Massive Issues Today
Resolved 💬 27 comments Opened Jun 12, 2025 by vanman2024 Closed Jun 12, 2025
💡 Likely answer: A maintainer (igorkofman, collaborator)
responded on this thread — see the highlighted reply below.
Here’s a clean and complete bug report template based on your input:
---
Environment
- Platform: <!-- Select one: Linux / macOS / Windows / WSL / Other -->
- Claude CLI version: <!-- output of
claude --version--> - Operating System: <!-- e.g., Windows 11, Ubuntu 22.04 -->
- Terminal: <!-- e.g., iTerm2, Windows Terminal, etc. -->
Bug Description
Repeated 503 errors and "Offline" issues throughout the day. MCP servers are not functioning as expected. This issue also occurred yesterday.
The following messages are being returned repeatedly during usage:
⎿ API Error (503 no healthy upstream) · Retrying in 5 seconds… (attempt 4/10)
⎿ API Error (503 no healthy upstream) · Retrying in 8 seconds… (attempt 5/10)
⎿ API Error (503 no healthy upstream) · Retrying in 18 seconds… (attempt 6/10)
⎿ API Error (503 no healthy upstream) · Retrying in 36 seconds… (attempt 7/10)
⎿ API Error (503 no healthy upstream) · Retrying in 33 seconds… (attempt 8/10)
⎿ API Error (503 no healthy upstream) · Retrying in 40 seconds… (attempt 9/10) Steps to Reproduce
- Start Claude CLI from terminal
- Trigger a task that uses MCP
- Observe repeated upstream connection failures
Expected Behavior
- MCP servers should connect and remain healthy
- Tasks should execute without interruption or retry delays
Actual Behavior
- Server reports 503 errors ("no healthy upstream")
- Task execution is halted or delayed
- System reports as "Offline" despite being previously functional
Additional Context
- Issue began intermittently but is now consistent
- No major config or CLI changes made between working state and failure state
- Same behavior observed across sessions and reboots
- Other network-dependent tools function normally
---
Let me know if you want to auto-fill any sections or submit this to a particular support channel.
27 Comments
Experiencing the same issue
same
!Image
Same here as well.
I have seen this before but it usually corrects itself, you having any isssues with the MCP servers as well? Windows did an update yesterday maybe this is what is causing it
Hopefully they fix this soon
Even without MCP servers I get these messages
I also have the same issue
It keeps flashing saying offline this sucks I love claude code but this is a pretty massive interuption I am sure they are on it, Boris will get her fixed.
+1
I went out for dinner - everything was fine - and when I came back, bam! This message keeps looping.
Same here
the whole internet is basically down today
Same here
!Image
Same here
⎿ API Error (503 no healthy upstream) · Retrying in 5 seconds… (attempt 4/10)
⎿ API Error (503 no healthy upstream) · Retrying in 10 seconds… (attempt 5/10)
⎿ API Error (503 no healthy upstream) · Retrying in 18 seconds… (attempt 6/10)
⎿ API Error (503 no healthy upstream) · Retrying in 39 seconds… (attempt 7/10)
⎿ API Error (503 no healthy upstream) · Retrying in 39 seconds… (attempt 8/10)
⎿ API Error (503 no healthy upstream) · Retrying in 39 seconds… (attempt 9/10)
⎿ API Error (503 no healthy upstream) · Retrying in 37 seconds… (attempt 10/10)
⎿ API Error: 503 no healthy upstream
╭────────────────────────────────────────────────────────────────╮
│ > │
╰────────────────────────────────────────────────────────────────╯
? for shortcuts ◯
same :/
Have you seen this before? Does claude code have a status page? I have a feeling it was this windows update that happened overnight that caused this
Come on I just got subscription three hour ago!!
same as above for me
Its unusual its usually very stable
Here is their status page: https://status.anthropic.com/
They are indeed aware of problems.
+1
+1
+1
And the issue is (mainly) the Google Cloud Platform outage: https://status.cloud.google.com/incidents/ow5i3PPK96RduMcb1SsW
OpenAi, Anthropic and Google AI all have issues with their services...
Hey folks, yes there was a massive GCP outage today. It is fully resolved now.
I am not getting the upstream error anymore, but it's still not working It just keeps spinning with no execution ,it just keeps flashing offline intermittently and ridciolously slow to exectute in vs code
This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.