Description of problem: As an OpenShift admin running multiple routers, I'd like the ability to randomize the order with which nodes and routes are loaded in order to ensure more even load distribution across the cluster. Version-Release number of selected component (if applicable): OCP 3.2
Proposed PR: https://github.com/openshift/origin/pull/14008
Thanks Brian, can you walk me through that data a little? Is there a particular endpoint that we should focus on there?
https://github.com/openshift/origin/pull/14008
hi, I did not see the fixed version in the https://errata.devel.redhat.com/advisory/28089 and I checked the latest 3.6 env with haproxy image id f2b5e752f72f the fixed PR did not be merged
Verified this bug on openshift v3.6.131 when enable the shuffle, the backend will be random list: cookie ffc46b9e81c946e896a03e31ac25ac9d insert indirect nocache httponly server pod:test-rc-pmh7d:test-service:10.129.0.73:8080 10.129.0.73:8080 cookie 6cdc8fd2b67c5c9d4e866be5480a2445 weight 100 check inter 5000ms server pod:test-rc-hlhw3:test-service:10.129.0.74:8080 10.129.0.74:8080 cookie 794a225214957229fca1711f6cfe7e28 weight 100 check inter 5000ms server pod:test-rc-5tz1z:test-service:10.129.0.72:8080 10.129.0.72:8080 cookie 263e0d26415531881bcdc76c0b4491c7 weight 100 check inter 5000ms server pod:test-rc-khl12:test-service:10.129.0.76:8080 10.129.0.76:8080 cookie 5d28f95106d4403385d68ebd54d6b80b weight 100 check inter 5000ms server pod:test-rc-2j4mm:test-service:10.129.0.75:8080 10.129.0.75:8080 cookie 8f5cff8b0847362a39a32c8c17159079 weight 100 check inter 5000ms
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/RHEA-2017:1716