Description of problem: The endpoint information didn't be removed from haproxy.config when service and pod is deleted. please note, first delete service and then delete pod. I'm afraid at this time another people create one pod with IP is same as above pod example: user A create pod/service/route and assume pod ip is (10.129.0.2). now A delete service firstly and then delete pod. the route will cannot be accessed. but the endpoints information still exist in haproxy.config. Check the haproxy.config: <--snip--> server 302789aed6899093dadb89a311076244 10.129.0.2:8080 check inter 5000ms cookie 302789aed6899093dadb89a311076244 <--snip--> now user B create one pod and the pod IP is just to (10.129.0.2). then user A can access his route to user B's Pod Version-Release number of selected component (if applicable): oc v3.2.0.43 kubernetes v1.2.0-36-g4a3f9c5 How reproducible: always Steps to Reproduce: 1. Given the haproxy router is running 2. create pod/service/router in project test 3. Delete the service 4. Delete the pod 5. check the haproxy.config in router container Actual results: As $Description Expected results: the endpoints information should be removed once the service is deleted. Additional info:
Fixed bug and fixed test case PR9408
https://github.com/openshift/origin/pull/9408
openshift/origin/pr9408 -- merged and closed.
This has been merged and is in OSE v3.3.0.8 or newer.
verified this bug on openshift version openshift v3.3.0.8 kubernetes v1.3.0+57fb9ac etcd 2.3.0+git
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, 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-2016:1933