Closed ddides closed 4 years ago
Was thinking about this a little more and thought it was worth mentioning -- the logs are being forwarded to Elasticsearch ultimately and viewed on Kibana, and that's where I'm seeing that they are individual lines as well.
@ddides the plugin expects proper Golang panic (exception/crash), which looks like:
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0xffffffff addr=0x0 pc=0xe01da]
goroutine 1 [running]:
main.main()
/tmp/sandbox382288397/prog.go:10 +0x1a
you can find the defined pattern here: https://github.com/GoogleCloudPlatform/fluent-plugin-detect-exceptions/blob/master/lib/fluent/plugin/exception_detector.rb#L97-L98
In your case, this is programmatically generated StackTrace debug.PrintStack()
, which is not and panic.
Thanks, @Trane9991! That is absolutely correct. Sounds like WAI.
Thanks for the information! Sorry I missed this response back in March.
I'm hoping this is an issue with something I'm doing, but I cannot figure out for the life of me what's wrong with my configuration. I've tried a few different fluentd base images and have had no luck.
fluent.conf
:Container logs after starting up:
Log file (Go
debug.PrintStack()
):For good measure, I also tried starting up a container running a Python container and generating an error with that to see if it was an issue with the Go patterns. I had the same results:
I'm not sure what I'm doing wrong here, so any guidance (configuration or even just how to debug a fluentd plugin) would be super appreciated! It looks like the plugin is being called though as it removes the
raw
tag successfully.