Velocity 09: John Allspaw and Paul Hammond, "10+ Deploys Pe

[@alessiopiergiacomi]

Who else got here from the reference in "The Phoenix Project" novel? :)

[@juliocanares]

I got here from the Learning Path Modern DevOps course in Udemy. Imma read the phoenix project book too.

[@LanceMarks]

Exactly. I spend way too much time Googling Eric's rants.

[@dannyknowlton]

Me! Fantastic book ... life-changing!

[@BusterrLTU]

Myself. Lithuania :) recommended book by greate colleague Sambath from India :)

[@Zen-2000-p6r]

me :)

[@thapakazi_]

I watched the talk second time and curated the timestamp if you are in tldr; 3:26 Traditional thinking - devs' job is to add new features to the site - ops' job is to keep site stable and fast 4:06 ops job is to enable business (this is dev's job too) 4:32 The business requires change But change is the root of most of outage two paths: - discourage change in the interest of stability OR - allow change to happen as often as it needs to 5:30 lowering the risk of change via tools and culture deploy logs: - who / when / what Tools | 7:50 - Automated Infrastructure - Shared version control | 9:52 - One step build and deploy | 11: 20 - Feature Flags | 16:01 - Shared Metrics | 22:00 - IRC and IM robots | 28:00 Culture | 29:30 - Respect | 29:50 - Trust | 34:05 - Healthy attitude about failure | 39:30 - Avoiding Blame | 41:45 45:35 Summary Happy Shipping :) 🚀🚀🚀

[@yahaidotdev]

Got here while taking the IBM Introduction to DevOps course, part of the IBM DevOps and Software Engineering Professional specialization.

[@DevOpsCraftsman]

Yes, I saw it recently too!

[@stevendekoffi]

same here

[@bobajob]

Getting a little old but still relevant today. This presentation was my lightbulb moment, after reading the Phoenix Project and getting started with Infrastructure-as-Code. Thanks for helping to kickstart the DevOps movement :)

[@domemvs]

A historical moment

[@k0ns0l]

Old is GOLD!

[@dmitriypo1]

I got here from the reference in The DevOps Handbook :-)

[@grossws]

It's still great talk more than decade later. Even the very small companies should onboard these ideas and look if their processes could be improved to better themselves

[@EduardoFavioANGELES]

today, this video is like Devops Canon

[@tortillatom9308]

That Paul Hammond guy seems pretty cool

[@vev]

from a book

[@chakreshsingh]

I got here after reading the book "Cyber Security Engineering: A Practical Approach for Systems and Software Assurance"

[@nanthinimuniapan6605]

IT Revolution DevOps Guide brought me here

[@DmitriyKl]

It’s been 15 years and Jenkins (Hudson) still looks the same today

[@JoseBarahonajrab66]

starting my devops job in two weeks lml

[@NinjadoLinux]

Muito bom, recomendo que todos os profissionais de TI assistam.

[@jpsecundino2235]

Is this kind of "levers" (17:50) in code still a good practice today?

[@BrycePelletier]

Packt Devops Video 1.1 brought me here.

[@dmlled]

single step build for the win!

[@AndresLeonRangel]

there is some hard truth behind achieving this: you need leadership and NOT bosses involved. You need support from leadership and good healthy influence that transforms the culture. This hard truth applies to most organizations and specially to the very toxic companies out there.

[@piscesinblack]

man this has not aged well

[@SachinGanpat]

In what way?

[@piscesinblack]

Their humor

[@DmitriyKl]

@@SachinGanpat14:38

[@Marma92]

Git2Kube brought me here !

[@vev]

The phoenix Project Gene Kim

[@johnmorgan4368]

I don't really understand the "databases suck" line...

[@kevinmburnett]

the answer to the question "why did the site go down?" is [almost] always related to "the database."

[@KelvinMeeks]

Because (usually) deploying data model changes often require taking the database down (i.e. breaking schema changes) - vs. document-centric type databases, NoSQL databases, etc. - which can be more supportive of Zero Downtime strategies. As a goal of DevOps - to release new features more frequently - a supporting goal is to reach a velocity of releasing [n] times per day, where [n] is a number we want to increase in value over time - needing to take a database down for schema changes - ergo, sucks.

[@tomsantos5982]

i have an interesting beginner studen question to do. is this the twenty one pilot's vocalist?

[@trackisgood]

Explain more what they mean exactly by automated infrastructure. Does this mean using the same image for all servers running in AWS or Azure?

[@KelvinMeeks]

Software Defined Infrastructure. Dynamic, Elastic - adjusting based on changing dynamics of the environment, demand/load, etc. Avoiding the traditional model of needing to submit a request to an Infrastructure team to get just one more server added to a cluster - and being told it will be tens-of-thousands of dollars of additional cost (e.g. the H/W has to be ordered, configured, installed/deployed, integrated into the monitoring fabric, reviewed by security, and procurement will need 6 weeks to complete your request... vs. Having an Infrastructure model that can be automatically configured/dynamically adjusted by Command Line Scripting (CLI) performed as part of the normal deployment process (or as an ongoing self-adjusting, predicted, back-pressure-adjusted model)

[@dunzek943]

@@KelvinMeeks 3 years later. Thanks!

[@japaoshiro]

#SemanaDevOps #LINUXtips #CodeOPS #VAIIII

[@alvaromoe]

Who else came here from another place? :)

[@martamasson8941]

Awesome :)

[@Freedonsharry]

#SemanaDevOps #LINUXtips #CodeOPS #VAIIII