We heavily rely on AWS EventBridge in our (cross-accounts) organization to connect pieces of our infrastructure, we recently noticed that our events were not being shared across accounts starting on Tuesday's 6th of April.
After troubleshooting this issue, we realized that the issue did not come from us but most likely from AWS. Each of our accounts has an EventBridge bus with a Resource-based policy with a set of rules and conditions, one of the conditions we used was the following:
"Condition": {
"StringEqualsIfExists": {
"events:source": "com.domain.*"
}
}
We have had to change the "StringEqualsIfExists" by "StringLike" and our connectivity issue appears to be resolved now.
Can anybody confirm whether this change was planned/announced?
I can't imagine the impact this might have had on our infrastructure if we used this condition for production-critical flows.
Q: Why have you used StringEqualsIfExists rather than StringLike?
A: We used the following official documentation