openstack
/
puppet-tripleo
Code Issues Proposed changes

Allow tuning timeouts for rabbitmq pacemaker resource 

This change introduces several timeout parameters so that users can
tune operation timeouts about rabbimtq resource in pacemaker.

Change-Id: Iaecdc0adb8455b2e660624f19a42e6dede5b931d
This commit is contained in:
Takashi Kajinami 2021-06-08 20:22:28 +09:00
parent 63a9008914
commit 26ee01a0d9
2 changed files with 68 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
manifests/profile/pacemaker/rabbitmq.pp
View File

@ -63,6 +63,21 @@
# for more details.
# Defaults to hiera('step')
#
# [*start_timeout*]
# (Optional) Maximum time in second for starting up a rabbitmq server
# before pacemaker considers the operation timed out.
# Defaults to 200
#
# [*monitor_timeout*]
# (Optional) Maximum time in second for monitoring a rabbitmq server
# before pacemaker considers the operation timed out.
# Defaults to undef (use the default value in resource agent)
#
# [*stop_timeout*]
# (Optional) Maximum time in second for stopping a rabbitmq server
# before pacemaker considers the operation timed out.
# Defaults to 200
#
class tripleo::profile::pacemaker::rabbitmq (
$erlang_cookie = hiera('rabbitmq::erlang_cookie'),
$user_ha_queues = hiera('rabbitmq::nr_ha_queues', 0),
@ -74,6 +89,9 @@ class tripleo::profile::pacemaker::rabbitmq (
$notify_nodes = hiera('oslo_messaging_notify_node_names', []),
$pcs_tries = hiera('pcs_tries', 20),
$step = Integer(hiera('step')),
$start_timeout = 200,
$monitor_timeout = undef,
$stop_timeout = 200,
) {
if $rpc_scheme == 'rabbit' {
$bootstrap_node = $rpc_bootstrap_node
@ -130,12 +148,27 @@ class tripleo::profile::pacemaker::rabbitmq (
$nr_ha_queues = $user_ha_queues
$params = "set_policy='ha-all ^(?!amq\\.).* {\"ha-mode\":\"exactly\",\"ha-params\":${nr_ha_queues}}'"
}
$op_start_params = $start_timeout ? {
undef => undef,
default => "start timeout=${start_timeout}s"
}
$op_monitor_params = $monitor_timeout ? {
undef => undef,
default => "monitor timeout=${monitor_timeout}s"
}
$op_stop_params = $stop_timeout ? {
undef => undef,
default => "stop timeout=${stop_timeout}s"
}
$op_params = join(delete_undef_values([$op_start_params, $op_monitor_params, $op_stop_params]), ' ')
pacemaker::resource::ocf { 'rabbitmq':
ocf_agent_name => 'heartbeat:rabbitmq-cluster',
resource_params => $params,
clone_params => 'ordered=true interleave=true',
meta_params => 'notify=true',
op_params => 'start timeout=200s stop timeout=200s',
op_params => $op_params,
tries => $pcs_tries,
location_rule => {
resource_discovery => 'exclusive',

35
manifests/profile/pacemaker/rabbitmq_bundle.pp
View File

@ -104,6 +104,22 @@
# [*force_ocf*]
# (optional) Use --force when creating the ocf resource via pcs
# Defaults to false
#
# [*start_timeout*]
# (Optional) Maximum time in second for starting up a rabbitmq server
# before pacemaker considers the operation timed out.
# Defaults to 200
#
# [*monitor_timeout*]
# (Optional) Maximum time in second for monitoring a rabbitmq server
# before pacemaker considers the operation timed out.
# Defaults to undef (use the default value in resource agent)
#
# [*stop_timeout*]
# (Optional) Maximum time in second for stopping a rabbitmq server
# before pacemaker considers the operation timed out.
# Defaults to 200
#
class tripleo::profile::pacemaker::rabbitmq_bundle (
$rabbitmq_docker_image = undef,
$rabbitmq_docker_control_port = 3122,
@ -127,6 +143,9 @@ class tripleo::profile::pacemaker::rabbitmq_bundle (
$tls_priorities = hiera('tripleo::pacemaker::tls_priorities', undef),
$bundle_user = 'root',
$force_ocf = false,
$start_timeout = 200,
$monitor_timeout = undef,
$stop_timeout = 200,
) {
# is this an additional nova cell?
if hiera('nova_is_additional_cell', undef) {
@ -327,11 +346,25 @@ class tripleo::profile::pacemaker::rabbitmq_bundle (
$ha_policy = merge($ha_queues_policy, $rabbitmq_extra_policies)
$ocf_params = "set_policy='ha-all ^(?!amq\\.).* ${to_json($ha_policy)}'"
$op_start_params = $start_timeout ? {
undef => undef,
default => "start timeout=${start_timeout}s"
}
$op_monitor_params = $monitor_timeout ? {
undef => undef,
default => "monitor timeout=${monitor_timeout}s"
}
$op_stop_params = $stop_timeout ? {
undef => undef,
default => "stop timeout=${stop_timeout}s"
}
$op_params = join(delete_undef_values([$op_start_params, $op_monitor_params, $op_stop_params]), ' ')
pacemaker::resource::ocf { 'rabbitmq':
ocf_agent_name => 'heartbeat:rabbitmq-cluster',
resource_params => $ocf_params,
meta_params => 'notify=true container-attribute-target=host',
op_params => 'start timeout=200s stop timeout=200s',
op_params => $op_params,
tries => $pcs_tries,
location_rule => {
resource_discovery => 'exclusive',