The /var/log/openvswitch/ovs-vswitchd.log around the time the puppet manifest is being applied should have some more information to confirm this. Are there any "ERR" messages?
It would also be useful to have the output of the following from the compute node. It might be useful to lock it first to get out of the reboot loop.
cat /proc/cmdline
ls /sys/devices/system/node/node0/hugepages/
ls /sys/devices/system/node/node1/hugepages/
This is a similar symptom to https:/ /bugs.launchpad .net/starlingx/ +bug/1796380 , which to me points to a memory issue, or lack thereof.
The /var/log/ openvswitch/ ovs-vswitchd. log around the time the puppet manifest is being applied should have some more information to confirm this. Are there any "ERR" messages?
It would also be useful to have the output of the following from the compute node. It might be useful to lock it first to get out of the reboot loop.
cat /proc/cmdline
ls /sys/devices/ system/ node/node0/ hugepages/ system/ node/node1/ hugepages/
ls /sys/devices/
cat /sys/devices/ system/ node/node0/ hugepages/ hugepages- 1048576kB/ nr_hugepages system/ node/node0/ hugepages/ hugepages- 1048576kB/ free_hugepages system/ node/node1/ hugepages/ hugepages- 1048576kB/ nr_hugepages system/ node/node1/ hugepages/ hugepages- 1048576kB/ free_hugepages
cat /sys/devices/
cat /sys/devices/
cat /sys/devices/
sudo ovs-vsctl get Open_vSwitch . other_config