Ready for more DevOpsDays?

DevOpsDays DC will return June 8-9, 2016.

Devops and Ninja an Outage

Facilitator:

Scribe: kyle mueller

contact M Kyle Mueller - @_kylemueller questions is the answer automating failovers

is there anything for peer point between 2sites how to force fight.

Depends on the outage

cloud Treat as non-peramanent - possible with hardware

Physical - Find the redundancy -Peers fighting with each other

Networks trouble detecting peers and ISPs metrics from users , from network akamai backbone cross internet and bounce , can go horrible wrong

Alerting what can you catch with alerting and monitoring, where is the value in monitoring to little or two much, no reaction after or before. Informational alerts - id like to know when it happens but i cant do anything about it

Other graceful degredation - software service inteliegently communicated to each other

retrospectives - can we improve or gain quick resolution to repair if happens again.

Internal management - ie twitter failwail CDN 404 based page admit to being down or having issues

Block possible with dam to waterfall issue to allow fix of down issue

automatic remmediation

apply the pain where it needs to go , dont look back at best practices, do them. All pains are ours. Loud alerting on service or application point to dev

internal service/app built with health check

microservices, health checks internal to application - and its connections