[ 
https://issues.apache.org/jira/browse/NIFI-8785?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17387244#comment-17387244
 ] 

Pierre Villard commented on NIFI-8785:
--------------------------------------

I didn't reproduce the issue when playing with the Schema Registry of Confluent 
Cloud but I added another approach that uses the /id/versions API which can be 
used to completely remove the calls to the /subjects API (it could be because 
of a specific subject name strategy according to 
[https://docs.confluent.io/platform/current/schema-registry/serdes-develop/index.html#subject-name-strategy]
 - in particular the TopicRecordNameStrategy strategy may require manual 
subject name registration when the topic is a replicated topic.

After a discussion with [~atavitov], my proposed changes are working but I'm 
still not sure why the original /subjects API call is not working as expected. 
I added a bunch of debug logs in case we have additional issues on the 
controller service.

> AvroReader controller can't get confluent schema IDs except only the latest
> ---------------------------------------------------------------------------
>
>                 Key: NIFI-8785
>                 URL: https://issues.apache.org/jira/browse/NIFI-8785
>             Project: Apache NiFi
>          Issue Type: Bug
>          Components: Extensions
>    Affects Versions: 1.12.1
>         Environment: CentOS7
>            Reporter: Arsen Tavitov
>            Priority: Major
>         Attachments: Kafka_Confluent_Schema_Suffering.json, nifi-app.rar
>
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> While trying to read Confluent Schema Registry Avro files(from Kafka) with 
> different schema IDs with AvroReader controller error "could not get Schema 
> with id: <schema id>" appears.
> NiFi version: 1.12.1
>  Confluent Schema Registry version: 5.5.2
> Using the following test flow in Nifi
> ConsumeKafka_2_6 1.12.1 ---> ConvertRecord 1.12.1
> In ConvertRecord processor we are using AvroReader controller as record 
> reader.
>  AvroReader controller properties:
>  * schema access strategy: confluent content-encoded schema reference
>  * schema registry: Confluent Schema Registry controller 
>  * schema name: ${schema.name} (default value)
>  * schema version: no value set
>  * schema branch: no value set
>  * schema text: ${schema.text} (default value)
>  * cash size: 1000 (default value)
> Similar issue:
> [https://community.cloudera.com/t5/Support-Questions/Nifi-ConsumeKafakRecord-0-11-Throws-Error-quot-Could-not/td-p/185636]
>  



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to