New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fixed lp:1442257 - port of #2366 to 1.25 #2392
Conversation
container.ConfigName: "juju", | ||
container.ConfigLXCDefaultMTU: "9000", | ||
|
||
"use-aufs": "false", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sadly no constant when this setting has been introduced.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yeah :/
LGTM 👍 |
|
Status: merge request accepted. Url: http://juju-ci.vapour.ws:8080/job/github-merge-juju |
Build failed: Does not match ['fixes-1457031'] |
the blocker is Fix Committed, so |
Status: merge request accepted. Url: http://juju-ci.vapour.ws:8080/job/github-merge-juju |
Build failed: Does not match ['fixes-1457031'] |
|
Status: merge request accepted. Url: http://juju-ci.vapour.ws:8080/job/github-merge-juju |
Fixed lp:1442257 - port of #2366 to 1.25 This includes 2 fixes for 1.25: * http://pad.lv/1442257 - instead of inheriting LXC containers' host's primary NIC's MTU value for container NICs, now we have an optional integer environment setting "lxc-default-mtu", which when set to a positive number will cause all LXC (but not KVM) NICs to use that MTU. This is a straightforward port of #2365 to 1.25. * Related, but not a fix for http://pad.lv/1443942 (originally proposed as #2190 but it didn't land in time) - only add iptables SNAT rules on the container hosts for EC2 environments, as this breaks MAAS and it's not needed (fallout from the original fix #2071 for bug http://pad.lv/1442801). Live tested on MAAS and EC2 with the following configurations: * bootstrap on trusty, add a precise node * bootstrap on precise, add a trusty node * in both cases above, do a test with and without the address-allocation feature flag * deploy workloads and relate them in LXC and KVM More unit tests around the enableNAT logic will be added later, as described bug lp:1443942.
This includes 2 fixes for 1.25:
primary NIC's MTU value for container NICs, now we have an optional
integer environment setting "lxc-default-mtu", which when set to a
positive number will cause all LXC (but not KVM) NICs to use that
MTU. This is a straightforward port of Fixed lp:1442257 - lxc-default-mtu environ setting added #2365 to 1.25.
as +address-allocation ff: don't use SNAT for LXC/KVM on MAAS #2190 but it didn't land in time) - only add iptables SNAT rules
on the container hosts for EC2 environments, as this breaks MAAS and
it's not needed (fallout from the original fix Mangle source IP address for EC2 traffic that leaves the VPC - fixes LXC on EC2 #2071 for bug
http://pad.lv/1442801).
Live tested on MAAS and EC2 with the following configurations:
feature flag
More unit tests around the enableNAT logic will be added later,
as described bug lp:1443942.