When running the deploy-hosts playbook the playbook fails on TASK [lxc_container_create : Execute first script]
When looking at the log it appears that the Repository rdo-qemu-ev is listed more than once in the configuration after running $RHT_PKG_MGR check-update.
I have followed the rebuild guide listed under "Rebuilding an AIO" on this page https://docs.openstack.org/openstack-ansible/rocky/user/aio/quickstart.html but that has had no effect
Here is the log from one of the containers, all containers have the same error
fatal: [aio1_utility_container-301e2034 -> 172.29.236.100]: FAILED! => {"attempts": 5, "changed": true, "cmd": ["/var/lib/lxc/aio1_utility_container-301e2034/container-first-run.sh"], "delta": "0:00:03.119944", "end": "2019-03-23 12:12:01.606533", "msg": "non-zero return code", "rc": 100, "start": "2019-03-23 12:11:58.486589", "stderr": "\nexport CPID=\"$(lxc-info -Hpn aio1_utility_container-301e2034)\"\n\nfunction update_backup_resolvers {\n if [[ -e \"/proc/${CPID}/root/etc/resolv.conf.setup\" ]]; then\n mv /proc/${CPID}/root/etc/resolv.conf.setup /proc/${CPID}/root/etc/resolv.conf\n elif [[ -e \"/proc/${CPID}/root/etc/resolv.conf\" ]]; then\n cp /proc/${CPID}/root/etc/resolv.conf /proc/${CPID}/root/etc/resolv.conf.setup\n fi\n}\n\n# If the setup is complete return as the process is complete\nif [[ -f \"/var/lib/lxc/aio1_utility_container-301e2034/setup.complete\" ]]; then\n exit 0\nfi\n\n# Backup the resolvers as the setup script is not executed within an NS.\nupdate_backup_resolvers\n\n# Use the host resolvers\ncat /etc/resolv.conf > /proc/${CPID}/root/etc/resolv.conf\n\n# Execute\nnsenter --mount=/proc/${CPID}/ns/mnt --pid=/proc/${CPID}/ns/pid \\\n --uts=/proc/${CPID}/ns/uts --ipc=/proc/${CPID}/ns/ipc \\\n -- /opt/container-setup.sh\n\n# Generate the lxc container create prep commands\n\n\n# Create dir \"/var/backup\"\nmkdir -p \"/var/backup\"\n\n\n# Run extra commands\nwhich dnf &>/dev/null && RHT_PKG_MGR='dnf' || RHT_PKG_MGR='yum'\ntimeout 90 $RHT_PKG_MGR check-update\nFailed to set locale, defaulting to C\nRepository rdo-qemu-ev is listed more than once in the configuration", "stderr_lines": ["", "export CPID=\"$(lxc-info -Hpn aio1_utility_container-301e2034)\"", "", "function update_backup_resolvers {", " if [[ -e \"/proc/${CPID}/root/etc/resolv.conf.setup\" ]]; then", " mv /proc/${CPID}/root/etc/resolv.conf.setup /proc/${CPID}/root/etc/resolv.conf", " elif [[ -e \"/proc/${CPID}/root/etc/resolv.conf\" ]]; then", " cp /proc/${CPID}/root/etc/resolv.conf /proc/${CPID}/root/etc/resolv.conf.setup", " fi", "}", "", "# If the setup is complete return as the process is complete", "if [[ -f \"/var/lib/lxc/aio1_utility_container-301e2034/setup.complete\" ]]; then", " exit 0", "fi", "", "# Backup the resolvers as the setup script is not executed within an NS.", "update_backup_resolvers", "", "# Use the host resolvers", "cat /etc/resolv.conf > /proc/${CPID}/root/etc/resolv.conf", "", "# Execute", "nsenter --mount=/proc/${CPID}/ns/mnt --pid=/proc/${CPID}/ns/pid \\", " --uts=/proc/${CPID}/ns/uts --ipc=/proc/${CPID}/ns/ipc \\", " -- /opt/container-setup.sh", "", "# Generate the lxc container create prep commands", "", "", "# Create dir \"/var/backup\"", "mkdir -p \"/var/backup\"", "", "", "# Run extra commands", "which dnf &>/dev/null && RHT_PKG_MGR='dnf' || RHT_PKG_MGR='yum'", "timeout 90 $RHT_PKG_MGR check-update", "Failed to set locale, defaulting to C", "Repository rdo-qemu-ev is listed more than once in the configuration"], "stdout": "Loaded plugins: fastestmirror, ovl, priorities, rpm-warm-cache\nLoading mirror speeds from cached hostfile\n * base: mirror.econdc.com\n * updates: mirrors.clouvider.net\n\ncpp.x86_64 4.8.5-36.el7_6.1 updates\ndbus.x86_64 1:1.10.24-13.el7_6 updates\ndbus-libs.x86_64 1:1.10.24-13.el7_6 updates\ngcc.x86_64 4.8.5-36.el7_6.1 updates\nlibblkid.x86_64 2.23.2-59.el7_6.1 updates\nlibgcc.x86_64 4.8.5-36.el7_6.1 updates\nlibgomp.x86_64 4.8.5-36.el7_6.1 updates\nlibmount.x86_64 2.23.2-59.el7_6.1 updates\nlibsmartcols.x86_64 2.23.2-59.el7_6.1 updates\nlibstdc++.x86_64 4.8.5-36.el7_6.1 updates\nlibuuid.x86_64 2.23.2-59.el7_6.1 updates\nnss-pem.x86_64 1.0.3-5.el7_6.1 updates\nshadow-utils.x86_64 2:4.1.5.1-25.el7_6.1 updates\nutil-linux.x86_64 2.23.2-59.el7_6.1 updates", "stdout_lines": ["Loaded plugins: fastestmirror, ovl, priorities, rpm-warm-cache", "Loading mirror speeds from cached hostfile", " * base: mirror.econdc.com", " * updates: mirrors.clouvider.net", "", "cpp.x86_64 4.8.5-36.el7_6.1 updates", "dbus.x86_64 1:1.10.24-13.el7_6 updates", "dbus-libs.x86_64 1:1.10.24-13.el7_6 updates", "gcc.x86_64 4.8.5-36.el7_6.1 updates", "libblkid.x86_64 2.23.2-59.el7_6.1 updates", "libgcc.x86_64 4.8.5-36.el7_6.1 updates", "libgomp.x86_64 4.8.5-36.el7_6.1 updates", "libmount.x86_64 2.23.2-59.el7_6.1 updates", "libsmartcols.x86_64 2.23.2-59.el7_6.1 updates", "libstdc++.x86_64 4.8.5-36.el7_6.1 updates", "libuuid.x86_64 2.23.2-59.el7_6.1 updates", "nss-pem.x86_64 1.0.3-5.el7_6.1 updates", "shadow-utils.x86_64 2:4.1.5.1-25.el7_6.1 updates", "util-linux.x86_64 2.23.2-59.el7_6.1 updates"]}
I am also getting the same problem on CentOS 7
Trying to deploy OSA with lxc containers.
The strange thing is that it only happened after a few restarts of the deployment process, whereas this did not occur on the first tries. Maybe the first tries did not get far enough to reach the bug though.