The improvement in the new version is significant, and there is no regression.
The improvement is still insufficient for our use case, as I noted would be likely in the updated description. For our use case of 32 LXC containers running on a 32 core (16 hyperthreaded) EC2 box, 21 ephemeral containers were able to correctly complete, with 11 timing out. Previously, a maximum of 11 were able to correctly complete experimentally, and 20+ timed out. In sum, we will need something more--a configurable timeout, or a much larger timeout, like 300. However, the proposed package is good--a very nice improvement in many ways--and I don't see any reason not to release it. -- You received this bug notification because you are a member of Ubuntu Server Team, which is subscribed to lxc in Ubuntu. https://bugs.launchpad.net/bugs/994752 Title: lxc-start-ephemeral's use of dhcp lease table is fragile To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/lxc/+bug/994752/+subscriptions -- Ubuntu-server-bugs mailing list Ubuntu-server-bugs@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-server-bugs