AppSheet Partial Availability - 2025-05-12

 
Issue Summary: From 7:20am PT to 8:20am PT on 2025-05-12 AppSheet had a partial outage in some serving regions due to a security improvement being dialed up causing an overload in traffic in an authentication service internally. 

Impact: A percentage of traffic had failures "The connection to the source path has been lost."
Root Cause: Dialing up a feature that improved segregation and latency of authentication refreshes.
Timeline: 7:20am to 8:20am PT
Next Steps: We have fully dialed up the change and improved the scaling responsiveness of the authentication component internally.

6 11 993
11 REPLIES 11

I asked this on a separate post, but I’ll also ask here, is there a timeline for a service status dashboard?

Yes - we are planning to be on the workspace status dashboard in the next 2 weeks.

I really appreciate how quickly this communication came out compared to incidents in the past. Thank you for taking our feedback and working to improve the platform!

khiemtg_0-1747190321508.png

It seems this error happened again

Thailand now going down

Appsheet dow right now. Huhu

Yeah same, can’t login via the app or website & app dev backend.  

Rolling out more enhanced security across the regions? 

Currently at YVR(BC,CA)

20:20 - Still just stuck in sync web, app, Admin backend 

I’m not at my computer right now, but I tested a few apps and they are syncing without an error.

I’m not having an issue. (US - East Coast)

Dear @Adam_Stevens   and Google AppSheet team,
The recent wave of unstable updates has caused serious disruptions to our operations. Each new change seems to introduce new system crashes or regressions, leading to confusion and frustration not only for developers but for our end users as well.
This is no longer a minor inconvenience—my customers are growing increasingly concerned and are actively considering alternative platforms. The lack of visible quality control and testing is deeply troubling for a product positioned as enterprise-ready.
We urge you to treat this matter with the seriousness it deserves. Please ensure that all updates are properly tested and that clear, timely communication is maintained with your developer community. We rely on Google for stability and professionalism—please don’t undermine that trust.

@Adam_Stevens 

Does this failure affect Automation?
In particular, did the Scheduled Bot work as planned?

Are there any other points that we, the users, need to check due to this failure?