, Datadog, etc.), you can configure shorter retention periods in Coder since logs are preserved externally. See Capturing/Exporting Audit Logs for details on exporting logs.

Database Maintenance

After enabling retention policies, you may want to run a VACUUM operation on your PostgreSQL database to reclaim disk space. See Maintenance Procedures for guidance.

Keeping Data Indefinitely

To keep data indefinitely for any data type, set its retention value to 0:

retention:
  audit_logs: 0s           # Keep audit logs forever
  connection_logs: 0s      # Keep connection logs forever
  api_keys: 0s             # Keep expired API keys forever
  workspace_agent_logs: 0s # Keep workspace agent logs forever

ai_gateway:
  retention: 0s            # Keep AI Gateway records forever

Monitoring

The purge process logs deletion counts at the DEBUG level. To monitor retention activity, enable debug logging or search your logs for entries containing the table name (e.g., audit_logs, connection_logs, api_keys).