openstack/openstack-ansible-os_octavia
Code Issues Proposed changes

Update systemd service for longer TimeoutStopSec

Browse Source 
This patch updates the default systemd service definition for the Octavia
processes. It increases the TimeoutStopSec to 300 seconds to allow more time
for the in-flight flows (failover, etc.) to finish and gracefully shutdown.
If the cloud has performance issues, flows may run beyond their normal
sub-minute runtime.

Change-Id: I5708c05db2c9f13491f5af0cd0b8c5632c537608
This commit is contained in:
Michael Johnson 2018-03-20 16:51:23 -07:00
parent e37b81cb05
commit b1d02c6ba9
1 changed files with 6 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
templates/octavia-systemd-init.j2
View File

@ -17,7 +17,12 @@ ExecStart={{ octavia_bin }}/{{ item.value.service_name }} {{ item.value.program_
{% endif %} {% endif %}
# Give a reasonable amount of time for the server to start up/shut down # Give a reasonable amount of time for the server to start up/shut down
TimeoutSec=120 TimeoutStartSec=120
# Give extra time for shutdown to allow flows to finish (failover, etc.)
# This timer expiring early can lead to load balancers stuck in PENDING_*
# states as the flows were interrupted before completing.
# This setting depends on the performance of your cloud.
TimeoutStopSec=300
Restart=on-failure Restart=on-failure
RestartSec=2 RestartSec=2