Skip to content

How to: page us (for urgent support)

During your customer-onboarding, we supplied you with an email address for urgent support. It should look similar to {your cluster}-{some random characters}@engflow.pagerduty.com.

If you take away just 1 thing from this how-to: email that address.

Our production on-call will receive a page, and should start to engage with you within minutes1.

What is a page

A page will be loud and interruptive (think of your phone playing a red alert klaxon or air-raid siren on max volume), waking a person up from sleep if necessary.

We ask you to be mindful of our responders when choosing to use this workflow. Thank you!

Examples
  • "builds are running slowly / failing compared to x ago"
  • "no users can see invocations being added to the UI"
  • "we see all builds stuck in the queue, blocking x people from doing y"

Help us to help you

The most important thing is to email the paging-address so we can get the ball rolling helping you.

There are some things you can include in that contact, or have waiting for when we engage, that will help speed us up:

What's happening?

  • What symptoms are you seeing?
  • When did this start happening?
  • Is this recently-bad compared to the same time yesterday / last-week?

Including material like logs, charts, other data will help to quickly and clearly communicate.

What's the impact severity for you?

  • What does this block or slow down?
  • How many people are affected?

Severity examples

(non-exhaustive lists)

Severity 1
  • builds are consistently slower on RE than without RE
  • build queue is growing but not being serviced
  • a particular feature stops being functional (including being too slow to reasonably use compared to "normal")
  • most/all people are blocked due to an issue with our service
  • security: endpoints are routeable or accessible to people they should not be
Severity 2
  • a particular feature stops being functional, but there is a workaround
  • some subset of users are blocked or significantly slowed down due to an issue with our service
  • builds are slower than "usual"
Severity 3+

Please refer to how to: get day-to-day support.

Has anything at your side changed recently?

We'll be looking at what changed on our side recently - the product, or the infrastructure it's hosted on.

We have no visibility into your source control or build-implementation, though; this is where we will need guidance from you.

  1. The response-time SLA varies according to the level of support included in your contract. ↩