mirror of
https://github.com/xcat2/confluent.git
synced 2024-11-25 11:01:09 +00:00
Halt autonomous discovery handling while configmanager is down
This avoids triggering a potential large amount of churn on transiently "unknown" systems that are actually discovered.
This commit is contained in:
parent
94b8559777
commit
d4c535d038
@ -648,6 +648,8 @@ def detected_models():
|
||||
|
||||
|
||||
def _recheck_nodes(nodeattribs, configmanager):
|
||||
if not cfm.config_is_ready():
|
||||
return
|
||||
if rechecklock.locked():
|
||||
# if already in progress, don't run again
|
||||
# it may make sense to schedule a repeat, but will try the easier and less redundant way first
|
||||
@ -766,6 +768,9 @@ def eval_detected(info):
|
||||
def detected(info):
|
||||
global rechecker
|
||||
global rechecktime
|
||||
if not cfm.config_is_ready():
|
||||
# drop processing of discovery data while configmanager is 'down'
|
||||
return
|
||||
# later, manual and CMM discovery may act on SN and/or UUID
|
||||
for service in info['services']:
|
||||
if service in nodehandlers:
|
||||
|
Loading…
Reference in New Issue
Block a user