Add another known issue to UCP

This commit is contained in:
Joao Fernandes 2018-05-25 10:04:43 -07:00 committed by Joao Fernandes
parent 8b099eca23
commit 649c09b038
1 changed files with 2 additions and 3 deletions

View File

@ -51,16 +51,15 @@ upgrade your installation to the latest release.
iptables -t nat -D KUBE-MARK-DROP -j MARK --set-xmark 0x8000/0x8000
iptables -t filter -D KUBE-FIREWALL -m comment --comment "kubernetes firewall for dropping marked packets" -m mark --mark 0x8000/0x8000 -j DROP
```
* `ucp-kube-controller-manager` emits a large number of container logs.
* Excessive delay is seen when sending `docker service ls` via UCP client
bundle on a cluster that is running thousands of services.
* Inter-node networking may break on Kubernetes pods while the `calico-node`
pods are being upgraded on each node. This may cause up to a few minutes of
networking disruption for pods on each node during the upgrade process,
depending on how quickly `calico-node` gets upgraded on those nodes.
* `ucp-interlock-proxy` may fail to start when two or more services are
configured with two or more backend hosts. [You can use this workaround](https://success.docker.com/article/how-do-i-ensure-the-ucp-routing-mesh-ucp-interlock-proxy-continues-running-in-the-event-of-a-failed-update).
## Version 3.0.0 (2018-04-17)