You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Install an agent with the resulting policy and notice that everything works fine even though 1000 bytes is certainly too low for a functioning agent. Alternately, once #3781 is merged, set any value you like and observe that it is ignored and replaced with Agent's default of 104857600.
After digging some more I'm not sure this is exactly a "bug" so much as a confusingly documented missing feature. It looks like the intended way to set max_message_size is to edit the elastic-agent.yml for the installed Agent and change that parameter, and that it is intentionally ignored in Fleet policies.
This has confused some users as well as some Agent engineers 😄 We should probably either support this in policies or more clearly document that this parameter is local-only.
If we pick increasing max_message_size as the solution to #2460 then we should support it from Fleet to make it easy to change.
One other possible complication is that the max message size can apply to both the client and server sides and I think the existing parameter only affects the server side. Probably we need to provide this to clients as well.
The
agent.grpc.max_message_size
setting in a Fleet policy is ignored by the Agent.To reproduce, create a policy setting
max_message_size
:Install an agent with the resulting policy and notice that everything works fine even though 1000 bytes is certainly too low for a functioning agent. Alternately, once #3781 is merged, set any value you like and observe that it is ignored and replaced with Agent's default of
104857600
.grpc.max_message_size
in Fleet policy is ignored #3782The text was updated successfully, but these errors were encountered: