Make minimum default num_engine_workers>=4
Downstream test environments are frequently having failing stacks with error messages like: MessagingTimeout: resources[0]: Timed out waiting for a reply to message ID ... These environments generally have 1 or 2 cores, so only spawn one or two engine workers. This deadlocks with stacks that have many nested stacks due to engine->engine RPC calls. Even our own functional tests don't work reliably with less than 4 workers, and the workaround has been to set that explicitly in pre_test_hook.sh. This change sets the default minimum number of workers to 4, but still matches workers to cores for larger servers. This change also moves the default evaluation to heat.cmd.engine so that generated configuration doesn't get a inappropriate default value. Change-Id: Iae6b3956bad414406d901bb2213c9ec230ff4304 Closes-Bug: #1526045
This commit is contained in:
parent
e86b91b05f
commit
adb2121795
@ -24,6 +24,7 @@ eventlet.monkey_patch()
|
|||||||
|
|
||||||
import sys
|
import sys
|
||||||
|
|
||||||
|
from oslo_concurrency import processutils
|
||||||
from oslo_config import cfg
|
from oslo_config import cfg
|
||||||
import oslo_i18n as i18n
|
import oslo_i18n as i18n
|
||||||
from oslo_log import log as logging
|
from oslo_log import log as logging
|
||||||
@ -66,8 +67,11 @@ def main():
|
|||||||
profiler.setup('heat-engine', cfg.CONF.host)
|
profiler.setup('heat-engine', cfg.CONF.host)
|
||||||
gmr.TextGuruMeditation.setup_autorun(version)
|
gmr.TextGuruMeditation.setup_autorun(version)
|
||||||
srv = engine.EngineService(cfg.CONF.host, rpc_api.ENGINE_TOPIC)
|
srv = engine.EngineService(cfg.CONF.host, rpc_api.ENGINE_TOPIC)
|
||||||
launcher = service.launch(cfg.CONF, srv,
|
workers = cfg.CONF.num_engine_workers
|
||||||
workers=cfg.CONF.num_engine_workers)
|
if not workers:
|
||||||
|
workers = max(4, processutils.get_worker_count())
|
||||||
|
|
||||||
|
launcher = service.launch(cfg.CONF, srv, workers=workers)
|
||||||
if cfg.CONF.enable_cloud_watch_lite:
|
if cfg.CONF.enable_cloud_watch_lite:
|
||||||
# We create the periodic tasks here, which mean they are created
|
# We create the periodic tasks here, which mean they are created
|
||||||
# only in the parent process when num_engine_workers>1 is specified
|
# only in the parent process when num_engine_workers>1 is specified
|
||||||
|
@ -15,7 +15,6 @@
|
|||||||
import os
|
import os
|
||||||
|
|
||||||
from eventlet.green import socket
|
from eventlet.green import socket
|
||||||
from oslo_concurrency import processutils
|
|
||||||
from oslo_config import cfg
|
from oslo_config import cfg
|
||||||
from oslo_log import log as logging
|
from oslo_log import log as logging
|
||||||
|
|
||||||
@ -81,7 +80,6 @@ service_opts = [
|
|||||||
default=5,
|
default=5,
|
||||||
help=_('Maximum depth allowed when using nested stacks.')),
|
help=_('Maximum depth allowed when using nested stacks.')),
|
||||||
cfg.IntOpt('num_engine_workers',
|
cfg.IntOpt('num_engine_workers',
|
||||||
default=processutils.get_worker_count(),
|
|
||||||
help=_('Number of heat-engine processes to fork and run.'))]
|
help=_('Number of heat-engine processes to fork and run.'))]
|
||||||
|
|
||||||
engine_opts = [
|
engine_opts = [
|
||||||
|
@ -27,7 +27,6 @@ if [ "$ENABLE_CONVERGENCE" == "true" ] ; then
|
|||||||
fi
|
fi
|
||||||
|
|
||||||
echo -e 'notification_driver=messagingv2\n' >> $localconf
|
echo -e 'notification_driver=messagingv2\n' >> $localconf
|
||||||
echo -e 'num_engine_workers=4\n' >> $localconf
|
|
||||||
echo -e 'hidden_stack_tags=hidden\n' >> $localconf
|
echo -e 'hidden_stack_tags=hidden\n' >> $localconf
|
||||||
echo -e 'encrypt_parameters_and_properties=True\n' >> $localconf
|
echo -e 'encrypt_parameters_and_properties=True\n' >> $localconf
|
||||||
|
|
||||||
|
Loading…
Reference in New Issue
Block a user