How many Inputs Graylog 5.2 support maximum

Graylog Central (peer support)
1

GrayLog Support

I have searched all over GrayLog community forum and google not found the exact answer.

I am using Graylog 5.2 on Ubuntu 22.0 LTS - with decent RAM 64GB and 2TB hdd,

I have different requirement each Service should use different Port so i can stream Line the Data in correct Manner while processing the Logs.
So the question - How many maximum inputs i can configure - what is the drawback if i use 50 inputs ?

Thank you
R!

2

I have never heard that 50 would give you issues, but there is overhead to each input.

What is the reason you need all the services running on different ports?

The bigger concern is it sounds like you are running both opensearch and graylog on the same server, that can definitely give you issues, what is the volume of logs you are expecting?

3

Appreciated your quick reply here.
For now kind of PoC i am doing see all in One Server can able to defeat my aim to Build Log Server (For now i have 1 Log Server - syslog-ng for all the Logging - setup 7 yeas back working as expected - with some Limitations)
Now planning to move to Graylog as many thing changed in the new Graylog 5.2 (impressed)

The reason i have more than 50 inputs - because i am thinking multi-tenant for now - (this is a design consideration question) - what i am thinking of deploy correct or not (based on industry experience like you).

Once this working - i am reading the documents for Graylog+MongoDB Cluster - OpenSearch Cluster ( start with 3node both and test it).

For now i am expecting 10Gb to 20G Logs. (Hope Opensource can accommodate)
or Opensource have any Limitation of the logs go beyond. 100GB ?

Compute and building new server not a problem have enough capacity.

4

any takers ? regarding the post ?

5

Open-source does not have a size limit so you will not have a problem with that part.

The normal recommendation would be to have as few inputs as possible, only create a new input when an input setting needs to be changed (store raw log message etc). Then use pipeline rules to route the logs to different streams based on rules to decide what customer each is from.

6

Make sense, appreciated your input.

7

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.