[Uptime] Add availability alerts

[andrewvc]

Personas / User Stories As a system administrator I want to be proactively notified if my availability breaches SLAs and SLOs

Susie Sysadmin is in charge of deploying and monitoring 20-30 internal services that her organization runs. She uses uptime to provide alerts for when these services fail. She'd also like to know about when these services are not meeting SLOs. She'd like to know if services have been generally unreliable or slow, so she's established an SLOs for these services having availability of 98% over a given month. She'd like to receive alerts for when services are at risk of breaching these thresholds or have breached them already.

Ishtar Inframan supports a IaaS platform at a large company that uses Uptime to provide availability and up/down monitoring across the company to many teams. Alerting for availability breaches would go to both Ishtar and the team whose service is at risk, and let them see how apps are generally doing in terms of health.

Carrie is a sysadmin who wants to monitor SLOs for third party service providers, for instance adobe tag manager. In the past their Javascript embeddable has been unavailable and she wants proactively monitor it for availability as well as get up/down alerts. When it comes time for renewal these QoS stats could be integral to the discussions.

ACs

• Availability alerting will be configurable in existing up/down alerts flyout
• Show the 'grouping' portion of the alert first (Kuery bar + filters)
• Show the up/down alerting section second

  MATCHING MONITORS ARE DOWN > 5 times
  WITHIN THE last 15 minutes

• Show the new availability section last

  MATCHING MONITORS ARE UP IN < [99%] of checks
  WITHIN THE LAST [30 days]

• The up/down and availability sections could be disabled with a check box. For existing alerts this box should be disabled by default on availability. New alerts should have both enabled
• Message variable changes:
  • context.message: Down Monitors: TOP_3_MONITORS(DOWN > N TIMES IN THE LAST N MINUTES)\nMonitors Below Availability Threshold (THRESHOLD %): TOP_3_UNAVAILABLE_MONITORS. We should display monitors as NAME_OR_ID( URL )

[elasticmachine]

Pinging @elastic/uptime (Team:uptime)