Hi, guys.
This might not be a Camel-specific error at all, but given that we only see
this in one of my Camel + Spring Boot apps I thought it might be worth a shot
to ask. If not a Camel or Spring issue, maybe it's an OpenShift issue others
have seen?
I'm in a RHEL environment, using Camel 3.8.0, Spring Boot 2.4.2, and OpenShift
3.x.
Brief description: This Boot/Camel app (MyApp) is deployed in an OpenShift
environment. The seda://call-service route is called many times, and the route
in turns makes an HTTP call to another service (MyService) in the same
OpenShift deployment. 'MyService' is scaled up to 10 pods.
MyApp basically generates about a 1000 calls to seda://call-service which then
generates a 1000 HTTP calls which are load-balanced by OpenShift.
After several weeks of running, we eventually get the following
"NoRouteToHostException", which is typically some kind of router problem. But
as I said, MyService is in the same OpenShift project as MyApp and there is no
router between them. I log the full URL of the failed HTTP call, and when I
click on it in the browser the call works; i.e. MyService is just fine and
there seems to be no connectivity issue.
When I restart MyApp, all the HTTP calls are once again fine! So it's like
some resource is being used up over time...
What the heck could it be? Where can I even begin poking around next? Is
there an OpenShift log I should be looking at?
Anyway, thanks for any tips.
The error is pasted below.
Thank you very much.
Ron
-----
[...]
seda://call-service: Call to service failed. Exception: Cannot assign requested
address (Address not available)
seda://call-service: o.apache.http.impl.execchain.RetryExec: I/O exception
(java.net.NoRouteToHostException) caught when processing request to
{s}->https://HOST:PORT: Cannot assigned requested address (Address not
available)
seda://call-service: o.apache.http.impl.execchain.RetryExec: Retrying request
to {s}->https://HOST:PORT
[...]
