Sentry is a developer-first error tracking and performance monitoring platform. Sentrys Slack integration incorrectly records the incoming request body in logs. This request data can contain sensitive information, including the deprecated Slack verification token. With this verification token, it is possible under specific configurations, an attacker can forge requests and act as the Slack integration. The request body is leaked in log entries matching event == slack.* && name == sentry.integrations.slack && request_data == *
. The deprecated slack verification token, will be found in the request_data.token
key. SaaS users do not need to take any action. Self-hosted users should upgrade to version 24.5.0 or higher, rotate their Slack verification token, and use the Slack Signing Secret instead of the verification token. For users only using the slack.signing-secret
in their self-hosted configuration, the legacy verification token is not used to verify the webhook payload. It is ignored. Users unable to upgrade should either set the slack.signing-secret
instead of slack.verification-token
. The signing secret is Slacks recommended way of authenticating webhooks. By having slack.singing-secret
set, Sentry self-hosted will no longer use the verification token for authentication of the webhooks, regardless of whether slack.verification-token
is set or not. Alternatively if the self-hosted instance is unable to be upgraded or re-configured to use the slack.signing-secret
, the logging configuration can be adjusted to not generate logs from the integration. The default logging configuration can be found in src/sentry/conf/server.py
. Services should be restarted once the configuration change is saved.
Information written to log files can be of a sensitive nature and give valuable guidance to an attacker or expose sensitive user information.
While logging all information may be helpful during development stages, it is important that logging levels be set appropriately before a product ships so that sensitive user data and system information are not accidentally exposed to potential attackers. Different log files may be produced and stored for: