Additional context
Node API is having kafka producer settings with KAFKA_PRINCIPAL_ID. Same machine we are running kinit command with different CACHE_PRINCIPAL_ID to get krb token for different purpose.
When we publish a message to topic it is getting published successfully using the above configuration, but some time it is trying to publish using the CACHE_PRINCIPAL_ID instead of KAFKA_PRINCIPAL_ID and it is getting denied.
Environment Information
node-rdkafka Configuration Settings
{
"producerConfig": { "bootstrap.servers": "MY_BROKER_DETAILS", "client.id": "localhost", "debug": "consumer,cgrp,fetch,msg,eos", "event_cb": true, "message.max.bytes": 1048576, "max.in.flight.requests.per.connection": 1, "sasl.mechanism": "GSSAPI", "sasl.kerberos.principal": "KAFKA_PRINCIPAL_ID", "sasl.kerberos.service.name": "kafka", "sasl.kerberos.keytab": "MY_KRB_KEYTAB", "sasl.kerberos.kinit.cmd": "kinit -V -R -t \"%{sasl.kerberos.keytab}\" -k %{sasl.kerberos.principal} || kinit -V -t \"%{sasl.kerberos.keytab}\" -k %{sasl.kerberos.principal}", "sasl.kerberos.min.time.before.relogin": 90000, "security.protocol": "sasl_plaintext", "compression.codec": "lz4", "delivery.timeout.ms": 2147483647, "dr_msg_cb": true, "enable.idempotence": false, "linger.ms": 100, "message.send.max.retries": 3, "request.timeout.ms": 180000, "retry.backoff.ms": 200 }, "topicConfig": { "acks": -1, "compression.type": "lz4" } }
Additional context Node API is having kafka producer settings with KAFKA_PRINCIPAL_ID. Same machine we are running kinit command with different CACHE_PRINCIPAL_ID to get krb token for different purpose.
When we publish a message to topic it is getting published successfully using the above configuration, but some time it is trying to publish using the CACHE_PRINCIPAL_ID instead of KAFKA_PRINCIPAL_ID and it is getting denied.
I appreciate any help one this.