Uploaded image for project: 'Knative Serving'
  1. Knative Serving
  2. SRVKS-99

When cold start takes longer than 30s then --> gateway timeout

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done
    • Icon: Undefined Undefined
    • v0.7.0
    • None
    • None
    • Serverless Sprint 168, Serverless Sprint 169

      When accessing a Knative service via an OpenShift route, and the service startup takes longer than 30s then the user runs into a Gateway timeout:

      curl -v http://random-generator.demo.apps.rhuss-dev.devcluster.openshift.com
* Rebuilt URL to: http://random-generator.demo.apps.rhuss-dev.devcluster.openshift.com/
*   Trying 34.194.121.221...
* TCP_NODELAY set
* Connected to random-generator.demo.apps.rhuss-dev.devcluster.openshift.com (34.194.121.221) port 80 (#0)
> GET / HTTP/1.1
> Host: random-generator.demo.apps.rhuss-dev.devcluster.openshift.com
> User-Agent: curl/7.54.0
> Accept: */*
>
* HTTP 1.0, assume close after body
< HTTP/1.0 504 Gateway Time-out
< Cache-Control: no-cache
< Connection: close
< Content-Type: text/html
<
<html><body><h1>504 Gateway Time-out</h1>
The server didn't respond in time.
</body></html>
* Closing connection 0

      happened on a OCP4 cluster in us-east-1 (curled from Germany).

      It's the router timeout of 30s which kills the connection. The next call obviously works.

              sashture Savita .
              rhuss1@redhat.com Roland Huss
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Created:
                Updated:
                Resolved: