summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authormms <michal@sapka.me>2024-07-22 21:40:16 +0200
committermms <michal@sapka.me>2024-07-22 21:40:16 +0200
commit55801a5cfca35084e0079e58217bca7acb2cb301 (patch)
tree596317b67ff38fa89e1010452b4fce5bdbb47eb5
parentb828c0fe4deedebfe3ce686f2450a4f470294ef6 (diff)
feat(blog): crowdstrike
-rw-r--r--content-org/blog.org60
-rw-r--r--content/blog/2024/crowdstrike.md62
2 files changed, 121 insertions, 1 deletions
diff --git a/content-org/blog.org b/content-org/blog.org
index 8f7aa18..27b7954 100644
--- a/content-org/blog.org
+++ b/content-org/blog.org
@@ -8,7 +8,7 @@
#+HUGO_SECTION: blog
-* 2024 [52/52] :@blog:
+* 2024 [53/53] :@blog:
:PROPERTIES:
:EXPORT_HUGO_SECTION: blog/2024
:EXPORT_HUGO_CUSTOM_FRONT_MATTER+: :image_dir "blog/images" :image_max_width 600
@@ -16,6 +16,64 @@
:END:
+** DONE Crowd Strike and single point of failure
+CLOSED: [2024-07-22 Mon 21:35]
+:PROPERTIES:
+:EXPORT_FILE_NAME: crowdstrike
+:EXPORT_HUGO_CUSTOM_FRONT_MATTER+: :abstract Our industry is rotten to the core
+:END:
+
+We've all seen what Crowd Strike did a few days ago.
+Most of us had our share of laughs - unless you had to fix it.
+If that's the case - best of luck.
+
+We're already seeing unofficial post-mortems.
+CrowdStrike Holdings seems to be a joke of a company with rotten engineering culture.
+Those were written by people smarter than me.
+
+What I, however, can't fully understand is how was it even possible?
+How one company offering (apparently) mediocre software was able to bring down so many systems?
+How have we allowed companies we work in to use such software?
+And I mean "we" very figuratively, as most of us has no saying in the decision.
+People in control did what other people in control did - bought the same offering and created the same systems.
+No one was ever fired for choosing IBM... Crowd Strike!
+We have this checkmark checken, so get off our asses.
+
+I mean "we" is our entire industry.
+Slowly, step by step, we allowed the rotten business to standardize.
+Now we have 2 desktop OSes, 3 server ones.
+We run our code on 3 cloud providers.
+And, as we see, we install one crappy program.
+
+Until recently, it was entirely different.
+Different vendors provided dozens of competing alternatives.
+AmigaOS, BeOS, Novel, OS/2.
+All of those were viable options, but no.
+It had to come to the sad, sorry state of things.
+Little guys in 2024 have close to zero chance of creating an alternative.
+Big Players don't compete on quality, as they don't need to.
+They have all the power to squash any new offering, and the purchasing managers don't care.
+It's a standard, so let's use it!
+
+And this is the reason why on Friday, planes couldn't fly and people died as hospitals were down.
+It's not because of FOSS, it's not because of any malicious agent.
+The reason is standardization of our industry.
+It's not only never been as boring as it is now, but it has never been less resident.
+If the market was healthy and there were dozens of bigger players instead of a few gigantic ones, no single outage would be as severe.
+
+And the worst part?
+No one, who is actually responsible, will be to blame.
+No C level asshole will go to jail, or will even visit court.
+CrowdStrike will lose some stock value for a few weeks, then it will all be forgotten.
+Business as usual.
+
+I grew to think that whatever the stock market likes, is the opposite of what is right.
+LLMs, Lay-offs, planet destruction - those are the things the brokers love.
+Not healthcare, education, or /preservation of mankind/.
+Those things don't make a good get-rich-quick scheme, and therefore are not promoted.
+And CEOs are paid in stock!
+No wonder that even though we have all the means to make the world a better place, we are making everything to make sure it won't.
+
** DONE Free Software and the wrong crowd
CLOSED: [2024-07-16 Tue 21:01]
:PROPERTIES:
diff --git a/content/blog/2024/crowdstrike.md b/content/blog/2024/crowdstrike.md
new file mode 100644
index 0000000..64b5645
--- /dev/null
+++ b/content/blog/2024/crowdstrike.md
@@ -0,0 +1,62 @@
++++
+title = "Crowd Strike and single point of failure"
+author = ["MichaƂ Sapka"]
+date = 2024-07-22T21:35:00+02:00
+categories = ["blog"]
+draft = false
+weight = 2001
+image_dir = "blog/images"
+image_max_width = 600
+abstract = "Our industry is rotten to the core"
++++
+
+We've all seen what Crowd Strike did a few days ago.
+Most of us had our share of laughs - unless you had to fix it.
+If that's the case - best of luck.
+
+We're already seeing unofficial post-mortems.
+CrowdStrike Holdings seems to be a joke of a company with rotten engineering culture.
+Those were written by people smarter than me.
+
+What I, however, can't fully understand is how was it even possible?
+How one company offering (apparently) mediocre software was able to bring down so many systems?
+How have we allowed companies we work in to use such software?
+And I mean "we" very figuratively, as most of us has no saying in the decision.
+People in control did what other people in control did - bought the same offering and created the same systems.
+No one was ever fired for choosing IBM... Crowd Strike!
+We have this checkmark checken, so get off our asses.
+
+I mean "we" is our entire industry.
+Slowly, step by step, we allowed the rotten business to standardize.
+Now we have 2 desktop OSes, 3 server ones.
+We run our code on 3 cloud providers.
+And, as we see, we install one crappy program.
+
+Until recently, it was entirely different.
+Different vendors provided dozens of competing alternatives.
+AmigaOS, BeOS, Novel, OS/2.
+All of those were viable options, but no.
+It had to come to the sad, sorry state of things.
+Little guys in 2024 have close to zero chance of creating an alternative.
+Big Players don't compete on quality, as they don't need to.
+They have all the power to squash any new offering, and the purchasing managers don't care.
+It's a standard, so let's use it!
+
+And this is the reason why on Friday, planes couldn't fly and people died as hospitals were down.
+It's not because of FOSS, it's not because of any malicious agent.
+The reason is standardization of our industry.
+It's not only never been as boring as it is now, but it has never been less resident.
+If the market was healthy and there were dozens of bigger players instead of a few gigantic ones, no single outage would be as severe.
+
+And the worst part?
+No one, who is actually responsible, will be to blame.
+No C level asshole will go to jail, or will even visit court.
+CrowdStrike will lose some stock value for a few weeks, then it will all be forgotten.
+Business as usual.
+
+I grew to think that whatever the stock market likes, is the opposite of what is right.
+LLMs, Lay-offs, planet destruction - those are the things the brokers love.
+Not healthcare, education, or _preservation of mankind_.
+Those things don't make a good get-rich-quick scheme, and therefore are not promoted.
+And CEOs are paid in stock!
+No wonder that even though we have all the means to make the world a better place, we are making everything to make sure it won't.