You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@avro.apache.org by "Tobias Fielitz (Jira)" <ji...@apache.org> on 2022/06/15 08:52:00 UTC
[jira] [Commented] (AVRO-2126) avro.io.SchemaResolutionException: Can't access branch index 36 for union with 2 branches
[ https://issues.apache.org/jira/browse/AVRO-2126?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17554467#comment-17554467 ]
Tobias Fielitz commented on AVRO-2126:
--------------------------------------
Same issue here:
{code:java}
avro.errors.SchemaResolutionException: Can't access branch index 23 for union with 2 branches
Writer's Schema: [
"null",
"string"
]
Reader's Schema: [
"null",
"string"
]{code}
Skipping first 5 bytes already.
Did anyone solve this? Know how to debug or even know what this error means?
I am fetching the schema directly from the kafka schema registry and a message from the queue with matching topic. I don't see why there should be any mismatch.
> avro.io.SchemaResolutionException: Can't access branch index 36 for union with 2 branches
> -----------------------------------------------------------------------------------------
>
> Key: AVRO-2126
> URL: https://issues.apache.org/jira/browse/AVRO-2126
> Project: Apache Avro
> Issue Type: Bug
> Components: python
> Affects Versions: 1.8.2
> Environment: Python 3.5.2, avro-python3 1.8.2
> Reporter: Jurgis Pods
> Priority: Major
>
> I get the following error when decoding messages from Kafka via avro-python3:
> {code}
> avro.io.SchemaResolutionException: Can't access branch index 36 for union with 2 branches
> Writer's Schema: [
> "null",
> "string"
> ]
> Reader's Schema: [
> "null",
> "string"
> ]
> {code}
> Relevant code producing this error:
> {code}
> import avro.schema
> import avro.io
> import io
> def parse_avro(value, schema):
> bytes_reader = io.BytesIO(value)
> decoder = avro.io.BinaryDecoder(bytes_reader)
> reader = avro.io.DatumReader(schema)
> return reader.read(decoder)
> # msg: Message from Kafka, schema: avro.schema.Schema from Confluent Schema Registry
> parse_avro(msg.value, avro_schema)
> {code}
> I have no idea where the number 36 might come from.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)