https://datahubproject.io logo
Join Slack
Powered by
I tried pulling the latest version of the code (no...
# troubleshoot
b
I tried pulling the latest version of the code (not the releases) off the repo to check out view ACL, but it seems like the search bar is broken after ingesting some datasets checking mySQL and ES, both datasources have the new dataset records already. GMS is not returning any errors in logs graphql search also not returning me any results after i ingested the datasets. * returns me the results BEFORE ingestion took place.
e
Whats the exact behavior you see? Is it that newly ingested datasets are not showing up on search?
b
yes, the newly ingested datasets are not showing up. (this was based on the repo pulled 24h ago)
e
hmn is this still the case?
I don’t recall any changes we made to how search retrieves things
b
yup, still happening. i pruned all the volumes, containers, images, and started datahub via
Copy code
docker-compose -p datahub -f docker-compose.yml -f docker-compose.override up -d
then followed by ingestion of sample data:
Copy code
cd ./ingestion && docker-compose up -d
logged in as datahub, unable to find anything via search bar except for the Datahub and John Doe entity. ES index is showing datasets, gms is not showing anything unusual, all containers are running, tried to run the datahub-upgrade job via
Copy code
./datahub-upgrade.sh -u NoCodeDataMigration
but it says:
Copy code
The bean 'kafkaProducerFactory', defined in class path resource [org/springframework/boot/autoconfigure/kafka/KafkaAutoConfiguration.class], could not be registered. A bean with that name has already been defined in URL [jar:file:/datahub/datahub-upgrade/bin/datahub-upgrade.jar!/BOOT-INF/lib/factories.jar!/com/linkedin/gms/factory/kafka/KafkaProducerFactory.class] and overriding is disabled.
30minutes later, still no search results
what worked for me so far, is to bring down all the containers, delete the broker volume, and restart all the containers again, then the search will work again. Previously, I brought down the containers, but was unable to bring it up again because broker complained
Copy code
[36mbroker                    |[0m [2022-04-01 07:11:11,234] ERROR Fatal error during KafkaServer startup. Prepare to shutdown (kafka.server.KafkaServer)
[36mbroker                    |[0m kafka.common.InconsistentClusterIdException: The Cluster ID V9WljOYIR7OA8OF2v41BiA doesn't match stored clusterId Some(FbqCKN4xQLuM-8w7GjM36A) in meta.properties. The broker is trying to join the wrong cluster. Configured zookeeper.connect may be wrong.
[36mbroker                    |[0m 	at kafka.server.KafkaServer.startup(KafkaServer.scala:220)
[36mbroker                    |[0m 	at io.confluent.support.metrics.SupportedServerStartable.startup(SupportedServerStartable.java:114)
[36mbroker                    |[0m 	at io.confluent.support.metrics.SupportedKafka.main(SupportedKafka.java:66)
[36mbroker                    |[0m [2022-04-01 07:11:11,245] INFO shutting down (kafka.server.KafkaServer)
[36mbroker                    |[0m [2022-04-01 07:11:11,250] INFO [ZooKeeperClient Kafka server] Closing. (kafka.zookeeper.ZooKeeperClient)
so i tried removing the broker volume, and it solved both the broker down and search not working issue together
2 Views
Open in Slack
PreviousNext
Powered by