[fluentd] Elasticsearch output enabled even when service disabled
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
kolla-ansible |
Fix Released
|
Medium
|
Scott Shambarger | ||
Ussuri |
Fix Released
|
Medium
|
Unassigned | ||
Victoria |
Fix Released
|
Medium
|
Unassigned | ||
Wallaby |
Fix Committed
|
Medium
|
Unassigned | ||
Xena |
Fix Released
|
Medium
|
Scott Shambarger |
Bug Description
Seen on v11, but appears in v12/master too...
Deploying all-in-one with the following in global.yml (elasticsearch disabled by default):
kolla_internal_
kolla_internal_
fluentd container contains td-agent.conf with elasticsearch output <store> entries, and fluentd repeatedly fails to connect to it, from fluentd.log:
#0 Could not communicate to Elasticsearch, resetting connection and trying again. Connection refused - connect(2) for 192.168.60.10:9200 (Errno:
THE FIX:
It appears that there's an error in ansible/
- name: Copying over td-agent.conf
vars:
log_
{{ ( enable_
( elasticsearch_
not enable_monasca | bool }}
log_direct_
elasticsearch_
Clearly a name is being compared to an IP. It appears the comparison should be:
elasticsear
... that, or elasticsearch_
Changed in kolla-ansible: | |
assignee: | nobody → Scott Shambarger (sshambar) |
Looks like a good bug with an easy solution, as proposed - compare with the FQDN. Can you propose a fix to gerrit?