There are already some VMs running in that network connected to the according L3 router. I see haproxy happily running. After launching a new VM, haproxy has a different PID. The same happens, when a VM is destroyed.
That means, that at least there is no haproxy listening for 3.5 seconds. Nova tells me, that launching a VM takes sometimes only 5 seconds. So it might happen, that the VM already runs cloud-init while haproxy is not running within the L3 router network namespace. eeks!
There are already some VMs running in that network connected to the according L3 router. I see haproxy happily running. After launching a new VM, haproxy has a different PID. The same happens, when a VM is destroyed.
root@network2:~# while (true); do ps afx | grep 48d5 | grep haproxy | wc -l; sleep 0.5; done
1
1
0
0
0
0
0
0
0
1
1
That means, that at least there is no haproxy listening for 3.5 seconds. Nova tells me, that launching a VM takes sometimes only 5 seconds. So it might happen, that the VM already runs cloud-init while haproxy is not running within the L3 router network namespace. eeks!
Are there any possible solutions for that?