Discover and read the best of Twitter Threads about #DevOps
Most recents (16)
Next up is my colleague @drensin on why #SRE and #DevOps are friends! (likely quoting me and @sethvargo a bunch) #VelocityConf
@drensin @sethvargo [ed: full disclosure, this is a sponsored talk.]
How many people have heard of SRE? like 80% of people. But 3 years ago, it was crickets. #VelocityConf
How many people have heard of SRE? like 80% of people. But 3 years ago, it was crickets. #VelocityConf
"Obviously in the age of Twitter, they must fight, right? But no, they actually reinforce each other."
And you should look at @lizthegrey and @sethvargo's video series for more details.
But the key message is: they developed independently but landed in same place #VelocityConf
And you should look at @lizthegrey and @sethvargo's video series for more details.
But the key message is: they developed independently but landed in same place #VelocityConf
So let’s take an org that is experiencing 50%+ drag due to technical debt.
You would think that would ring alarm bells everywhere, right? It would be all hands on deck fixing the issue, right?
Well... (1/13) #leanagile #devops
You would think that would ring alarm bells everywhere, right? It would be all hands on deck fixing the issue, right?
Well... (1/13) #leanagile #devops
“How could YOU ever let it get this way?”
“Well, WE’VE been warning YOU about it since.”
“But YOU never said it would be THIS bad!”
“We did. You’d ask for a plan. We’d give you a plan. You’d say no way. And then we’d settle on a band-aid like 10% time” (2/13) #leanagile #devops
“Well, WE’VE been warning YOU about it since.”
“But YOU never said it would be THIS bad!”
“We did. You’d ask for a plan. We’d give you a plan. You’d say no way. And then we’d settle on a band-aid like 10% time” (2/13) #leanagile #devops
Typically, someone was raising the alarm, somewhere. But it all got watered down...
the org was experiencing growth, hiring lots of people, launching new products. So much noise. Little signal.
“Oh, it’s just developers complaining...” (3/13) #leanagile #devops
the org was experiencing growth, hiring lots of people, launching new products. So much noise. Little signal.
“Oh, it’s just developers complaining...” (3/13) #leanagile #devops
Many incidents happen during or right after the release.
Here are some ideas to improve MTTR and on-call considering this argument
[thread] #OnCall #MTTR #DevOps
Here are some ideas to improve MTTR and on-call considering this argument
[thread] #OnCall #MTTR #DevOps
If to-be-updated services are not covered in the on-call rotations, fail your delivery pipeline.
Add a temporary 15min on-call rotation covering the developers who have issues in the release. Most probably they have something to do with the upcoming alerts.
It's #YorkshireDay! Yes it is. What better way to celebrate than a MASSIVE thread that I'm calling the 'Entirely-anecdotal-Kane-Fulton-#Yorkshire-Startup-Tech-Power-List-100!' These talented folk have been a joy to work with and are doing awesome things in our county. Let's go! 
100. @officialtaras Taras Lanchev - An absolute startup champion who has overseen Startup Weekend in Leeds, exited his cryto tee shirt business Cryto Daddy and is now working on a meeting scheduling app called Focus.
99. @GlobalDigitalWK Global Digital Week's James Munn. Former president of Leeds @UniversityLeeds Enterprise Society, James helps out by mentoring at events such as Startup Weekend, has exited his own startups and gets involved in superb initiatives such as the current Pitch Pot.
Reminder: the talk on intro to #SRE culture and how it relates to #DevOps is at 4:35pm in Moscone West W2005! #GoogleNEXT18
Starting now! What a crowd, and still more people coming in! Love it when people are interested in SRE! #GoogleNEXT18
And off they go!
“How does #ux work with #agile” ?
Step 1: Read the Agile Manifesto for Software Development agilemanifesto.org
Step 2: Read the Principles behind the Agile Manifesto: agilemanifesto.org/principles.html
Step 3: For a modern twist, check out Modern Agile: modernagile.org
1/3
Step 1: Read the Agile Manifesto for Software Development agilemanifesto.org
Step 2: Read the Principles behind the Agile Manifesto: agilemanifesto.org/principles.html
Step 3: For a modern twist, check out Modern Agile: modernagile.org
1/3
...you’ll quickly notice there is no mention of Product Owners, Sprints, User Stories, Story Points, Scrum Masters, Demos, Jira, etc. And no mention of #ux, #datascience, #businessanalyts #devops etc.
There is one key thing though... 2/3
There is one key thing though... 2/3
$AVGO makes a big bet on @CAinc to build integrated digital solutions that combine chip, connectivity, and software. At the very least, there should be more embedded APIs & services as well as joint solutions development coming out of this.
@CAInc has traditionally been strong in #DevOps, #API Management, & #ITOps: the management layer focused on processor, network, & IT SLA performance. Now, Broadcom has a wide range of software tools to track the monitoring and usage of its hardware.
I'll be interested in seeing how @CAInc's Internet of Things & mobile app dev departments expand once $AVGO completes this transaction as well. Getting CAs dev tools closer to silicon can only help performance and help $AVGO to provide out-of-the-box IoT solutions
There are many ways to inspire a “sense of urgency” that have nothing to do with estimates/deadlines.
#agile #lean #softwaredevelopment
#agile #lean #softwaredevelopment
1. First ... giving enough context so people give a fuck. That’s a start. Otherwise, you can’t expect people to “go the extra mile”...
#agile #softwaredevelopment #ux #devops
#agile #softwaredevelopment #ux #devops
2. Let them STOP doing things that are less valuable. Nothing says “you’re free to focus and care” more than freeing people from context switching and multi-tasking.
“What would you like to stop doing, to leave time for this?”
#agile #softwaredevelopment #ux #DevOps
“What would you like to stop doing, to leave time for this?”
#agile #softwaredevelopment #ux #DevOps
If your business relies on an MSP or SI to run your SDLC and release cycles, you are doing it wrong. #DOES18
The MSP and SI have an incentive to artificially limit the amount of changes you can push in any environment, and then up charge based on the number of (totally artificial) allocated build / deploys per month.
Why would anyone want to be beholden to managed services running the change process inside your business....when it can be fully automated?
{narrator voice}: the MSP and SI are not aligned to YOUR operational objectives.
#DOES18
{narrator voice}: the MSP and SI are not aligned to YOUR operational objectives.
#DOES18
Inspired by @patio11 @RachelTobac @HydeNS33k @holman @sehurlburt here is a list "Quick Things Many People Find Too Obvious To Have Told You Already" aka "Things I wish someone had told me earlier"
I've often heard that #DevOps is all about #empathy and I agree.
As an operations person, the most helpful empathetic developers I ever saw were the ones that were told: "20% of your bonus depends on a rating of you from the Operations people"
As an operations person, the most helpful empathetic developers I ever saw were the ones that were told: "20% of your bonus depends on a rating of you from the Operations people"
I didn't believe this for a long time but you can 100% start a blog, write interesting posts and get people to pay you money to tell you more about what's in those blog posts.
Put another way: there are videos of people putting together Duplo on YouTube with MILLIONS of views.
Put another way: there are videos of people putting together Duplo on YouTube with MILLIONS of views.
What if we could just use real code rather than domain specific languages for immutability? What are the advantages of this approach? [ed: p.s. I'm super excited to have met her in person finally!!!] #gluecon
Previously worked at Azure on Azure Functions. Letting people be really productive using serverless offerings.
Where did people struggle? Doing deployments in repeatable fashions. The problem is not the atoms of code, it's deploying the dependency graph. #gluecon
Where did people struggle? Doing deployments in repeatable fashions. The problem is not the atoms of code, it's deploying the dependency graph. #gluecon
For me a serverless (S-) architecture is much more than a event-driven function as a service (FaaS)...
(Thread)
#serverless #faas #architectures #thread #lambda #fastMovingTeams #ownership #devops
(Thread)
#serverless #faas #architectures #thread #lambda #fastMovingTeams #ownership #devops
A S- architecture is an abstraction on top of a (potentially very complex) infrastructure, so that devs can focus on building great features and products and run their code in production without focusing on boring configuration and scalability.
This doesn't mean there are no servers and there is no care for infrastructure: quite the opposite indeed.
Personal rage thread on #Monitoring vs. #Observability. Courtesy of this delineastion: "Monitoring is telling you your systems are broken, Observability is being able to ask why?"
1/
1/
Telling me a system is broken is *worthless* if I can't ask it why it's broken.
"Yo, your shit's broken" has ALWAYS been BAD monitoring.
At times, it may be the only monitoring we have, but it's not the end goal.
2/
"Yo, your shit's broken" has ALWAYS been BAD monitoring.
At times, it may be the only monitoring we have, but it's not the end goal.
2/
Every time I see #Monitoring vs #Observability I wonder if they knew sysadmins could program before #DevOps became a thing. I've been a sysadmin for almost 20 yrs. On day 1, I thought "running the same commands over and over again is dumb, here's some #Perl."
3/
3/
Words matter... (1/5)
Notice how words like discovery, design, delivery, build, and run often get coopted to silo / segment / divide teams/ppl
To draw imaginary lines around skills. To protect kingdoms even. To make pretty left to right diagrams.
#ux #design #agile #devops
Notice how words like discovery, design, delivery, build, and run often get coopted to silo / segment / divide teams/ppl
To draw imaginary lines around skills. To protect kingdoms even. To make pretty left to right diagrams.
#ux #design #agile #devops
First plenary talk: @nicolefv and @jezhumble on measurement. "If you don't know where you're going, it doesn't matter how fast you get there." #SREcon
Outline of the talk: (1) where am I going, (2) why do we care, (3) improve performance/quality, (4) measure performance, (5) culture & how to measure. #SREcon
How ironic is it that some deployments suffer from downtime caused by failing HA mechanisms ? #sysadmin #devops
Ask yourself if you really need HA and consider how much downtime you might run into due to complexity, upgrades, etc. Sometimes, the raw simplicity of a single node can give you better uptime and make maintenances/operations drastically easier. #sysadmin #devops
I'm not advocating against HA: I'm saying it has a cost and going HA doesn't suddenly remove downtime from your vocabulary.
