You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The DID Resolver caches DID Resolution results for 3600 seconds or 1 hour. In a test environment, we've run into a scenario where a shorter TTL would have been convenient but I have doubts that this translates to any real world scenarios.
For context, the situation we encountered is related to fixes made in #3371. We had an ACA-Py instance with a public DID working with a mediator. In did:sov terms, that means an attrib txn with endpoint and routing keys in it. We then wanted to be able to spin up the ACA-Py instance with a new mediator and then connect again from another agent. The fixes in #3371 enable this but from the other agent, since we'd connected once before, it was using its cached version of the did:sov document, resulting in a failure to connect.
Having a caching layer enables us to use the DID Resolver at multiple layers (credential issuance, verification, DIDComm messaging) without needing to worry too much about the efficiency of retrieving the DID Document.
Interested to hear if others have thoughts on what a good TTL would be.
The text was updated successfully, but these errors were encountered:
The DID Resolver caches DID Resolution results for 3600 seconds or 1 hour. In a test environment, we've run into a scenario where a shorter TTL would have been convenient but I have doubts that this translates to any real world scenarios.
For context, the situation we encountered is related to fixes made in #3371. We had an ACA-Py instance with a public DID working with a mediator. In did:sov terms, that means an attrib txn with endpoint and routing keys in it. We then wanted to be able to spin up the ACA-Py instance with a new mediator and then connect again from another agent. The fixes in #3371 enable this but from the other agent, since we'd connected once before, it was using its cached version of the did:sov document, resulting in a failure to connect.
Having a caching layer enables us to use the DID Resolver at multiple layers (credential issuance, verification, DIDComm messaging) without needing to worry too much about the efficiency of retrieving the DID Document.
Interested to hear if others have thoughts on what a good TTL would be.
The text was updated successfully, but these errors were encountered: