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
It happens that currently the agent has only two different mechanisms to avoid missing logs with logcollector
The only_future_events option to keep reading files since the last timestamp when the agent was running
The leaky bucket, that holds in memory the events until the connection is re-established
The second in particular has some drawbacks, for example:
A restart of the agent during the disconnection period causes the loss of all events in memory
The disconnection period could be long, and the buffer insufficient
That's why @protocolpaladin proposed a configurable log retention threshold with high capacity (GBs) to store the logs until they can be sent to the manager.
DoD
The corresponding team analyzes the feature request
The feature is included in the backlog or rejected
The text was updated successfully, but these errors were encountered:
Description
An user has made a feature request at #23377.
It happens that currently the agent has only two different mechanisms to avoid missing logs with logcollector
only_future_events
option to keep reading files since the last timestamp when the agent was runningThe second in particular has some drawbacks, for example:
That's why @protocolpaladin proposed a configurable log retention threshold with high capacity (GBs) to store the logs until they can be sent to the manager.
DoD
The text was updated successfully, but these errors were encountered: