Hi team, I have a simple question : What`s the difference between MCE/MAE consumer job inside de GMS and standalone, please? is it possible inside GMS ( env MCE/MAE = true) get endpoins metrics like this http://localhost:9091/actuator/metrics ? thank you

Inside and standalone are mainly for scaling purposes. If your workload gets bigger it might make sense to run standalone to scale those out independently

Hi @square-activity-64562, tks for explanation and about second question when you run standalone have endpoints possibility like git below, but inside gms (not standalone) we have one endpoint to try get metrics or not , please? https://github.com/linkedin/datahub/tree/master/metadata-jobs/mae-consumer-job

@gentle-florist-49869, if you run the embedded mae/mce in gms with docker, then check the tcp/ip port, you will not see

9091

and

9090

port.

@acceptable-architect-70237 yes, this exactly question, when run embedded mae/mce in gms , how cab I check the metrics for mae/mce consumer job please?

my understanding is that , with embedded

mae/mce

, gms itself has the capability of consuming

mce

and

mae

so we won't get a metrics as standalone's mae or mce's.

Unfortunately, GMS does not use spring boot, so doesn’t have the actuator metrics exposed. However, you can try setting ENABLE_PROMETHEUS=true in gms, which exports JMX metrics in Prometheus format

@acceptable-architect-70237 @early-lamp-41924 tks guys, I will try to enable prometheus and looking for about metrics. After when I move my lab to k8s probably create MAE/MCE standalone service/pod