Fluentd Google Group

Hi Kentaro, So I did some more digging into Fluentd and Prometheus client code. Here is a solution I

Hi, I've assumed that @type redisoutput plugin [1] so I've thought that insert_key_prefix may

Hi, Could you explain in detail to reproduce the steps for your issue (what you did/what you didn

It depends on elasticsearch-ruby version. See https://github.com/elastic/elasticsearch-ruby#

From td-agent log, td-agent seems to receive broken log. log shows your data is '{{...}"}

Currently, no limit for payload size. How big is your record? Could you show an example? On Tue, Jan

I have tried it but not working .. services hung to restart. On Mon, Jan 11, 2021 at 10:29 PM Kentaro

With multiprocess plugin. Each process can use "cmdline --log filepath" to special the td-

Awesome! I'm happy to hear that :) - Hiroshi 2020年12月29日火曜日 5:54:44 UTC+9 Darren Spruell: Further

Hi, It seems that there is a tkgi metadata parser [1], but does not fit for you? [1] https://github.

Hi, ``` n.times do begin ... rescue ... else break end sleep ... end ``` Simply just repeat and break

Hi, How do you configure @type elasticseach in match section? Then did you see something wrong (

Hi, It seems that ruby will not be supported officially in distroless. Add a Ruby distroless? https:/

This happens no matter what port is use. On Sunday, December 27, 2020 at 9:45:40 PM UTC-8 Kelvin

Hi, It's fresh install but I cannot start it. See the logs attached. Thanks, </source> <

This is hard to agree. Because your statement conflicts with the log below which was successfully

We are thinking to migrate v0.12 to newer version. Our production has more than 100 files of

Hi, It seems that there is no such a plugin yet [1] so it may need to collect memory usage by

There is a constant TCP connection from the app to fluent server. So no memory on extra sockets...

Thanks Mr. Fiber for sharing the PPT, it helps a lot. On Wed, 16 Dec 2020 at 18:55, Mr. Fiber <

Please paste the configuration and log example. Without these information, we only say "

> So does fluentd update the pos_file as soon as it reads the event in the input? No. in_tail will

> Before shutdown and reassign workers, you can send USR1 signal to flush buffers forcibly. Oh, I

Hi, How about using curator? It's a CLI tool to manage Elasticsearch indices. documentation:

Hi, FYI: It seems that it is already known issue and currently no workaround. impossible to use

Hi, Even though it is not bundled by default, and I'm not sure to match your needs, you may be

Hi, It is not direct answer for your questions, but some may help you. There is performance-related

Hi, I have no idea for cleaning indices on Kibana UI, but I found that elastic's curator can

Hi, It seems that fluentd does not support Basic auth in in_http. So, it may be easy to use <

Hi, We have integrated Prometheus and retrieving metrics from fluentd. Now we need to test different