X-Git-Url: https://gerrit.opnfv.org/gerrit/gitweb?a=blobdiff_plain;f=mcp%2Fconfig%2Fstates%2Fmaas;h=f472c015427affe3d0a9499a615f98ab5788d348;hb=refs%2Fchanges%2F37%2F49237%2F3;hp=01f8ee0cd8bf5bef8d94758f2696815b78cfdb98;hpb=08bb570b39b123c5a5adf70e34e05bb43a6e3387;p=fuel.git diff --git a/mcp/config/states/maas b/mcp/config/states/maas index 01f8ee0cd..f472c0154 100755 --- a/mcp/config/states/maas +++ b/mcp/config/states/maas @@ -1,29 +1,34 @@ -#!/bin/bash -set -x - -function wait_for() { - local total_attempts=$1; shift - local cmdstr=$@ - local sleep_time=10 - echo "[NOTE] Waiting for cmd to return success: ${cmdstr}" - for attempt in $(seq "${total_attempts}"); do - eval "${cmdstr}" && break || true - echo -n '.'; sleep "${sleep_time}" - done -} +#!/bin/bash -e +# shellcheck disable=SC1090,SC2155 +############################################################################## +# Copyright (c) 2017 Mirantis Inc., Enea AB and others. +# All rights reserved. This program and the accompanying materials +# are made available under the terms of the Apache License, Version 2.0 +# which accompanies this distribution, and is available at +# http://www.apache.org/licenses/LICENSE-2.0 +############################################################################## + +CI_DEBUG=${CI_DEBUG:-0}; [[ "${CI_DEBUG}" =~ (false|0) ]] || set -x +ERASE_ENV=${ERASE_ENV:-0} + +source "$(dirname "${BASH_SOURCE[0]}")/../../scripts/globals.sh" +source "$(dirname "${BASH_SOURCE[0]}")/../../scripts/lib.sh" # Wait for MaaS commissioning/deploy to finish, retry on failure function maas_fixup() { local statuscmd="salt 'mas01*' --out yaml state.apply maas.machines.status" - wait_for 180 "${statuscmd} | tee /dev/stderr | " \ - "grep -Eq '((Deployed|Ready): 5|status:Failed|status:Allocated)'" + local ncount=$(salt --out yaml 'mas01*' pillar.get maas:region:machines | \ + grep -cE '^\s{2}\w+:$') + wait_for 30 "${statuscmd} | tee /dev/stderr | " \ + "grep -Eq '((Deployed|Ready): ${ncount}|status: (Failed|Allocated))'" local statusout=$(eval "${statuscmd}") local fcnodes=$(echo "${statusout}" | \ - grep -Po '(?<=system_id:)(.*)(?=,status:Failed commissioning)') + grep -Pzo 'status: Failed commissioning\n\s+system_id: \K.+\n') for node_system_id in ${fcnodes}; do salt -C 'mas01*' state.apply maas.machines.delete \ pillar="{'system_id': '${node_system_id}'}" + sleep 10 done if [ -n "${fcnodes}" ]; then salt -C 'mas01*' state.apply maas.machines @@ -31,12 +36,15 @@ function maas_fixup() { fi local fdnodes=$(echo "${statusout}" | \ - grep -Po '(?<=system_id:)(.*)(?=,status:(Failed deployment|Allocated))') + grep -Pzo 'status: (Failed deployment|Allocated)\n\s+system_id: \K.+\n') + local rnodes=$(echo "${statusout}" | \ + grep -Pzo 'status: Ready\n\s+system_id: \K.+\n') for node_system_id in ${fdnodes}; do salt -C 'mas01*' state.apply maas.machines.mark_broken_fixed \ pillar="{'system_id': '${node_system_id}'}" + sleep 10 done - if [ -n "${fdnodes}" ]; then + if [ -n "${fdnodes}" ] || [ -n "${rnodes}" ]; then salt -C 'mas01*' state.apply maas.machines.deploy return 1 fi @@ -44,65 +52,50 @@ function maas_fixup() { return 0 } -# MaaS rack/region controller, node commissioning -salt -C 'mas01*' cmd.run "add-apt-repository ppa:maas/stable" +# Optionally destroy MaaS machines from a previous run +if [ "${ERASE_ENV}" -gt 1 ]; then + set +e; dnodes=$(salt 'mas01*' --out yaml state.apply maas.machines.status | \ + grep -Pzo '\s+system_id: \K.+\n'); set -e + cleanup_uefi + for node_system_id in ${dnodes}; do + salt -C 'mas01*' state.apply maas.machines.delete \ + pillar="{'system_id': '${node_system_id}'}" + sleep 10 + done +fi +# MaaS rack/region controller, node commissioning salt -C 'mas01*' state.apply linux,salt,openssh,ntp -salt -C 'mas01*' state.apply linux.network.interface salt -C 'mas01*' state.apply maas.pxe_nat salt -C 'mas01*' state.apply maas.cluster -salt -C 'cfg01*' cmd.run \ - "route add -net 192.168.11.0/24 gw ${MAAS_IP:-192.168.10.3}" wait_for 10 "salt -C 'mas01*' state.apply maas.region" salt -C 'mas01*' state.apply maas.machines -wait_for 10 maas_fixup +# MaaS node deployment +wait_for 20 maas_fixup # cleanup outdated salt keys salt-key --out yaml | awk '!/^(minions|- cfg01|- mas01)/ {print $2}' | \ xargs -I{} salt-key -yd {} -# MaaS node deployment -salt -C 'mas01*' state.apply maas.machines.deploy -wait_for 10 maas_fixup - salt -C 'mas01*' pillar.item\ maas:region:admin:username \ maas:region:admin:password -# KVM, compute node prereqs (libvirt first), VCP deployment -salt -C '* and not cfg01* and not mas01*' saltutil.sync_all - -salt -C 'kvm*' pkg.install bridge-utils -salt -C 'kvm*' state.apply linux.network -salt -C 'kvm*' system.reboot -wait_for 90 "! salt 'kvm*' test.ping | tee /dev/stderr | fgrep -q 'Not connected'" - -salt -C '* and not cfg01* and not mas01*' state.apply linux,ntp - -salt -C 'kvm*' state.sls libvirt - -salt -C '* and not cfg01* and not mas01*' state.apply salt -salt -C 'kvm*' saltutil.sync_all -salt -C 'kvm*' state.sls salt.control - -vcp_nodes=$(salt --out yaml 'kvm01*' pillar.get salt:control:cluster:internal:node | \ - awk '/\s+\w+:$/ {gsub(/:$/, "*"); print $1}') - -# Check all vcp nodes are available +# Check all baremetal nodes are available rc=1 -while [ $rc -ne 0 ]; do +attempt=0 +total_attempts=10 +while [ $rc -ne 0 ] && [ ${attempt} -lt ${total_attempts} ]; do + bm_nodes=$(salt --out yaml 'mas01*' pillar.get maas:region:machines | \ + awk '/^\s+\w+[[:digit:]]+:$/ {gsub(/:$/, "*"); print $1}') rc=0 - for node in $vcp_nodes; do + for node in $bm_nodes; do salt "$node" test.ping 2>/dev/null || { rc=$?; break; }; done sleep 5 + ((attempt+=1)) done wait_for 10 "salt -C '* and not cfg01* and not mas01*' saltutil.sync_all" -wait_for 10 "salt -C 'E@^(?!cfg01|mas01|kvm|cmp00).*' state.apply salt" -wait_for 10 "salt -C 'E@^(?!cfg01|mas01|kvm|cmp00).*' state.apply linux,ntp" - -wait_for 10 "salt -C 'E@^(?!cfg01|mas01|kvm|cmp00).*' ssh.set_auth_key ${SUDO_USER} \ - $(awk 'NR==1{print $2}' $(eval echo ~${SUDO_USER}/.ssh/authorized_keys))"