From 9593981149aededef621ebda17962b1e02318328 Mon Sep 17 00:00:00 2001 From: Michele Baldessari Date: Sun, 25 Sep 2016 10:49:15 +0200 Subject: Start mongod before calling ceilometer-dbsync Currently we in major_upgrade_controller_pacemaker_2.sh we are calling ceilometer-dbsync before mongod is actually started (only galera is started at this point). This will make the dbsync hang indefinitely until the heat stack times out. Now this approach should be okay, but do note that when we start mongod via systemctl we are not guaranteed that it will be up on all nodes before we call ceilometer-dbsync. This *should* be okay because ceilometer-dbsync keeps retrying and eventually one of the nodes will be available. A completely clean fix here would be to add another step in heat to have the guarantee that all mongo servers are up and running before the dbsync call. Change-Id: I10c960b1e0efdeb1e55d77c25aebf1e3e67f17ca Closes-Bug: #1627453 --- extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh b/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh index 6bb2fa73..e91c7679 100755 --- a/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh +++ b/extraconfig/tasks/major_upgrade_controller_pacemaker_2.sh @@ -32,6 +32,13 @@ fi start_or_enable_service galera check_resource galera started 600 +# We need mongod which is now a systemd service up and running before calling +# ceilometer-dbsync. There is still a race here: mongod might not be up on all nodes +# so ceilometer-dbsync will fail a couple of times before that. As it retries indefinitely +# we should be good. +# Due to LP Bug https://bugs.launchpad.net/tripleo/+bug/1627254 am using systemctl directly atm +systemctl start mongod +check_resource mongod started 600 if [[ -n $(is_bootstrap_node) ]]; then tstart=$(date +%s) -- cgit 1.2.3-korg