The PR merge made into "4.5.0-0.nightly-2020-08-13-011355" release. With the fix in place it is noted that the ingress operator now goes into degraded and triggers the alerts when the router goes down for conditions like misconfigured routes:
$ oc -n openshift-ingress get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
router-default-5d7dffdbf-4f5p6 1/1 Running 0 5h31m 10.131.0.14 ip-10-0-155-247.us-east-2.compute.internal <none> <none>
router-default-5d7dffdbf-dpcxv 1/1 Running 0 5h31m 10.128.2.3 ip-10-0-164-143.us-east-2.compute.internal <none> <none>
router-internalapps-796c7b8bd8-brgnj 1/2 CrashLoopBackOff 33 114m 10.129.2.13 ip-10-0-223-101.us-east-2.compute.internal <none> <none>
$ oc -n openshift-ingress logs router-internalapps-796c7b8bd8-brgnj -c router --tail 5
E0813 10:12:10.322714 1 haproxy.go:416] can't scrape HAProxy: dial unix /var/lib/haproxy/run/haproxy.sock: connect: no such file or directory
E0813 10:12:10.333434 1 limiter.go:165] error reloading router: exit status 1
[ALERT] 225/101210 (18) : parsing [/var/lib/haproxy/conf/haproxy.config:343] : timer overflow in argument '999d' to 'timeout server' (maximum value is 2147483647 ms or ~24.8 days)
[ALERT] 225/101210 (18) : Error(s) found in configuration file : /var/lib/haproxy/conf/haproxy.config
[ALERT] 225/101210 (18) : Fatal errors found in configuration.
The prometheus UI now has the "template_router_reload_fails" metric displaying the reload failures.
Created attachment 1711306 [details]
Prometheus graph data from patched cluster version
Marking this bug as "verfied" in reference to https://github.com/openshift/router/pull/167.
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory (OpenShift Container Platform 4.5.7 bug fix update), and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.