SKALA NETWORKS
SKALA NETWORKS
I opened an issue on Helm to try to get more people to talk about this, but it doesn't seem to pick up any interest... https://github.com/helm/helm/issues/31147
Would running a deployment not be lighter? In any case, we need to have the content of the netfilter synced to avoid disruption when a node/nat gw pod crashes. If...
Could we use a tool like conntrackd and only enable it on "handovers" when we want to switch traffic from one nat-gw to another? This mode could work great for...
How do we ensure compatibility with "old" gateways? Rollout a new API version?
Hey, starting the thread again because I think this is a really important subject for Kube-OVN. Right now, VPC NAT GW have the following issues: - They're Statefulsets, which means...
Yes, statefulsets never come back up when a node is down by design. Deployments will.
@zhangzujian Hi, I'll probably start working on the feature this week, I need HA NAT gateways in production, as running everything off one Pod is not ideal, especially since node...
I'll handle this whenever I get the time, I'm working on a lot of different stuff at the same time and I couldn't get to it properly. I already have...
I'm not sure I understand your message, are you proposing an alternative to what was discussed here or are you making a call for better ideas? BFD/ECMP was pretty established...
I'm honestly confused by the doc on what the VIPs are meant to do when declaring SLRs. I thought they were used to reserve IPs in the IPAM to avoid...