Skip to content
This repository has been archived by the owner on Nov 24, 2021. It is now read-only.

Client configurations? #2

Open
brockn opened this issue Jun 4, 2018 · 4 comments
Open

Client configurations? #2

brockn opened this issue Jun 4, 2018 · 4 comments

Comments

@brockn
Copy link
Contributor

brockn commented Jun 4, 2018

From the base log appender:

curl -X POST -H 'Content-Type: application/json' -d '{"category": "'$category'","timestamp": '$timestamp', "level": "'$level'", "message": "'$message'", "threadName": "'$threadName'"}' http://0.0.0.0:9005/log?application=$application

It'd be nice to get this from a client config, eg /etc/pulse/conf/env.sh so I'll I'd have to use as a user is:

source /opt/cloudera/parcels/PULSE/lib/appenders/logger.sh

and I'd be running.

@bpmcd
Copy link
Contributor

bpmcd commented Jun 18, 2018

Another client asked today about the same possibility, but for the log4j configuration in a Spark app.

@afoerster
Copy link
Contributor

@brockn task created

@bpmcd task created, for this to work we would have to reload log4j configuration at runtime so a little more challenging.

@brockn
Copy link
Contributor Author

brockn commented Jun 19, 2018

reload log4j configuration at runtime so a little more challenging.

feels like a non-goal

@afoerster
Copy link
Contributor

I haven't quite figured this one out yet. I can create an environment file using the parcel, but that doesn't have information about the host running the log-collector role. The descriptor.sdl has a 'peerConfigGenerator' to get hostnames running a role, but it will only share the peer config with other roles of the same type.

If Pulse was depending on another service this would be easy, for example Kafka depends on Zookeeper so it's easy for Kafka to get the Zookeeper hosts.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants