Description of problem: Installation for aws m5/c5 server is failed because of trying to insert 'xen_netfront' Version-Release number of the following components: openshift v3.11.0-0.28.0 openshift-ansible-3.11.0-0.28.0.git.0.730d4be.el7.noarch.rpm How reproducible: Always Steps to Reproduce: 1 set below parameters vm_type: m5.large image: qe-rhel-7-release Actual results: TASK [openshift_storage_glusterfs : load kernel modules] *********************** Thursday 06 September 2018 11:11:10 +0800 (0:00:00.135) 0:02:29.580 **** fatal: [ec2-18-232-57-9.compute-1.amazonaws.com]: FAILED! => {"changed": false, "msg": "Unable to start service systemd-modules-load.service: Job for systemd-modules-load.service failed because the control process exited with error code. See \"systemctl status systemd-modules-load.service\" and \"journalctl -xe\" for details.\n"} fatal: [ec2-52-90-226-187.compute-1.amazonaws.com]: FAILED! => {"changed": false, "msg": "Unable to start service systemd-modules-load.service: Job for systemd-modules-load.service failed because the control process exited with error code. See \"systemctl status systemd-modules-load.service\" and \"journalctl -xe\" for details.\n"} fatal: [ec2-34-229-212-114.compute-1.amazonaws.com]: FAILED! => {"changed": false, "msg": "Unable to start service systemd-modules-load.service: Job for systemd-modules-load.service failed because the control process exited with error code. See \"systemctl status systemd-modules-load.service\" and \"journalctl -xe\" for details.\n"} Expected results: Installation should be successfull. Additional info: [root@ip-172-18-5-48 ~]# systemctl status systemd-modules-load.service ● systemd-modules-load.service - Load Kernel Modules Loaded: loaded (/usr/lib/systemd/system/systemd-modules-load.service; static; vendor preset: disabled) Active: failed (Result: exit-code) since 三 2018-09-05 23:11:10 EDT; 15min ago Docs: man:systemd-modules-load.service(8) man:modules-load.d(5) Process: 7270 ExecStart=/usr/lib/systemd/systemd-modules-load (code=exited, status=1/FAILURE) Main PID: 7270 (code=exited, status=1/FAILURE) 9月 05 23:11:10 ip-172-18-5-48.ec2.internal systemd[1]: Starting Load Kernel Modules... 9月 05 23:11:10 ip-172-18-5-48.ec2.internal systemd-modules-load[7270]: Inserted module 'dm_thin_pool' 9月 05 23:11:10 ip-172-18-5-48.ec2.internal systemd-modules-load[7270]: Inserted module 'target_core_user' 9月 05 23:11:10 ip-172-18-5-48.ec2.internal systemd-modules-load[7270]: Failed to insert 'xen_netfront': No such device 9月 05 23:11:10 ip-172-18-5-48.ec2.internal systemd[1]: systemd-modules-load.service: main process exited, code=exited, status=1/FAILURE 9月 05 23:11:10 ip-172-18-5-48.ec2.internal systemd[1]: Failed to start Load Kernel Modules. 9月 05 23:11:10 ip-172-18-5-48.ec2.internal systemd[1]: Unit systemd-modules-load.service entered failed state. 9月 05 23:11:10 ip-172-18-5-48.ec2.internal systemd[1]: systemd-modules-load.service failed.
All openshift-ansible is doing is writing out /etc/modules-load.d/glusterfs.conf with the contents of the following dm_thin_pool dm_multipath target_core_user {% if (groups.glusterfs is defined and inventory_hostname in groups.glusterfs) or (groups.glusterfs_registry is defined and inventory_hostname in groups.glusterfs_registry) %} dm_snapshot dm_mirror Then we restart `systemd-modules-load.service` when that file changes. Was anything else done on the host to load the xen_netfront module? Can you gather the contents of /etc/modules-load.d ?
What AMI are you using on this host type?
(In reply to Scott Dodson from comment #1) > All openshift-ansible is doing is writing out > /etc/modules-load.d/glusterfs.conf > with the contents of the following > > dm_thin_pool > dm_multipath > target_core_user > {% if (groups.glusterfs is defined and inventory_hostname in > groups.glusterfs) or (groups.glusterfs_registry is defined and > inventory_hostname in groups.glusterfs_registry) %} > dm_snapshot > dm_mirror > > Then we restart `systemd-modules-load.service` when that file changes. > > Was anything else done on the host to load the xen_netfront module? Can you > gather the contents of /etc/modules-load.d ? Just FYI, m5/c5 instances are VMs based on KVM Hypervisor, so don't need to load xen_netfront module any more. Please refer to this bug 1497392 for more information. And m5/c5 instance requires ENA driver in kernel, please make sure it's integrated and the ena support are enabled with AMI or instance. More information about EC2 instances please refer to or contact my team: https://docs.engineering.redhat.com/pages/viewpage.action?spaceKey=XENQE&title=RHEL+on+AWS+EC2
Yes, c5/m5 and upcoming new instances are not Xen based, two required modules are ena and nvme.
Vitaly, Lin, So, the solution here is "Use a more recent AMI" ?
(In reply to Scott Dodson from comment #7) > Vitaly, Lin, > > So, the solution here is "Use a more recent AMI" ? Yes, I just double checked RHEL7.5 ami-680d9010 (us-west-2) and there is no xen_netfront in /etc/modprobe.d (it was there before BZ#1497392).
Vitaly, Thanks! CLOSED NOTABUG the AMI in use is not compatible with the machine time specified. Newer AMI versions have been updated to work appropriately on the new machine times please update the AMI.
Another workaround consists on disabling the module from being loaded as is not present on the RHEL 7.4 AMI because is not xen anymore (mentioned on previous comments). Steps to solve the issue: - echo "blacklist xen_netfront" >> /etc/modprobe.d/local-blacklist.conf - echo "install xen_netfront /bin/false" >> /etc/modprobe.d/local-blacklist.conf More info: https://access.redhat.com/solutions/41278