Replies: 2 comments 1 reply
-
Hello, Sorry but there is no such option I'm afraid. |
Beta Was this translation helpful? Give feedback.
-
Events are first put in the internal ring buffer before they are rendered and delivered to the remote destination. This buffer can hold up to 8192 events by default but you can adjust its size using the The appender will automatically try to reconnect when the connection is lost. In the meantime, events are enqueued in the ring buffer until a connection is reestablished. What happens when the buffer is full depends on the RingBuffer Full configuration property. The default configuration is to drop new events until some space becomes available but you can also configure it to block if you don't want to loose anything. I would say that you should adapt the ring buffer size to match your logging rate and the expected/acceptable downtime of your remote logstash. Ideal you would also setup a second logstash instance to which the appender would connect when the first becomes unavailable (see Multiple Destinations for various strategies). Hope this helps. Note: the writeBufferSize controls the size of an intermediate byte buffer that sites in front on the socket output stream - it stores bytes instead of events. It is not meant to handle a temporary lost of connectivity... |
Beta Was this translation helpful? Give feedback.
-
Hi Team,
I am using LogstashTcpSocketAppender for sending logs to centralized log server. I wanted to configure backpressure and TCP connection failure scenarios. Like i can write the logs in filesystem in case TCP socket is not available. . Please let me know how to handle this.
Thanks in advance.
Aj
Beta Was this translation helpful? Give feedback.
All reactions