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
$ time curl -I https://vc.transmute.world/.well-known/did.json
HTTP/2 404
vary: Origin
access-control-allow-origin: *
content-type: application/json; charset=utf-8
x-cloud-trace-context: 8b460a2fbbca1ecbb9007453ae7e5cb2;o=1
content-length: 93
date: Thu, 25 Mar 2021 13:02:28 GMT
server: Google Frontend
real 0m40.265s
user 0m0.037s
sys 0m0.011s
$ time curl -I https://vc.transmute.world/.well-known/did.json
HTTP/2 404
vary: Origin
access-control-allow-origin: *
content-type: application/json; charset=utf-8
x-cloud-trace-context: b536ab1fb35d3b28df28b4316d97eb9c;o=1
content-length: 93
date: Thu, 25 Mar 2021 13:02:37 GMT
server: Google Frontend
real 0m0.389s
user 0m0.033s
sys 0m0.008s
$ time curl -i https://vc.transmute.world/.well-known/did.json | head
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
HTTP/2 200 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
vary: Origin
access-control-allow-origin: *
content-type: application/json; charset=utf-8
x-cloud-trace-context: 3062a3c24f2e4685bb158f85a9433f37;o=1
date: Thu, 25 Mar 2021 13:02:53 GMT
server: Google Frontend
content-length: 5066
{"@context":"https://w3id.org/did/v0.11","id":"did:web:vc.transmute.world","publicKey":[{
[...]
It looks like the web server may be spinning up on demand, since subsequent requests respond faster.
Is vc-http-api meant to test verifying a VC where resolving the issuer's DID may have high latency? Or should the issuer have high availability, like the vc-http-api endpoints themselves?
Are implementations encouraged to locally cache DID documents rather than resolve them dynamically, to prevent such timeouts or other issues - similar to how JSON-LD context files are treated?
The text was updated successfully, but these errors were encountered:
Was 20 a conscious choice, or an offhand assumption ("SURELY resolving a webpage won't take more than 20 seconds on a properly-configured web server")? Offhandedly, my first instinct is to set it to 60 and mention it (and other adjustable timeouts) in the troubleshooting section...
The
did:web
issuer of VC case 5 sometimes responds slowly, causing the test to fail due to timeout.Here is an example of the timeout happening when running the tests in GitHub Actions:
https://github.com/spruceid/didkit/runs/2187670579?check_suite_focus=true#step:14:53
Test timeouts are set to 20 seconds:
https://github.com/w3c-ccg/vc-http-api/blob/b4df10dfdce98b453a2333e7ec1728a10bcc54d1/packages/vc-http-api-test-server/services/jest/cli.js#L4
Timing fetching the DID document with
curl
shows it taking up to 40 seconds:Timed DID resolution
Some time later...
It looks like the web server may be spinning up on demand, since subsequent requests respond faster.
Is
vc-http-api
meant to test verifying a VC where resolving the issuer's DID may have high latency? Or should the issuer have high availability, like thevc-http-api
endpoints themselves?Are implementations encouraged to locally cache DID documents rather than resolve them dynamically, to prevent such timeouts or other issues - similar to how JSON-LD context files are treated?
The text was updated successfully, but these errors were encountered: