hosting

GET /v1/sites/{siteId}/web-logs

Get the tail of a site's access and error logs.

All hosting endpoints

All developer docs →

Authentication

Send an API key as a bearer token. The key must carry the sites.view permission; a key without it is refused with 403, not 404.

This endpoint takes no organisation id. Your key already identifies the organisation it belongs to, and the response is scoped to it.

Try it

Replace anything in angle brackets with your own values, and the key placeholder with a key from your dashboard.

curl -X GET https://api.zinndigital.com/v1/sites/{siteId}/web-logs \
  -H "Authorization: Bearer zdk_live_…"

Signed in? The API console in your dashboard fills in your real organisation id and your own key, and runs the request against the live API so you can see the actual response. Open this endpoint in the API console

Details

The customer's own web-server logs for their own site. Until this endpoint existed a customer could not see their access log at all, which is why "why is my site slow" was a support ticket rather than a tab. ⛔ The path is /web-logs, deliberately not /logs. /v1/ops/logs is Loki and staff-only, and two endpoints differing by a prefix are two endpoints somebody confuses in a runbook at three in the morning. ⛔ access is a list of the customer's visitors' IP addresses — personal data held under the customer's lawful basis, not ours. It is served only to a principal entitled to the site, must not be cached by the client, and both streams are cut to their most recent bytes so a busy site cannot answer with tens of megabytes; truncated says so rather than implying the tail is the whole. unreadable is true when the vendor answered with a body naming neither log — which is not the same as a site with no traffic, and must never render as two empty panes. ⭐ Serves sites on our own servers as well as vendor-hosted ones (W43-54). A fleet site's access log is its own account's file; its error log is filtered for that account on the host, so no other tenant's lines are ever transferred. Each stream carries its own *_available, because on the fleet one log can be readable while the other is not. Requires sites.view and sites.panel_access. 404 for a site that does not exist or is not in scope.

Parameters

NameTypeRequiredWhat it is
siteId (path)UuidYesSite ID (UUIDv7).

Response

NameTypeRequiredWhat it is
accessstringYesThe tail of the access log, or "" when the vendor published none.
errorstringYesThe tail of the error log, or "" when the vendor published none. On a misconfigured application it can carry absolute paths and credentials in a stack trace.
truncatedbooleanYesEither stream was cut to its most recent bytes. Say so rather than implying the tail is the whole log.
unreadablebooleanYesNeither log could be read. ⛔ Not the same as a site with no traffic, and must never render as two empty panes.
access_availablebooleanYesWhether the access log could be read. On our own servers the two logs are different files and one can be readable while the other is not, so access: "" with this false means…
error_availablebooleanYesWhether the error log could be read — the same distinction, for error.

Errors this endpoint can return

401 · 403 · 404 · 429 · 503