Description of problem:
After upgrading from 4.3 to 4.4 nightly, the server directive in haproxy.conf backend is not picking up endpoints/pods.
~~~~
Good haproxy.config file
~~snippet~~~
backend be_edge_http:da-saas-dev-apigee:da-saas-dev-us-login-ui
http-request set-header X-Forwarded-Proto-Version h2 if { ssl_fc_alpn -i h2 }
http-request add-header Forwarded for=%[src];host=%[req.hdr(host)];proto=%[req.hdr(X-Forwarded-Proto)]
cookie 4d647911d32fe9d240f81a98ba8a37a5 insert indirect nocache httponly secure
server pod:login-ui-4-gc4tt:login-ui-svc:172.20.57.101:8081 172.20.57.101:8081 cookie 3a6baad05e3bfd9d4b5ae8a65489c811 weight 256 <======== pod entry
Bad haproxy.config file
~~~snippet~~~
http-request add-header Forwarded for=%[src];host=%[req.hdr(host)];proto=%[req.hdr(X-Forwarded-Proto)]
cookie 0d9795a3241921a94276bd5cdbc98145 insert indirect nocache httponly
..... <========== No pod entry
~~~~
Version-Release number of selected component (if applicable):
4.4.0-0.nightly-2020-07-30-212136
How reproducible:
Intermittent. The customer noticed that they were able to delete routes and occasionally they would populate the haproxy.conf on their ingress routers.
Steps to Reproduce:
1. Create route/svc with pod backing it
2. oc rsh pod and vi haproxy.conf
3.
Actual results:
server directive is missing / routes not working
Expected results:
server directive exists with pod/endpoint and routes redirect to the pod/endpoint.
Additional info:
must-gather, ingress controller logs, haproxy logs, haproxy.config will be linked.
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 4.6 GA Images), and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.
https://access.redhat.com/errata/RHBA-2020:4196
Description of problem: After upgrading from 4.3 to 4.4 nightly, the server directive in haproxy.conf backend is not picking up endpoints/pods. ~~~~ Good haproxy.config file ~~snippet~~~ backend be_edge_http:da-saas-dev-apigee:da-saas-dev-us-login-ui http-request set-header X-Forwarded-Proto-Version h2 if { ssl_fc_alpn -i h2 } http-request add-header Forwarded for=%[src];host=%[req.hdr(host)];proto=%[req.hdr(X-Forwarded-Proto)] cookie 4d647911d32fe9d240f81a98ba8a37a5 insert indirect nocache httponly secure server pod:login-ui-4-gc4tt:login-ui-svc:172.20.57.101:8081 172.20.57.101:8081 cookie 3a6baad05e3bfd9d4b5ae8a65489c811 weight 256 <======== pod entry Bad haproxy.config file ~~~snippet~~~ http-request add-header Forwarded for=%[src];host=%[req.hdr(host)];proto=%[req.hdr(X-Forwarded-Proto)] cookie 0d9795a3241921a94276bd5cdbc98145 insert indirect nocache httponly ..... <========== No pod entry ~~~~ Version-Release number of selected component (if applicable): 4.4.0-0.nightly-2020-07-30-212136 How reproducible: Intermittent. The customer noticed that they were able to delete routes and occasionally they would populate the haproxy.conf on their ingress routers. Steps to Reproduce: 1. Create route/svc with pod backing it 2. oc rsh pod and vi haproxy.conf 3. Actual results: server directive is missing / routes not working Expected results: server directive exists with pod/endpoint and routes redirect to the pod/endpoint. Additional info: must-gather, ingress controller logs, haproxy logs, haproxy.config will be linked.