Skip to content

Conversation

@wolfpack9377
Copy link
Contributor

…ashes if provided a negative value

During the AWS outage today, the aws_credentials app ultimately was terminated because the supervisor reached the max restart intensity. The crash that was causing the restarts was:

,[{erlang,send_after,[-1000,<0.2314.0>,refresh_credentials],[{error_info,#{cause => time,module => erl_erts_errors}}]},{aws_credentials,setup_callback,1,

Unfortunately, the actual value being returned for the expiration during this period was not in the stack trace. I'm sure this is an exceptional case but if the app were to remain running, it could, in theory recover once the service disruption is over.

@onno-vos-dev
Copy link
Member

Hi @wolfpack9377 very well aware of this issue 😅 And was going to write a similar fix as you proposed myself tomorrow so thank you for the PR 🙏

I'll roll out a release in the morning with this fix. Considering AWS appears to be recovering I'd like some sleep soon 💤😅

@onno-vos-dev onno-vos-dev merged commit f4691cf into aws-beam:master Oct 21, 2025
6 checks passed
@onno-vos-dev
Copy link
Member

@wolfpack9377
Copy link
Contributor Author

Awesome! Thank you for the quick turnaround!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants