openstack/devstack
Code Issues Proposed changes

Set [scheduler]workers=$API_WORKERS

Browse Source 
Since blueprint placement-claims in Pike, the Nova FilterScheduler
uses the placement service to make resource allocation 'claims'
before sending the build request to the chosen compute host to
perform the legacy style resource claim. This allows us to safely
scale out the number of scheduler workers when using the FilterScheduler.

The [scheduler]workers option defaults to ncpu if using the
FilterScheduler (which is the default scheduler driver) so to avoid
out of memory issues, we need to set $API_WORKERS scheduler workers
if using the FilterScheduler in devstack.

Depends-On: Ifdcd363d7bc22e73d76d69777483e5aaff4036e3
Change-Id: Ieae234eb5388560b3f66bf60c156a91a8e831bc4
This commit is contained in:
Matt Riedemann
2018-02-18 17:14:48 -05:00
parent 22d70554ac
commit a3844240ab
1 changed files with 3 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
lib/nova
View File

@@ -413,6 +413,9 @@ function create_nova_conf {
iniset $NOVA_CONF DEFAULT rootwrap_config "$NOVA_CONF_DIR/rootwrap.conf"
iniset $NOVA_CONF scheduler driver "$SCHEDULER"
iniset $NOVA_CONF filter_scheduler enabled_filters "$FILTERS"
if [[ $SCHEDULER == "filter_scheduler" ]]; then
iniset $NOVA_CONF scheduler workers "$API_WORKERS"
fi
iniset $NOVA_CONF DEFAULT default_floating_pool "$PUBLIC_NETWORK_NAME"
if [[ $SERVICE_IP_VERSION == 6 ]]; then
iniset $NOVA_CONF DEFAULT my_ip "$HOST_IPV6"