Background:
Our a puppet server manages lots of agent nodes, it often failed to 
response in time due to too many requests squeeze in some part of period. 
We have set 'splay=true' but it seem to do no help. We hope to more 
pricisely control each node's agent request timepoint, that is, to make 
them evenly distribute across the run-interval period. But 'systemctl 
restart puppet.service' does not always trigger the puppet agent request 
immediately.

So the question are:
1. how do background puppet agent service determine its next request 
timepoint?
2. suppose 'runinterval=1h', with splay feature disabled, will all the 
future request time point be fixed exactly at x+1h, x+2h, x+3h, etc... ?
3. at puppet server node, is there any way to list which agent nodes are 
currently connecting or occupying server's instance, and which nodes are in 
queue?

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/puppet-users/59fd2837-9347-4b21-a2d5-79d9af274bb0n%40googlegroups.com.

Reply via email to