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
Is your feature request related to a problem? Please describe.
Problem: Something goes wrong and Fluent Bit is unable to send to your log destination. May be the destination goes down, or some network or other config prevents Fluent Bit from reaching it.
Proposed Solution: Specify a second output plugin as backup for another output. You'd be able to configure a timeout/threshold for when Fluent Bit switches to the backup output. (May be number of failed retries, number of errors, or something like the time since the first unsuccessful send).
Compared to other existing feature requests and open issues, I do not think this is very high priority. I'm interested in seeing if anyone else thinks this would be useful. Plus one if you agree and comment if you want to discuss how this would work!
The text was updated successfully, but these errors were encountered:
I feel like this is an important feature. A primary goal of Fluent Bit should be: Never loose data, and having backup output plugins would help achieve this goal.
I strongly agree. As many of the companies that uses fluent-bit will want to get all their data. This is important for statistics, analysis, anomaly detection, security monitoring and so on. So loosing a tiny bit of data, might sound insignificant, but can actually have major consequences.
Is your feature request related to a problem? Please describe.
Problem: Something goes wrong and Fluent Bit is unable to send to your log destination. May be the destination goes down, or some network or other config prevents Fluent Bit from reaching it.
Proposed Solution: Specify a second output plugin as backup for another output. You'd be able to configure a timeout/threshold for when Fluent Bit switches to the backup output. (May be number of failed retries, number of errors, or something like the time since the first unsuccessful send).
Compared to other existing feature requests and open issues, I do not think this is very high priority. I'm interested in seeing if anyone else thinks this would be useful. Plus one if you agree and comment if you want to discuss how this would work!
The text was updated successfully, but these errors were encountered: