I am working on a proof of concept. The situation is I am logging lots of data quite quickly from a short lived java app.
I have one graylog-server on a small AWS system. I have one elasticsearch server on a small AWS system. I have one graylog-web-server on a small AWS system. The AWS stats do not indicate the CPU, memory, disk or network are overly taxed. I have a test program that uses the Java gelfclient, using TCP. I have tried two tests (run each run several times). Test 1: Sends 10K log messages which are small (about 400-500 bytes). Test 2: Sends 100 messages which larger (20-25K). Test1 Results: 1. Sometimes there are a few (0-400) messages that "fail" meaning the gelclient returns a false when doing a trySend. 2. Waiting 5 seconds at end of the test gets about 4K of the 10K messages. Always the first 4K. Last 6K are all lost. 3. Waiting 10 seconds at end of the test gets about 8K of the 10K messages. Test2 Results: 1. No messages "fail" to send (trySend always returns true) 2. Waiting 5 seconds gets all 100 messages. 3. Not waiting after sending messages gets 8 of the 100 messages. My conclusion is that it is taking a while for the queue in the gelfclient to empty. I have tried to identify when the queue is empty but have not been successful. My application is to gather logs from short lived Java processes, so I need some way to knowing when the queue is empty and I can shut down the app. Any help is much appreciated. Thanks, Derrick Shadel -- You received this message because you are subscribed to the Google Groups "Graylog 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/graylog2/0c18d852-d69e-4554-be8b-05855851e671%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
