-
由 Sarah Yasonik 创作于
The endpoints configured via alert integrations have the ability to be rate-limited by project. For gitlab.com, the default values are now being utilized.
由 Sarah Yasonik 创作于The endpoints configured via alert integrations have the ability to be rate-limited by project. For gitlab.com, the default values are now being utilized.
代码所有者
将用户和群组指定为特定文件更改的核准人。 了解更多。
incident_management_rate_limits.md 1.67 KiB
type: reference
stage: Monitor
group: Respond
info: To determine the technical writer assigned to the Stage/Group associated with this page, see https://about.gitlab.com/handbook/engineering/ux/technical-writing/#assignments
Incident management rate limits (ULTIMATE SELF)
Introduced in GitLab 12.5.
You can limit the number of inbound alerts for incidents that can be created in a period of time. The inbound incident management alert limit can help prevent overloading your incident responders by reducing the number of alerts or duplicate issues.
As an example, if you set a limit of 10
requests every 60
seconds,
and 11
requests are sent to an alert integration endpoint within one minute,
the eleventh request is blocked. Access to the endpoint is allowed again after one minute.
This limit is:
- Applied independently per project.
- Not applied per IP address.
- Disabled by default.
Requests that exceed the limit are logged into auth.log
.
Set a limit on inbound alerts
To set inbound incident management alert limits:
- On the top bar, select Menu > Admin.
- On the left sidebar, select Settings > Network.
- Expand Incident Management Limits.
- Select the Enable Incident Management inbound alert limit checkbox.
- Optional. Input a custom value for Maximum requests per project per rate limit period. Default is 3600.
- Optional. Input a custom value for Rate limit period. Default is 3600 seconds.