weave-tc
weave-tc copied to clipboard
Amazon EC2
Hello Quentin,
Do you perhaps have any experience deploying this solution to AWS? We used kops to deploy everything and have been running into this issue. Now we launched this weave-tc as a DaemonSet in kube-system, and everything seems fine inside of the weave-tc pods. In other pods however the problem persists. We tried mounting /usr/lib/tc and /sbin/tc from the host machine to the same locations inside of the container.
Any ideas?
Best Regards,
Jordy
Hi there,
We do currently deploy this solution on AWS w/ weave, which solves the problem pretty much entirely. Did you verify that the tc & iptables rules are in place on the hosts? Do you notice the iptables rule marking packets successfully? Do you notice insert_failed increase when the packet loss occurs?
I've figured out that the DNS issues are only resolved when hostNetwork is true on other pods. Whenever it is not, it doesn't work. I'm using Weave 2.4.0, Kernel 4.9 and Kubernetes 1.10. As long as hostNetwork is true obviously iptables is fine and packet marking is okay.
I do have a new AMI ready to go with Linux 5.1.19 and Weave 2.5.2 which solves it too, it's just a little riskier to upgrade to.