This has gotten to a point where it doesn't really matter anymore. When a service crosses a certain reliability threshold it's like a phase change. The customer base eventually adapts to the situation. Anyone who still genuinely cares has moved to self hosted enterprise or something else by now. It was most tenuous for me when they almost met the SLA. Now that they've blown so far beyond it, the stress is mostly gone.
JsonDemWitOster 2 days ago [-]
I care but I can't move out because it's Orders from Up High which migrated us to Github. We haven't been here a year and it's not worth my neck to mirror our code in some kind of skunkworks forge instance. Woe. Woe is me.
exhilaration 2 days ago [-]
Our Big Giant Enterprise wants to move all our repos from ADO to GitHub for all the nifty AI features, but I'm told the frequent downtime is a major issue so we're slow-walking the migration.
kelseydh 2 days ago [-]
We have so many automated workflows and pipelines moving through Github Actions + other Github integrations it would be a giant headache to migrate. Not clear where we would go either. Gitlab??
SOLAR_FIELDS 1 days ago [-]
I’ve been running an SMB trending mid market out of gitlab for the greater part of a year with no issues. One of my favorite benefits is that ci runners are colocated with the self hosted instance on k8s so suddenly a whole huge slew of shit that you had to care about with GitHub, security, provenance and supply chain is just… not an issue.
Getting off the GitHub actions dependency is a feature, not a bug
PeterStuer 1 days ago [-]
Of course. Every non trivial migration is a 'headache'. But so is being down too many times.
At the very least explore and prepare for alternatives. Map out dependencies that are not trivial to replace. There's probably fewer of those than you think.
I had checked that page not long ago, and as far as I remember there were many "red" or "orange" days in the past 3 months. Now it's all green. That's concerning
tosti 2 days ago [-]
If they've actually improved the service, it would be fine. Is there reasonable plausability, reasonable doubt, or reasonable suspicion?
aweiher 24 hours ago [-]
it should not change any historical records
alfg 2 days ago [-]
Seems like if they changed the criteria of downtime. If you hover over the individual days you’ll see lot of degraded messages, but still green.
2 days ago [-]
hsbauauvhabzb 2 days ago [-]
I clicked through wayback machine and couldn’t see any strong indicators that uptime had been rewritten, but there are a lot of snapshots. If you can prove it, I’m interested.
darkwater 2 days ago [-]
As said, maybe I'm really hallucinating and mixed up the official and unofficial GH status pages, but there are definitely incidents published in the official one that are not counted as downtime or even partial degradation in the uptime counter.
coolgoose 2 days ago [-]
They have an incident reported today, but the status page for actions shows green :D that's fun.
Zealotux 2 days ago [-]
Soon we'll have to track when Github is up.
bernds74 2 days ago [-]
The London Underground have been doing it in reverse for a long time. "There is good service on the Piccadilly line."
Xymist 2 days ago [-]
The phrasing they chose is particularly amusing, since it's always false regardless of the day. They could have said "usual service", and left in the plausible acknowledgement of the fact that it's hot, cramped, miserable, grubby, loud, plague-ridden, mouse-ridden, and generally unpleasant.
JsonDemWitOster 2 days ago [-]
The Google SRE book offers the following as one of the reasons to not gun for 100% reliability (emphasis added):
> users typically don’t notice the difference between high reliability and extreme reliability in a service, because the user experience is dominated by less reliable components like the cellular network or the device they are working with. Put simply, a user on a 99% reliable smartphone cannot tell the difference between 99.99% and 99.999% service reliability!
I've been on a shaky relationship with my ISP of late. What brought me to this thread today is that I couldn't push to Github. Notably this isn't covered by their downtime report so, going by the available facts, it's _probably_ not Github's fault I couldn't push; and I've just been on my daily stand-up call and I got disconnected so frequently.
But looking beyond today's available facts, odds are there's a bigger problem GH is not mentioning in their status page. They say the current incident has to do with "unauthorized users" and I wonder if pushing a commit from my IDE client counts as an operation from an "unauthorized user" as I still have to authorize with my SSH key.
It's just insane I can't decide which between Github or German o2 should be the more reliable service!
ownagefool 2 days ago [-]
Github isn't having a debate over how many 9s they have, they're having a zero 9s problem.
I think there's 3 big themes with this, thought not
1. LLM tools have added considerable load.
2. LLM used by developers to increase velocity seem to be leading more outages. This calls into question the increased velocity.
3. Roadmaps focused on pushing features that aren't reliability problems. i.e. github moving to azure, or adding AI features.
All these same problems happen to orgs with other fads that aren't AI. Following fads is not good engineering.
Grombobulous 2 days ago [-]
Your comment made me think: if GitHub was a Google product with similar popularity and scaling trajectory, would we see similar reliability issues?
Absolutely not. Google has reliability practices so deeply ingrained in their company they’re like an involuntary reflex.
This is a management issue.
PeterStuer 1 days ago [-]
So they failed to manage growth. That is a business management problem first, and only a technical problem second. Yet Github management seems to constantly deflect to operations.
If you take on load (this is 100% by choice) beyond capacity, then obviously the system collapses.
Apparently Github is experiencing a huge increase in usage due to LLMs and this is the cause for a lot of their instability as of late.
PeterStuer 1 days ago [-]
'Experiencing' makes it sound like they have no deliberate choice. No, they let this happen, by choice. They could have prevented this, contractually, by pricing, by governance, but chose not to.
IshKebab 2 days ago [-]
> Put simply, a user on a 99% reliable smartphone cannot tell the difference between 99.99% and 99.999% service reliability!
Sure they can. If Google loads and Github doesn't, then it's clearly Github being down, not the mobile network.
Also not everyone uses a phone. My desktop & fibre internet has way better than 99% reliability.
spondyl 2 days ago [-]
"unauthorized" is a bit different than "unauthenticated". The former suggests trying to access something you don't have permission for while the latter suggests you're just not logged in.
At a guess, I could imagine some sort of failure of cached pages, which can be cached for signed out users but probably not for signed in users (as the rendered HTML would need to have user context like their avatar etc)
DamonHD 2 days ago [-]
First time that GitHub has ever been down for me (London, UK):
504 Gateway Time-out
The server didn't respond in time.
bonesss 2 days ago [-]
I just got the same in Western EU, refreshed and got a site, and now it's gone again with intermittent hiccups of life.
Honestly it's pretty mad to see, especially without a crisp failover.
DamonHD 2 days ago [-]
And it came back for me, at least for a refresh or two...
witx 2 days ago [-]
Slophub is vibing strong
exploraz 2 days ago [-]
> Following investigation, we are seeing that impact is limited to unauthenticated users when accessing Pull Requests or Issues. Our team continues to work towards mitigation with more updates to follow as we have them.
deviation 2 days ago [-]
We should have a pinned post, just for GitHub outages.
Rohunyyy 2 days ago [-]
GitHub is down for like the nth time this week? Time for some truly ground breaking statements like this is the result of LLM (yes it is I know)
JsonDemWitOster 2 days ago [-]
Well thankfully it's Monday so they've only been down for the _first_ time this week!
Keyframe 2 days ago [-]
What's going on with GitHub lately? I've never seen them having so much issues over the years as they do have now.
orwin 2 days ago [-]
Microsoft management/engineering practices+AI slop.
phplovesong 2 days ago [-]
That, but also AI tools have made github req/res go up by 100x. There is simply too muvh AI generated traffic.
I k ow for a fact that ANY other platform would fail faster than github if they had the same volume of http requests.
Keyframe 2 days ago [-]
I get that, but all of that was in there for quite some time though, no?
netdevphoenix 2 days ago [-]
Digital systems don't necessarily deteriorate immediately after the causal factors. Like technical debt, issues grow unnoticed and become visible gradually.
cyberjunkie 2 days ago [-]
Waiting for someone to say this was always the case, not just post Microsoft's takeover.
ramon156 2 days ago [-]
Welcome back everyone
phplovesong 2 days ago [-]
People seem to miss entirely that this is not (only) some slop code that makes github go down, but its the fact that they get 100x the number of requests since AI tools came to the devs daily workflow.
sscaryterry 2 days ago [-]
Now self-hosting, left the dumpster fire.
yanhangyhy 2 days ago [-]
shit i thought it was because my browser. sorry firefox..
why i am keep seeing github down news in HN?
exploraz 2 days ago [-]
Simply because of being a dependency of many things.
In my employment from the last 3 years, I saw twice that there were two migrations from internal systems to GitHub. I would think that companies are doing this for cost-cutting measures. It's not like something I am going to research but it'd be interesting that their recent issues are related to large migrations from in-house installations to github and doubly, if that is related to how large companies might be tightening up their spending in the past few years.
m-schuetz 2 days ago [-]
Anything I'd self-host would be down way more often than github.
ta8903 2 days ago [-]
Only for a scale approaching github's, otherwise a gitea instance or whatever doesn't have any interdependent components other than the server you host it on, which won't have nearly as low a downtime as github (though that's a low bar, a better way to phrase it would be saying it would pretty much never be down).
PeterStuer 1 days ago [-]
My on prem forgejo is way less down than github. I use both. You don't have to chose.
bigstrat2003 1 days ago [-]
I self host my own VCS repos. It's never down. It's not rocket surgery to self host, even for business stuff rather than personal.
lmm 2 days ago [-]
Monday is probably the best day for an outage? Everyone's in the office, so you're not disturbing them, and hung over, so you're not hurting their productivity much.
Rendered at 10:41:13 GMT+0000 (Coordinated Universal Time) with Vercel.
Getting off the GitHub actions dependency is a feature, not a bug
At the very least explore and prepare for alternatives. Map out dependencies that are not trivial to replace. There's probably fewer of those than you think.
API Requests with 4 nines of availability??
Issues with 99.96 uptime?
PR with 99.61% uptime last 90 days??
https://mrshu.github.io/github-statuses/ marks PRs at 95.89% in the same period as an example.
> users typically don’t notice the difference between high reliability and extreme reliability in a service, because the user experience is dominated by less reliable components like the cellular network or the device they are working with. Put simply, a user on a 99% reliable smartphone cannot tell the difference between 99.99% and 99.999% service reliability!
I've been on a shaky relationship with my ISP of late. What brought me to this thread today is that I couldn't push to Github. Notably this isn't covered by their downtime report so, going by the available facts, it's _probably_ not Github's fault I couldn't push; and I've just been on my daily stand-up call and I got disconnected so frequently.
But looking beyond today's available facts, odds are there's a bigger problem GH is not mentioning in their status page. They say the current incident has to do with "unauthorized users" and I wonder if pushing a commit from my IDE client counts as an operation from an "unauthorized user" as I still have to authorize with my SSH key.
It's just insane I can't decide which between Github or German o2 should be the more reliable service!
I think there's 3 big themes with this, thought not
1. LLM tools have added considerable load.
2. LLM used by developers to increase velocity seem to be leading more outages. This calls into question the increased velocity.
3. Roadmaps focused on pushing features that aren't reliability problems. i.e. github moving to azure, or adding AI features.
All these same problems happen to orgs with other fads that aren't AI. Following fads is not good engineering.
Absolutely not. Google has reliability practices so deeply ingrained in their company they’re like an involuntary reflex.
This is a management issue.
If you take on load (this is 100% by choice) beyond capacity, then obviously the system collapses.
Sure they can. If Google loads and Github doesn't, then it's clearly Github being down, not the mobile network.
Also not everyone uses a phone. My desktop & fibre internet has way better than 99% reliability.
At a guess, I could imagine some sort of failure of cached pages, which can be cached for signed out users but probably not for signed in users (as the rendered HTML would need to have user context like their avatar etc)
Honestly it's pretty mad to see, especially without a crisp failover.
I k ow for a fact that ANY other platform would fail faster than github if they had the same volume of http requests.
why i am keep seeing github down news in HN?
https://www.githubstatus.com/history
https://isgithubcooked.com/
Could not have happened on a worse day (Monday) and you can see how unreliable GitHub has been.
Better of self-hosting.
[0] https://news.ycombinator.com/item?id=48418183