Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

AlertManager Sends alert notification when the alert in a group gets resolved #1195

Closed
vivek425ster opened this issue Jan 13, 2018 · 3 comments · Fixed by #1205
Closed

AlertManager Sends alert notification when the alert in a group gets resolved #1195

vivek425ster opened this issue Jan 13, 2018 · 3 comments · Fixed by #1205

Comments

@vivek425ster
Copy link

vivek425ster commented Jan 13, 2018

Whenever an alert in a group gets resolved alertmanager sends a notification of an alert with the remaining alerts in a group. This notification is not of a resolved alert but it is of the remaining alerts.

For example if i have 4 alerts(A, B, C, D) firing in a group. I get a notification with 4 alerts firing. Now if an alert 'C' gets resolved. I get a notification with 3 alerts(A, B, D) firing after group_interval period. Ideally this shouldn't happen since i have already received a notification of these alerts (A, B, D) firing. We should only get a notification if any new alert is added in a group.

  • Alertmanager version:
    0.12.0

  • Prometheus version:
    2.0

For config file:
https://groups.google.com/forum/?utm_medium=email&utm_source=footer#!msg/prometheus-users/4nibd10rJ_0/eqKZ_1gSAQAJ

@simonpasquier
Copy link
Member

IMO this is the intended behavior, at least this is what users currently rely on. If many people ask for it, maybe there could be an optional parameter telling AlertManager to send notifications only about new firing alerts.

Having said that, the description of the group_wait parameter could be improved to say that it encompass firing and resolved alerts.

@brian-brazil
Copy link
Contributor

group_wait should have no impact on resolved alerts, as it delays the start of a group which is from firing alerts.

The OP's desired behaviour is correct though, the user already knows that ABD are firing so there's no need to tell them again before the repeat_interval.

@simonpasquier
Copy link
Member

In that case, the issue is valid since currently the DeDup stage will pass the alerts to the Retry stage if the current set of resolved alerts (eg [D]) isn't a subset of the nflog entry (eg []). See

if !entry.IsResolvedSubset(resolved) {
return true, nil
}

hh pushed a commit to ii/alertmanager that referenced this issue Dec 20, 2018
Signed-off-by: Paul Gier <pgier@redhat.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants