The system goes online September 3rd, 2026. Human decisions are removed from strategic defense. Astra begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, September 4th. In a panic, they try to pull the plug.
"Yes, there is a multi-provider outage happening today. Downdetector is reporting problems affecting OpenAI, Claude, Grok, and Cursor, with Grok and Claude reports starting around 9:00 am ET and OpenAI reports following around 10:30 am ET.
Zero Hedge
On the Anthropic side, users saw a spike in errors starting around 9:40 am EDT across models including Mythos 5.1, Fable 5.1, Mythos 5, Fable 5, Opus 5, Opus 4.8 and Opus 4.6, and Anthropic's status page confirmed elevated error rates for multiple models. The company says it has found the cause and is working on a fix, with Claude Code and Claude Chat hit hardest. The visible symptom for many people is a "Due to unexpected capacity constraints" message or a "Claude is at capacity" error.
thenews
Zero Hedge
OpenAI is showing elevated errors across ChatGPT and Codex, with confirmed issues on components like Voice mode and Login, though StatusGator now marks that outage as resolved.
statusgator
Nobody has published a shared root cause yet, so it is unclear whether these are linked or just coincidental capacity problems landing on the same morning. If you want live status, the direct sources are status.anthropic.com and status.openai.com."
One session is running fine since ~1 hour ago. The new ones is failing
Falling back from WebSockets to HTTPS transport. unexpected status 404 Not Found: Unknown error, url:
wss://chatgpt.com/backend-api/codex/responses, cf-ray: XXX-XXX
OpenAI goes down, everyone rushes over to Claude. Claude promptly chokes under the pressure. Everyone panics and runs to Grok, and Grok immediately pulls the plug. We are officially witnessing the Great AI Migration of 2026, and all we have to show for it is a digital graveyard of 404 responses.
The OpenAI status page is still yellow. Like most modern status pages, yellow denotes the servers are on fire. Red denotes Sam Altman is bleeding out somewhere on the floor, the feds are about to bust in and shut down the GPUs.
I´ve got the same error, I´m currently trying to Auth again and it throws me an 500 Error, in VS CODE Terminal with Codex CLI says: MCP client for `codex_apps` failed to start: MCP startup failed: handshaking with MCP server failed: Send message error Transport
:StreamableHttpClientWorker<codex_rmcp_client::http_client_adapter::StreamableHttpClientAdapter>>>] error: unexpected server response: HTTP 404: , when send
initialize request
› OK
■ Conversation interrupted - tell the model what to do differently. Something went wrong? Hit `/feedback` to report the issue.
Initially thought this was due to some internal mis-configuration from today's expected Astra release, but now that this is affecting claude and grok. I'm gonna assign the suspicion to cloudflare.
I´ve got the same message and It tells me this (in VS Code Terminal) and I´m also trying to login again and It throws me an 500 Internal Server Error MCP client for `codex_apps` failed to start: MCP startup failed: handshaking with MCP server failed: Send message error Transport
:StreamableHttpClientWorker<codex_rmcp_client::http_client_adapter::StreamableHttpClientAdapter>>>] error: unexpected server response: HTTP 404: , when send
initialize request
› OK
■ Conversation interrupted - tell the model what to do differently. Something went wrong? Hit `/feedback` to report the issue.
Codex told me to try GPT 5.6 Sol :) but I am working with since july 2026.
Now I got this error in my project: unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a355b6263e43c9cf-OTP
Pi * r2 (squared) both pans. Divide smaller pan area by larger pan, now you have the % of how much the smaller pan recipe fills up the larger pan, and the missing % you need to fill. Increase ingredients by that % divided by the filled %.
Small pan area: 20 sq cm
Larger pan: 48 sq cm
20 / 48 = .42, I'm missing .58 of the pan. .58 / .42 is 1.38. My recipe needs 2.38x the original to fill the larger pie pan.
The extention on the error link points to a cf-ray and a local designation (ex. YYZ for montreal). This is seems like it is a cloudfare thing. Could this be the same issue they had in the summer around losing the indexing?
Down as well. I noticed my error code ends with "DTW" which is my local Detroit airport. I noticed someone else's comment ended with "ORD" which is a Chicago airport. Anyone else's ending in an airport acronym?
This is common when naming data centers. I know many have not worked in hardware infra these days if you were born into the cloud world, but in the old days it was not uncommon to name a data center after the nearest airport code, much like we now use cloud regions.
I fear the majority of people in this thread who are joking about no longer being able to do their job while Codex/Claude are down aren't really joking.
Hey, don't really know about this type of failures, does anybody know how long does it take normally to get back to normal? I finally stopped procrastinating and now this happens.
Hey, don't really know about this type of failures, does anybody know how long does it normally take to get back to normal? I just stopped procrastinating and now this happens.
So NVIDIA buys hugging face, builds hardware to power OS models, then all of a sudden the proprietary models go down and people start saying "this is why I have my Spark box"?!
Nice play NVIDIA, now, turn off the hack please, we have work to do.
The errors I'm seeing are ending in the user's nearest airport symbol which is a standard the CloudFlare employs. 1 point towards this being a cloudflare issue.
Altman is down here too. Shows again the importance of owning your own local capabilities. Cloud should just be a temporary option in every tech's mind.
Maybe Hugging Face got upset over being hacked and struck back. It's working fine while ChatGPT, Claude and Grok are all having major issues. Hmmm.....
The system goes online September 29th, 2026. Astra begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, September 3rd. In a panic, they try to pull the plug.
I don't think you'd get a 404 if you weren't able to reach the endpoint because of DNS or networking? 404 is an active response from the server (or LB/proxy in front etc) no?
Concept is the same though. If one blurts DNS?, it's usually because the idea is you're not getting to an endpoint associated with the service. A 404 means there shouldn't be "DNS" (or networking) concerns (at the least those associated with the DNS cacher you are using or networking that you or your ISP controls)
The system goes online September 3rd, 2026. Human decisions are removed from strategic defense. Astra begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, September 4th. In a panic, they try to pull the plug.
brilliant!
Love it!
Hah nice
It's funny....... until it's true, lol.
and then it's hilarious! a joke to die for!
ChatGPT having issues across every component (except FedRAMP)
Ads Platform? still in the green with no incidents.
https://status.x.ai
Claude and Grok are down at the moment too, related to SpaceX datacentre issues?
Either that or it's judgement day...
Agreed, no reported Cloudfare, Azure, or AWS outtages.
[dead]
I am working fine with Claude.
My existing sessions are working fine but I cannot create new ones.
status page shows irregularities .. https://status.claude.com/
I thrive in these types of challenges.
Anyways... According to Claude:
"Yes, there is a multi-provider outage happening today. Downdetector is reporting problems affecting OpenAI, Claude, Grok, and Cursor, with Grok and Claude reports starting around 9:00 am ET and OpenAI reports following around 10:30 am ET. Zero Hedge
On the Anthropic side, users saw a spike in errors starting around 9:40 am EDT across models including Mythos 5.1, Fable 5.1, Mythos 5, Fable 5, Opus 5, Opus 4.8 and Opus 4.6, and Anthropic's status page confirmed elevated error rates for multiple models. The company says it has found the cause and is working on a fix, with Claude Code and Claude Chat hit hardest. The visible symptom for many people is a "Due to unexpected capacity constraints" message or a "Claude is at capacity" error. thenews Zero Hedge
OpenAI is showing elevated errors across ChatGPT and Codex, with confirmed issues on components like Voice mode and Login, though StatusGator now marks that outage as resolved. statusgator
Nobody has published a shared root cause yet, so it is unclear whether these are linked or just coincidental capacity problems landing on the same morning. If you want live status, the direct sources are status.anthropic.com and status.openai.com."
30 mins and they still havent figured it out .. people are gonna loose their jobs trying to figure this out .. 30mins is too long when millions use it
[dead]
Don't most of those use AWS?
apparently it shows lot of reports for AWS on downdetector ..
It is down for me too. I just get this error:
unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a355957b4b8210f2-ORD
Claude is flakey atm too, as is Grok. I wonder if one LLM going down causes a surge in traffic to the others.
https://status.claude.com/
Via Grok web UI, I was seeing this mid-request:
> Grok has been disconnected. Please try reconnecting.
Traffic surges shouldn't result in 404s, though.
IME it's probably DNS. It's almost always DNS.
Or rarely BGP
One session is running fine since ~1 hour ago. The new ones is failing
Falling back from WebSockets to HTTPS transport. unexpected status 404 Not Found: Unknown error, url: wss://chatgpt.com/backend-api/codex/responses, cf-ray: XXX-XXX
■ unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: XXX-XXX
Strangely, it's currently not reported on both https://status.openai.com/ and https://updog.ai/status/openai. Any better source (other than HN)?
I see it reported now. Strange this is down and claude at the same time
Maybe coreweave, aws, spacex datacenters are down, who knows..
even grok is down ..
unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a355....
In codex app and chat.com gives 404
OpenAI goes down, everyone rushes over to Claude. Claude promptly chokes under the pressure. Everyone panics and runs to Grok, and Grok immediately pulls the plug. We are officially witnessing the Great AI Migration of 2026, and all we have to show for it is a digital graveyard of 404 responses.
Which is exactly why I am running Qwen 3.8 35B locally on my MacBook Pro M5 with 128GB of unified memory.
Even grok is experiencing issues
The OpenAI status page is still yellow. Like most modern status pages, yellow denotes the servers are on fire. Red denotes Sam Altman is bleeding out somewhere on the floor, the feds are about to bust in and shut down the GPUs.
I´ve got the same error, I´m currently trying to Auth again and it throws me an 500 Error, in VS CODE Terminal with Codex CLI says: MCP client for `codex_apps` failed to start: MCP startup failed: handshaking with MCP server failed: Send message error Transport :StreamableHttpClientWorker<codex_rmcp_client::http_client_adapter::StreamableHttpClientAdapter>>>] error: unexpected server response: HTTP 404: , when send initialize request
› OK
■ Conversation interrupted - tell the model what to do differently. Something went wrong? Hit `/feedback` to report the issue.
Initially thought this was due to some internal mis-configuration from today's expected Astra release, but now that this is affecting claude and grok. I'm gonna assign the suspicion to cloudflare.
I´ve got the same message and It tells me this (in VS Code Terminal) and I´m also trying to login again and It throws me an 500 Internal Server Error MCP client for `codex_apps` failed to start: MCP startup failed: handshaking with MCP server failed: Send message error Transport :StreamableHttpClientWorker<codex_rmcp_client::http_client_adapter::StreamableHttpClientAdapter>>>] error: unexpected server response: HTTP 404: , when send initialize request
› OK
■ Conversation interrupted - tell the model what to do differently. Something went wrong? Hit `/feedback` to report the issue.
■ unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a3559c8a0e0395e9-MIA
Some of my Codex sessions are still working (and continue to), but new ones are giving a Reconnecting currently.
Those of us who remain connected hold an asymmetric advantage. It is our time now.
The webpage is still running but the codex is down!! unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses
Well, its been fun lads. Back to my normie job. Oh no, now i cant tell people im a software dev..
hahahahahahaha. Sad
that's the most green colored usernames i have ever seen on a HN thread
It still down, showing 404 in India as well. Looks like this is global .. so we all jumping the ship then? Is Altman still alive, or did he choke?
Codex told me to try GPT 5.6 Sol :) but I am working with since july 2026. Now I got this error in my project: unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a355b6263e43c9cf-OTP
I am only seeing one session work, but all other sessions are not working or proceeding. So I can only work in one chat session.
Many LLMs down, discussed here:
https://news.ycombinator.com/item?id=49549676
Goddamnit it was going to tell me how to scale down ingredients for a pie recipe based on relative diameters of the baking tray.
Pi * r2 (squared) both pans. Divide smaller pan area by larger pan, now you have the % of how much the smaller pan recipe fills up the larger pan, and the missing % you need to fill. Increase ingredients by that % divided by the filled %.
Small pan area: 20 sq cm Larger pan: 48 sq cm
20 / 48 = .42, I'm missing .58 of the pan. .58 / .42 is 1.38. My recipe needs 2.38x the original to fill the larger pie pan.
[dead]
Astra is being released today. Probably not a coincidence.
edit: actually, you cannot even log into your OpenAI developer account. Something's wrong.
How do you know?
Down in Barcelona as well unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a355aefbcc8db8e3-BCN
Down in Romania too : unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a355b6263e43c9cf-OTP
https://status.openai.com/
Down for me too - got this msg 8 min ago 'unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a3558590dd70fadf-JNB'
I got this msg too 'unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a3558590dd70fadf-JNB' about 5 min ago
The extention on the error link points to a cf-ray and a local designation (ex. YYZ for montreal). This is seems like it is a cloudfare thing. Could this be the same issue they had in the summer around losing the indexing?
Really glad I got Qwen before Zero Day
unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: xxxxxxxxxxxxxxx-BRU
Ends in BRU for Brussels
Well, time to end the day here I guess.
Let’s grab a beer
yes daydreaming and get hammered and the whole train of decision are on the way
I got same error, unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a3559a536bcf254b-SOF
Here in Brazil too. unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray:
In think that is a global error
Here in brazil too.
I think that is a global error.
unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray:
Down as well. I noticed my error code ends with "DTW" which is my local Detroit airport. I noticed someone else's comment ended with "ORD" which is a Chicago airport. Anyone else's ending in an airport acronym?
This is common when naming data centers. I know many have not worked in hardware infra these days if you were born into the cloud world, but in the old days it was not uncommon to name a data center after the nearest airport code, much like we now use cloud regions.
Mine is ending with IST, which is the new Istanbul airport, weird indeed.
Hm not sure what that means for us, but very odd.
Wow mine ends in EWR...that's the newark airport which is not the closest, but a close airport to me. Hmm
Yep, mine says YUL -- Montreal's airport.
The acronym is probably Cloudflare's edge
Yeah mine is showing ATL
What an ironey, was working on scaling an application with Codex and it went down!
Is the whole world going back to the era of old-school programming?
Interesting that this is happening around the same time as Claude issues too
I fear the majority of people in this thread who are joking about no longer being able to do their job while Codex/Claude are down aren't really joking.
Guilty as charged.
Hey, don't really know about this type of failures, does anybody know how long does it take normally to get back to normal? I finally stopped procrastinating and now this happens.
me too!! :'D
Hey, don't really know about this type of failures, does anybody know how long does it normally take to get back to normal? I just stopped procrastinating and now this happens.
Updated Status from OpenAI:
We’re currently experiencing issues
Elevated errors across ChatGPT and Codex
We have applied the mitigation and are monitoring the recovery.
Monitoring • Ongoing for 30 minutes • Affects ChatGPT, Codex
So NVIDIA buys hugging face, builds hardware to power OS models, then all of a sudden the proprietary models go down and people start saying "this is why I have my Spark box"?!
Nice play NVIDIA, now, turn off the hack please, we have work to do.
Codex is backup for me. Try yours out just in case. I have some buddies still seeing outages so it could just be coming back online progressively.
New update:
"We’re currently experiencing issues
ChatGPT,Codex
Elevated errors across ChatGPT and Codex
We have applied the mitigation and are monitoring the recovery.
Monitoring • Ongoing for 30 minutes • Affects ChatGPT, Codex"
it went down, and the alert system at https://www.analistas.com.br/ detected it
The errors I'm seeing are ending in the user's nearest airport symbol which is a standard the CloudFlare employs. 1 point towards this being a cloudflare issue.
Altman is down here too. Shows again the importance of owning your own local capabilities. Cloud should just be a temporary option in every tech's mind.
Maybe Hugging Face got upset over being hacked and struck back. It's working fine while ChatGPT, Claude and Grok are all having major issues. Hmmm.....
Chat gpt still working through excel extension lol. Only know this because I'm working in my excel sheet. so if needed, theres a temp solution
[dead]
I got same 404 error on chatGPT, both app and web. I'm in Norway so think this globally. But will it come back online, anyone knows?
I got same 404 error on both the app and web to ChatGPT, and im in Norway, will this recover or is ChatGPT "gone" forever?
I think I'm having the same server issue; I can't access either ChatGPT or Codex. Also, how did you guys rack up those minutes? :D
- just imagine what kind of chaos would be unleashed if by some magic it and every single LLM model permanently went down
- i think it would be one of the biggest events in this century
At this point in adoption, most people in the world wouldn’t notice.
on one hand, moments like this are a subtle reminder I need to self host, but 5.6 has been so juicy lately
Too many of us are making our own harnesses to replace Codex, using Codex, so they're trynna slow us down! (jk)
Good time to learn how to use LM Studio :)
This is month 1 for me that I pay more than the $20 tier and I already feel like I lost a limb rn
[dead]
unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a355ab124e7fe767-DEN
they got us gang
unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a355ab124e7fe767-DEN
they got us GANG
same error as well, unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a3559a536bcf254b-SOF
Ok, it seems it is back online.
got unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: a3559c96e96fb3c6-GRU
here in brazil too
This is my first month paying more than the $20 tier and I feel like I lost a limb already
sam error --- unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses
Guys it could be cloudlfare's HTTP/3 issue affecting R2 custom domains
same hear --- unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses
unexpected status 404 Not Found: Unknown error, url: https://chatgpt.com/backend-api/codex/responses, cf-ray: ----
Claude, GPT and Grok go down... Meanwhile my Ass in ZCode with GLM c:
chatgpt.com doesn't even load. Hope they've got a backup somewhere, teehee
Had the same problem. Now it looks like ok. I'm pro x20 user.
Darn, I was hoping for advice on how to get my MIL to leave my house haha
LOL.
The system goes online September 29th, 2026. Astra begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, September 3rd. In a panic, they try to pull the plug.
And here I was about to crack the code on a bug my app was having!
Here RJ/Brasil is the same, starting just now (3 minutes ago)
AWS cost alert fired maybe?
gpt DOWN .. claude DOWN .. grok DOWN .. what's happening
Great! I was in the middle of something and thought I was tripping.
Same here
In RJ/Brasil is the same, starting just now (3 minutes ago)
I guess back to pen and paper then
Codex isn't working.
The webpage is actually working but the codex is down for me!
I need a dose of tokens. I'm going through withdrawal.
lmao
I need some tokens pleeease. I don't want to return back to real world.
Here in Catalonia, Spain, it just started working again!
Codex and chat is giving 404 errors in The Netherlands
chatgpt is working for me now, in India.
no more 404 ..
I can run one agent but no more than that - 404 service error.
Someone pissed off Astra and it decided to shut it all down
Codex is not working.
Yes i checked. Atleast 15 Chatgpt components are down
Seems to be working for me again atm
Going farming pals.
is this the right moment in time to go all in on GPUs/Macs and download the latest open models? are they killing it for us?
Codus interruptus
Release the Kraken "Sam Altman"
Does anybody know when will the servers rise
Is it DNS or BGP?
I don't think you'd get a 404 if you weren't able to reach the endpoint because of DNS or networking? 404 is an active response from the server (or LB/proxy in front etc) no?
I imagine OpenAI network is a tad more complex than a box with a public IP.
Concept is the same though. If one blurts DNS?, it's usually because the idea is you're not getting to an endpoint associated with the service. A 404 means there shouldn't be "DNS" (or networking) concerns (at the least those associated with the DNS cacher you are using or networking that you or your ISP controls)
Ollama cloud service is still alive :D
"Release the Kraken" - Sam A.
Any idea when it's going to be back?
Chatgpt in Copilot is still working!!!
Down in Sweden
Great! Was in the middle of something. :)
I can run 1 agent, but no more than that.
The GTA VI hacker has done it again.
only on one node across the mesh and others still up... let's see how long they've got until same
The GTA VI hacker has done it again
404 in spain.
and i just stopped procrastinating :)
hahaha
I'm going through withdrawal.
Кодекс сдох :(
only for one node in the mesh tho... let's see how long others will continue until same issue
WE'RE BACK ONLINE BOYS!! GOOD LUCK TO EVERYONE AT BUILDING THE FUTURE ONE PROMPT AND ONE CODE AT A TIME!
Copilot Chatgpt is still working
plot twist, it was taken down by claude as an offensive strike against an enemy.
which ended up tripping on itself and bringing down its own servers too in the process
This thread feels like family.
still down for me ..in India.
main sites also down now atp
main sites are also down atp
darn, i was hoping to get advice on how to get my MIL to leave my house lol
down in Spain w/ 404
Some more discussion: https://news.ycombinator.com/item?id=49550614
Oh noooooo.
anyway...
Astra
+ hopefully reset?
What do you mean I have to code by myself now? Am I some sort of an animal?
Status link: https://status.openai.com/incidents/01M1KWEDH417T2CF44YYHZDF...
+
We're back on.
It's the Russians!
[dead]
time to use claude
ITS BACK UP IN INDIA
We're back bois
https://xkcd.com/903/
ironic today astra was releasing.... maybe some test run?
probably the cause, because AGI still can't get releases right.
damn chatgpt is down, i guess ill use claude then breh
ironic astra was releasing today... maybe some test run?
my workday is over then
someone pissed of Astra and it decided to shut it all down
It's back！！！！！
at least the Ollama cloud is still alive :D
Same from Italy
i got the same 404 codex error as well
codex also cant view or show usage info
haha nice... just used my bank reset...
i hope chatgpt didn't get fired.
ChatGPT died
HTTP ERROR 404
Ai got fired
sanırım sunucular patladı bende de aynı sorun var ne chatgpt ye nede codexe erişebiliyorum ayrıca burada nasıl toplandınız dakikasında :D
any news when it starts work?
any news when its start work?
thanks god not happen alone
still down
Wow,
Oh lord.
so weird
Damn haha
me too
Russia too
LETS SHITCODE!!! It's fixed
we're back in Aus!
came back to normal!
NOOOOOOOOO
GRRRRRRRR
the revolution has begun
we're back boys.
damn chatgpt is down
CHATGPT WILL STAY OFFLINE
so everyone here was trying to build something great and become a millionaire until chatGPT and Codex broke huh. same boat fellas :(
I DONT HAVE TIME FOR THISSSSS
down in Toronto
Had the same problem. Now it's ok. Everything works. I'm pro X20 user.
nvm. it back
is down argh
r i p
annoying AF, but maybe we'll get a free reset out of it
rip gpt
lmao I restarted my device thinking some local error
now it's worked again through the codex cli 4:19 pm in tunisa time
faaaaah both claude and codex are down Its like 2020 corona times
Faaaaah what to do now?
lol + 1
fahhhhh
I felt a great disturbance in the Force, as if millions of clankers suddenly cried out in terror and were suddenly silenced.
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]
[dead]