Reminders for Unresolved and Unclosed Incidents & Editing Incident Times

[Erica Dohring]

My team often forgets to close incidents and we often aren't disciplined about doing postmortems.

Is there a solution to be really annoying and try to enforce these behaviors?

For example, slack reminders or even alerts to oncall to close them.

Additionally, is it possible to edit a P-Level and incident time after closing or resolving?

1. Sometimes people close alerts/incidents when something is still ongoing by accident.

2. Sometimes an incident "starts" over slack. I don't want to necessarily start an incident and start paging the oncall if they are busy trying to tackle the issue at hand, but would love to have those timestamps correct for reporting.

[Erica Dohring]

Answer re: editing P-level and incident times

Resolved & Closed Incidents

✅ P-Level

✅ Impact started at + ended at -> which calculates impact duration

✅ Detected at -> which calculates impact detection

❌ Elapsed time <- does not seem editable (probably calculated from created at and updated at)

[Aysun Özgün Denizci]

Hey Erica,

In Actioner, we are designing Slack apps that integrate your toolset and help you automate your flows. In our app directory, we introduced a solution for incident management with PagerDuty. We also do have a solution for Opsgenie alerts and incidents, but it is currently missing notification flows.

I wanted to mention about these since I believe it is possible to build a solution that works with Opsgenie and Slack for all cases you mentioned.

This being said, we are also using Opsgenie for managing our alerts/incidents and facing similar problems. We have built some custom logic to handle these cases.

Let me proceed to explain how we are internally handling similar ones and how our product can help with the use-cases you asked about.

1. You can define a recurring job in Actioner that polls open incidents daily and notify responders in Slack when there is an incident that is not closed for X days.

2. Instead of closing an incident, you might add a label to your incidents, such as "closing" or "resolved". When this tag is added, Actioner can notify responders in Slack to adjust the P-level and then resolve & close it. This would also prevent accidental closing. Alternatively, you can set Actioner to close the incident X hours after such a tag is added.

3. You might dedicate a #problem-reporting Slack channel where your users (or stakeholders) can report an incident by selecting affected service. Accordingly, you can set Actioner to add "problem-reporting" tag to the incident reported in Slack and adjust Opsgenie not to notify responders for these ones, or apply a softer notification policy instead of an agressive one.

Hope these help! In case you'd like to explore more, you can ping me or reach out to us through our Contact page. We'd be happy to walk you through some steps to build such automations.

Best regards,

Aysun from Actioner

[Mansi]

Hi, 
We at pagerly.io have solved this.
We ensure that you are able to get proper SLA alerts for all tickets / incidents on Slack passing any due date. 
Do reach out to us for any requests/ customisation