Comment 12 for bug 1636708

First, I believe the original description of this case shows it's the same problem (or same fix at least) as bug 1573272 - the vlan is racing with its raw device bond to come up. The fix in that bug should address this problem, by forcing the bond interface to always come up first (it's forced during udev processing when udev detects a new vlan interface). So I think this bug can be marked as fixed or duplicate - can anyone who had the original problem please test with the latest vlan package to verify this is fixed?

Second, @tom-verdaat, re: comment 11, I set up a system with that config, and rebooted many times and saw no problem. As your description doesn't involve dhcp on the vlan interface - which is what caused the long-term locking of the ifupdown bond0 lock from the initial problem description - it sounds like a new issue. Especially since this is an older bug, and you only now started seeing the problem after upgrading to the latest vlan package. Can you open a new bug for your issue please?