They achieve all of this using 100% open-source infrastructure. If I remember correctly, it's all running on Codeberg-owned hardware as well, not some rented servers.
A company at which I once worked built a functioning server into the frame of a motorcycle. It was after I left, so I'm not sure of the details, including whether it had to be plugged in; but regardless, they called it "the world's fastest server!" and I think that's pretty funny.
Annoyingly I noticed that the status page only shows the past 22 minutes to 1 hour for the primary services. I have no idea why, and there doesn't seem to be a way to look further back. But the badge says 99.45% uptime over the last 14 days, so that's probably right.
To be fair MS makes orders of magnitude more money and has the benefit of operations at scale. Whereas codeberg's operational budget for 2025 was 100k euro and they still need to deal with DDoS and bot scraping. They also were running off a single server up until sept'25 when they had two donated hardware services which are now hooked up to make a 3 node ceph cluster.
more users means, they should do much better than the ones with less users (assuming each user is worth the same/requires same infra).
at the worst case, a bigger org could just copy paste a smaller orgs system a couple times to get the exact same uptime, with same budget per user*. The benefit of bigger orgs is, that they can consolidate these separate system a big system that is more stable AND costs less. If this wasn't true, we wouldn't have big orgs in the first place**.
* yes, it is NOT the same budget for the users. You can't JUST copy paste the system, you'd also need to think how you split it up. I know there are a million little things to nitpick here, but this can all be solved somewhat easily, and they wont change the overall argument.
** regulatory capture, lobbying, corruption and creating a monopoly could also be consider aspects of "consolidating into a bigger system". This doesn't mean why MS shouldn't be able to be better, it just explains why they aren't better.
They also have an history of incidents further down and as you can see they are very short, heck many aren't even incidents since they were on purpose for mantaince and features deployment
It never occurred to me before now but from here on out, there will probably always be some old part of the internet, crumbling and sparse, moldering and broken, populated by far fewer denizens than it was designed for.
I wonder if that'll just be the ever-fading "old folks" internet.
Oh sure, there will always be museums and monuments with little slices of the internet that was, but for the most part, the urge to repurpose old resources to new endeavors means that some parts of the internet will always fade away. I don't know if we'll ever start preserving it perfectly but we certainly aren't there yet.
If I use my phone (not android Auto), I can no longer say, "Navigate to ". It flat out does not work.
Navigate to Local Bakery Xyz.
I'm sorry I can't do that.
(It tries to open the non-existent app for the local bakery).
If I'm in the car that has android auto, it refuses to let me type while in drive (fair enough) and it recognizes the "Navigate to..." Instructions, but if I click on the Maps nav bar for voice and say my destination (it literally says, no text while driving speak your destination)... It tries to open the app.
This shit used to work, it's getting actively dumber.
I once tried to ask my phone to set an alarm. It said it did.
I checked the app. No alarm set.
I tried again, but with a timer. It said it was set.
Again, nothing.
I gave up on digital assistants after that. They took them out back and shot them, and what we see now is their rotted corpse puppeted by Shareholder Value™️ gone wrong. This was 2022.
"Users will completely understand the increased outages if we just eliminate the point-and-click UI that we've spent the last 30+ years getting them used to and instead give them a chat bot that they have to repeatedly type detailed instructions to for marginal results at best."
We're watching Microsoft ruin another company. Its like if EA or IBM buys something, its enshittified and rent seeking occurs for shareholders.
Once this Windows monopoly has passed due to the abysmal quality it will hopefully be over, and hopefully AI helps remove barriers to file portability to hasten their demise.
I think that's how a lot of the internet is dying right now because buying an IP then wringing every drop of value out of its dying corpse before dropping it is a good way to make money right now. This is very much a thing that happens outside of the internet too, and happened long before the internet existed. I think one of the cool things about the internet is how quickly word can spread about this kind of compromised company / product / whatever thing, though I think we need to get better at it. I'm not exactly sure how to accomplish that, it seems like an overwhelming problem, but I think about it a lot.
Not arguing that Gates isn't a total bastard with how he ran things but that was decades ago and he hasn't been in charge for just about as long.
We need to stop making him the target of rants and focus on the people who are actively destroying software and democracy. The current crop of ceos and board members are arguably worse with even less oversight and regulations combined with more wieldable power. Focusing on Gates is just diverting attention from these current monsters.
Server / service downtime. For a well managed company, you would expect these to be almost uniformly green, meaning that all servers are responding correctly almost all of the time. This graph has a lot of yellow and red, indicating severe instability in their services.
Not being able to keep servers running is something that typically happens to smaller companies that grow too fast for them to manage. Established companies are (or, IMO, should be...) expected to have near perfect (>99.99%) uptime, and this is indicative of some expertise loss for the company broadly.
TBF, no, established companies tend to have something between 99.9% and 99.99% of uptime. It only increases if the company is explicitly focused on it, at a large cost that usually needs to be paid by some customer.
But Github pretends to be one of those companies that focus on uptime. And it's also less than 99% right now. So yeah, the main point stands.
Yeah that's fair. It's part of the advertising in some sectors, but not all. A lot of the companies I've bought products from tend to advertise their uptime, and that's the type of company I think about when I think about uptime stats. However, a lot of the companies I've sold products to tended to not talk about it, and their uptime was often in the 2 nines to 3 nines, if not a lot worse. Somehow they still managed to keep going lol. Some of them anyway.
I actually manage servers and network services, and am familiar with the importance of five nines uptime. It seems like this kind of an interface is a failure in that it provides quick information to most people but doesn't include information for people with disabilities. I think it would be beneficial to have that visual interface with color information but to also include information that showed bars of different heights and widths.
I understand that the direct inclusion of all of the numbers could clutter the interface and make it less easy to immediately tell uptime, but even without the numbers it seems that significant improvements could be made to the way this information is presented with this layout.
yeah it's not a great design. It could easily be a line graph where the the vertical axis is a log scale representation of each day's uptime . You'd still be able to tell at a glance which services are stable (top marks across the board) and which services have dips. you could even still have the line change colors when it dips below a certain range if you just really wanted colors on it. it would even zoom infinitely* since the horizontal axis can be arbitrary buckets of time, and the vertical axis is always 0%-100% uptime
down to whatever your health check interval is, at which point the graph would just alternate between 0% and 100%, and up to the full lifetime of the service. still, it could be nice for analyzing some service failures.
At one place I worked we had a service that had been part of an acquired company that, as far as I could tell, had no one responsible for maintaining it, and it either zero or almost zero users, so it would go down for weeks at a time before somebody noticed and did something about it, usually because it needed a security patch. To this day I have no idea why it wasn't shut down but AFAIK it's still out there causing problems for whoever works there now.
We came up with a bunch of ways to describe its uptime: a service has one fortnine of reliability if it stays up for at least one continuous fortnight of the year, for instance. An absolute nine is nine days per year. Fractional nines were invented: a "quarter nine" was 25% of 90%uptime, or 22.5% total uptime.
Man when was the last time you used Windows? The regular restart criticism hilariously outdated
My work computer has mandatory updates from IT like every 2 weeks but when I ran Windows on my own PC, I'd go months without restarting. I've restarted my months-old Fedora install more times than that
I've had uptimes over 1000 days on some of my air gapped linux and BSD machines. Windows never liked going more than month or two, and now unless you turn off automatic updates you never get close to that wall.
I have a company laptop with win11 that some days can't go longer than 6 hours without a reboot because something stopped working. The ubuntu machine I use instead I restart once a month
I and strongly against windows and what microsoft is doing, But you are absolutely correct. If you stick to a build/update that's not trying to brick your NVME, windows desktop uptime is very reasonable.
We're not scoping on stability of thier updates, or the ability to update, just uptime on a run of the mill patched version, it goes as long as you'd need it to for most people.
Now, my linux desktop can go for very very long stretches without updates/reboots if I cared to do it. but windows 11 isn't bad in the way that 95, 98, 2000 were. I'd even argue that win10 was more stable or at the very least had far less breaking issues.
At least their status bars are, presumably, somewhat honest. It's pretty common for the status server being used to track various Lemmy instances to show all green even when the site has clearly been down several hours or even for days.
lol I’m about to kill my copilot subscription. Why have baby autocomplete when you have big daddy Claude now? I don’t even type anything anymore. The whole point of agentic swe is you don’t code by hand. If the ai does something wrong correct it so it doesn’t happen again.
79 Comments
NeatNit@discuss.tchncs.de · 205 pts · 136d
Meanwhile, over at Codeberg: https://status.codeberg.org/
They achieve all of this using 100% open-source infrastructure. If I remember correctly, it's all running on Codeberg-owned hardware as well, not some rented servers.
https://codeberg.org/about
sznowicki@lemmy.world · 162 pts · 136d
They were down for like entire day once because they moved that server to a new location by train. In a backpack.
MNByChoice@midwest.social · 70 pts · 136d
I am disappointed. A few servers have been moved via train and stayed online. Codeberg should do better.
toynbee@piefed.social · 66 pts · 136d
A company at which I once worked built a functioning server into the frame of a motorcycle. It was after I left, so I'm not sure of the details, including whether it had to be plugged in; but regardless, they called it "the world's fastest server!" and I think that's pretty funny.
ZombieChicken@reddthat.com · 5 pts · 135d
I have a dead moyorcycle. I want to do this now.
toynbee@piefed.social · 6 pts · 135d
On behalf of a company that hasn't been my employer for more than half my life, I give you permission.
ZombieChicken@reddthat.com · 6 pts · 135d
I don't need permission to do awesome things. All I need is the budget.
toynbee@piefed.social · 4 pts · 135d
Fair enough. With that I cannot help.
aarmea@lemmy.world · 6 pts · 135d
If that was their only downtime that year, that would have resulted in 99.7% uptime.
Iusedtobeanalien@lemmy.world · 16 pts · 135d
Migrations should always incur downtime
"Hey we're migrating, take a break for a week"
NeatNit@discuss.tchncs.de · 16 pts · 136d
Lol, awesome.
Annoyingly I noticed that the status page only shows the past 22 minutes to 1 hour for the primary services. I have no idea why, and there doesn't seem to be a way to look further back. But the badge says 99.45% uptime over the last 14 days, so that's probably right.
nialv7@lemmy.world · 28 pts · 136d
To be fair the number of users they serve is probably orders of magnitudes lower.
starshipwinepineapple@programming.dev · 36 pts · 136d
To be fair MS makes orders of magnitude more money and has the benefit of operations at scale. Whereas codeberg's operational budget for 2025 was 100k euro and they still need to deal with DDoS and bot scraping. They also were running off a single server up until sept'25 when they had two donated hardware services which are now hooked up to make a 3 node ceph cluster.
bort@sopuli.xyz · 7 pts · 135d
more users means, they should do much better than the ones with less users (assuming each user is worth the same/requires same infra).
at the worst case, a bigger org could just copy paste a smaller orgs system a couple times to get the exact same uptime, with same budget per user*. The benefit of bigger orgs is, that they can consolidate these separate system a big system that is more stable AND costs less. If this wasn't true, we wouldn't have big orgs in the first place**.
* yes, it is NOT the same budget for the users. You can't JUST copy paste the system, you'd also need to think how you split it up. I know there are a million little things to nitpick here, but this can all be solved somewhat easily, and they wont change the overall argument.
** regulatory capture, lobbying, corruption and creating a monopoly could also be consider aspects of "consolidating into a bigger system". This doesn't mean why MS shouldn't be able to be better, it just explains why they aren't better.
astropenguin5@lemmy.world · 15 pts · 136d
Tbf that only shows the past 14 days instead of past 30, but still
Axolotl_cpp@feddit.it · 7 pts · 135d
They also have an history of incidents further down and as you can see they are very short, heck many aren't even incidents since they were on purpose for mantaince and features deployment
bort@sopuli.xyz · 8 pts · 135d
oh fu
osanna@lemmy.vg · 2 pts · 135d
queerlilhayseed@piefed.blahaj.zone · 85 pts · 136d
We're watching the old internet fall apart.
queerlilhayseed@piefed.blahaj.zone · 43 pts · 136d
It never occurred to me before now but from here on out, there will probably always be some old part of the internet, crumbling and sparse, moldering and broken, populated by far fewer denizens than it was designed for.
I wonder if that'll just be the ever-fading "old folks" internet.
marcos@lemmy.world · 23 pts · 136d
You mean sourceforge?
queerlilhayseed@piefed.blahaj.zone · 19 pts · 136d
ascend@lemmy.radio · 16 pts · 136d
https://www.spacejam.com/1996/
queerlilhayseed@piefed.blahaj.zone · 10 pts · 136d
Oh sure, there will always be museums and monuments with little slices of the internet that was, but for the most part, the urge to repurpose old resources to new endeavors means that some parts of the internet will always fade away. I don't know if we'll ever start preserving it perfectly but we certainly aren't there yet.
plateee@piefed.social · 36 pts · 136d
We are Flowers for Algernoning our technology.
If I use my phone (not android Auto), I can no longer say, "Navigate to ". It flat out does not work.
Navigate to Local Bakery Xyz.
(It tries to open the non-existent app for the local bakery).
If I'm in the car that has android auto, it refuses to let me type while in drive (fair enough) and it recognizes the "Navigate to..." Instructions, but if I click on the Maps nav bar for voice and say my destination (it literally says, no text while driving speak your destination)... It tries to open the app.
This shit used to work, it's getting actively dumber.
This morning I got fed up and asked,
"Can I use you to navigate somewhere?"
"Dutch Bros"
(Opens the Dutch Bros app)
Dogiedog64@lemmy.world · 12 pts · 135d
I once tried to ask my phone to set an alarm. It said it did.
I checked the app. No alarm set.
I tried again, but with a timer. It said it was set.
Again, nothing.
I gave up on digital assistants after that. They took them out back and shot them, and what we see now is their rotted corpse puppeted by Shareholder Value™️ gone wrong. This was 2022.
synapse3252@sh.itjust.works · 2 pts · 133d
Oh my god, i'm not the only one!! Thank you for confirming i'm not incapable of using android auto. Such a stupid fucking bug
criss_cross@lemmy.world · 17 pts · 136d
It’s like all companies forgot that reliability is a core feature…
jubilationtcornpone@sh.itjust.works · 14 pts · 135d
"Users will completely understand the increased outages if we just eliminate the point-and-click UI that we've spent the last 30+ years getting them used to and instead give them a chat bot that they have to repeatedly type detailed instructions to for marginal results at best."
-- Vibe CEO's Everywhere
maplesaga@lemmy.world · 16 pts · 135d
We're watching Microsoft ruin another company. Its like if EA or IBM buys something, its enshittified and rent seeking occurs for shareholders.
Once this Windows monopoly has passed due to the abysmal quality it will hopefully be over, and hopefully AI helps remove barriers to file portability to hasten their demise.
queerlilhayseed@piefed.blahaj.zone · 10 pts · 135d
I think that's how a lot of the internet is dying right now because buying an IP then wringing every drop of value out of its dying corpse before dropping it is a good way to make money right now. This is very much a thing that happens outside of the internet too, and happened long before the internet existed. I think one of the cool things about the internet is how quickly word can spread about this kind of compromised company / product / whatever thing, though I think we need to get better at it. I'm not exactly sure how to accomplish that, it seems like an overwhelming problem, but I think about it a lot.
osanna@lemmy.vg · 7 pts · 135d
I REALLY hope MS crashes and burns. They're a shitstain company, and the shit Gates did as CEO was atrocious.
hornywarthogfart@sh.itjust.works · 1 pts · 134d
Not arguing that Gates isn't a total bastard with how he ran things but that was decades ago and he hasn't been in charge for just about as long.
We need to stop making him the target of rants and focus on the people who are actively destroying software and democracy. The current crop of ceos and board members are arguably worse with even less oversight and regulations combined with more wieldable power. Focusing on Gates is just diverting attention from these current monsters.
obvs@lemmy.world · 48 pts · 136d
I’m colorblind, but I’m curious to know what is being represented here.
queerlilhayseed@piefed.blahaj.zone · 78 pts · 136d
Server / service downtime. For a well managed company, you would expect these to be almost uniformly green, meaning that all servers are responding correctly almost all of the time. This graph has a lot of yellow and red, indicating severe instability in their services.
Not being able to keep servers running is something that typically happens to smaller companies that grow too fast for them to manage. Established companies are (or, IMO, should be...) expected to have near perfect (>99.99%) uptime, and this is indicative of some expertise loss for the company broadly.
marcos@lemmy.world · 41 pts · 136d
TBF, no, established companies tend to have something between 99.9% and 99.99% of uptime. It only increases if the company is explicitly focused on it, at a large cost that usually needs to be paid by some customer.
But Github pretends to be one of those companies that focus on uptime. And it's also less than 99% right now. So yeah, the main point stands.
queerlilhayseed@piefed.blahaj.zone · 16 pts · 136d
Yeah that's fair. It's part of the advertising in some sectors, but not all. A lot of the companies I've bought products from tend to advertise their uptime, and that's the type of company I think about when I think about uptime stats. However, a lot of the companies I've sold products to tended to not talk about it, and their uptime was often in the 2 nines to 3 nines, if not a lot worse. Somehow they still managed to keep going lol. Some of them anyway.
Gork@sopuli.xyz · 7 pts · 136d
Thanks. I was thinking it was something biological, or some sort of light spectrum and was getting confused.
queerlilhayseed@piefed.blahaj.zone · 2 pts · 135d
They have always reminded me of bright line spectrographs. Now that you mention it I see the resemblance to DNA tests too.
obvs@lemmy.world · 2 pts · 132d
I actually manage servers and network services, and am familiar with the importance of five nines uptime. It seems like this kind of an interface is a failure in that it provides quick information to most people but doesn't include information for people with disabilities. I think it would be beneficial to have that visual interface with color information but to also include information that showed bars of different heights and widths.
I understand that the direct inclusion of all of the numbers could clutter the interface and make it less easy to immediately tell uptime, but even without the numbers it seems that significant improvements could be made to the way this information is presented with this layout.
queerlilhayseed@piefed.blahaj.zone · 1 pts · 132d
yeah it's not a great design. It could easily be a line graph where the the vertical axis is a log scale representation of each day's uptime . You'd still be able to tell at a glance which services are stable (top marks across the board) and which services have dips. you could even still have the line change colors when it dips below a certain range if you just really wanted colors on it. it would even zoom infinitely* since the horizontal axis can be arbitrary buckets of time, and the vertical axis is always 0%-100% uptime
neuracnu@lemmy.blahaj.zone · 39 pts · 135d
Worst sorting algorithm ever.
BasicallyHedgehog@feddit.uk · 38 pts · 136d
https://mrshu.github.io/github-statuses/ offers a slightly more honest version with aggregate numbers
OrganicMustard@lemmy.world · 46 pts · 136d
90% uptime is abysmal
Any other company would be asked refunds from most clients
criss_cross@lemmy.world · 35 pts · 136d
LMAO 1 nine of reliability.
queerlilhayseed@piefed.blahaj.zone · 14 pts · 136d
At one place I worked we had a service that had been part of an acquired company that, as far as I could tell, had no one responsible for maintaining it, and it either zero or almost zero users, so it would go down for weeks at a time before somebody noticed and did something about it, usually because it needed a security patch. To this day I have no idea why it wasn't shut down but AFAIK it's still out there causing problems for whoever works there now.
We came up with a bunch of ways to describe its uptime: a service has one fortnine of reliability if it stays up for at least one continuous fortnight of the year, for instance. An absolute nine is nine days per year. Fractional nines were invented: a "quarter nine" was 25% of 90%uptime, or 22.5% total uptime.
Lifter@discuss.tchncs.de · 1 pts · 129d
Not even that. One 8.
osanna@lemmy.vg · 36 pts · 135d
yeah, but it's microsoft. what's the longest you've gone without rebooting windows? a couple days? It stands to reason.
glimse@lemmy.world · 4 pts · 135d
Man when was the last time you used Windows? The regular restart criticism hilariously outdated
My work computer has mandatory updates from IT like every 2 weeks but when I ran Windows on my own PC, I'd go months without restarting. I've restarted my months-old Fedora install more times than that
CanadaPlus@futurology.today · 7 pts · 135d
Outing yourself as a Windows abstainer isn't the worst thing, I guess.
glimse@lemmy.world · 4 pts · 135d
Yeah but it's like saying the iPhone sucks because it doesn't have copy and paste lol
onlinepersona@programming.dev · 3 pts · 135d
IPhones dont have copy and paste???
glimse@lemmy.world · 7 pts · 135d
They didn't at launch. It was a perk of Android at the time
BanMe@lemmy.world · 6 pts · 135d
It was bigger than copy/paste really, there were no contextual menus at all yet. So no place to stick the commands.
osanna@lemmy.vg · 1 pts · 135d
They do.
Source: posting from an iPhone
zod000@lemmy.dbzer0.com · 6 pts · 135d
I've had uptimes over 1000 days on some of my air gapped linux and BSD machines. Windows never liked going more than month or two, and now unless you turn off automatic updates you never get close to that wall.
glimse@lemmy.world · 1 pts · 135d
I had automatic updates on and rarely ever got prompted to restart. And when I did, I'd usually ignore it for as long as possible
keyez@lemmy.world · 2 pts · 134d
I have a company laptop with win11 that some days can't go longer than 6 hours without a reboot because something stopped working. The ubuntu machine I use instead I restart once a month
rumba@lemmy.zip · 2 pts · 135d
I and strongly against windows and what microsoft is doing, But you are absolutely correct. If you stick to a build/update that's not trying to brick your NVME, windows desktop uptime is very reasonable.
We're not scoping on stability of thier updates, or the ability to update, just uptime on a run of the mill patched version, it goes as long as you'd need it to for most people.
Now, my linux desktop can go for very very long stretches without updates/reboots if I cared to do it. but windows 11 isn't bad in the way that 95, 98, 2000 were. I'd even argue that win10 was more stable or at the very least had far less breaking issues.
Simulation6@sopuli.xyz · 1 pts · 134d
Win10 was stable, but Win11 was back into the reboot often mode for a while. Not sure if it has gotten better.
tehbilly@lemmy.dbzer0.com · 3 pts · 135d
Personally? Months. Regularly weeks. About the same as my servers. Uptime on a single machine isn't a metric of anything meaningful.
That said, GitHub ain't a single machine and the reliability issues are definitely not a good look.
Zexks@lemmy.world · 2 pts · 135d
My main runs anywhere between 3 to 6 months at a time before i reboot it.
netizen@programming.dev · 1 pts · 135d
A quarter of a century IIRC
flambonkscious@sh.itjust.works · 1 pts · 135d
Win95, maybe
m0darn@lemmy.ca · 34 pts · 135d
Lol I legit thought
desmosthenes@lemmy.world · 26 pts · 135d
https://status.claude.com/ not much better
desmosthenes@lemmy.world · 5 pts · 135d
380b
ArseAssassin@sopuli.xyz · 25 pts · 135d
InvalidName2@lemmy.zip · 19 pts · 136d
At least their status bars are, presumably, somewhat honest. It's pretty common for the status server being used to track various Lemmy instances to show all green even when the site has clearly been down several hours or even for days.
ryannathans@aussie.zone · 11 pts · 135d
Probably pings the servers instead of checking web server works
NewOldGuard@lemmy.ml · 1 pts · 136d
panda_abyss@lemmy.ca · 18 pts · 136d
Two 9’s, the pinnacle of reliability.
WhiskyTangoFoxtrot@lemmy.world · 19 pts · 136d
Five 9s with an 8 in front.
onlinepersona@programming.dev · 13 pts · 135d
Github users right now: I don't care, I'll depend on it harder now!
olafurp@lemmy.world · 2 pts · 134d
Harder daddy
UnfairUtan@lemmy.world · 6 pts · 136d
Gitlab is pretty much the same
diabetic_porcupine@lemmy.world · -5 pts · 135d
lol I’m about to kill my copilot subscription. Why have baby autocomplete when you have big daddy Claude now? I don’t even type anything anymore. The whole point of agentic swe is you don’t code by hand. If the ai does something wrong correct it so it doesn’t happen again.
macaw_dean_settle@lemmy.world · -8 pts · 135d
*yeah, not yea or nay. Do people no longer attend school?
Hominine@lemmy.world · 4 pts · 135d
Cocksure. Look it up.
zarkanian@sh.itjust.works · -11 pts · 136d
Nobody says "yea". Unless they're voting. Or talking about the size of something.
macaw_dean_settle@lemmy.world · -5 pts · 135d
You are correct. It is yeah, not yea or nay. It isn't a vote. Most people are stupid.
osanna@lemmy.vg · 2 pts · 135d
it's actually "yes."