I am using flume + kafka to sink the log data to hdfs. My sink data type is Avro. In avro schema (.avsc), there is 80 fields as columns.
So I created an external table like that
CREATE external TABLE pgar.tiz_biaws_fraud
PARTITIONED BY(partition_date INT)
ROW FORMAT SERDE 'org.apache.hadoop.hive.serde2.avro.AvroSerDe'
STORED AS INPUTFORMAT 'org.apache.hadoop.hive.ql.io.avro.AvroContainerInputFormat'
OUTPUTFORMAT 'org.apache.hadoop.hive.ql.io.avro.AvroContainerOutputFormat'
LOCATION '/data/datapool/flume/biaws/fraud'
TBLPROPERTIES ('avro.schema.url'='hdfs://xxxx-ns/data/datapool/flume/biaws/fraud.avsc')
Now, I need to add 25 more columns to avro schema. In that case,
if I create a new table with new schema which has 105 columns, I will have two table for one project. And if I add or remove some columns in coming days, I have to create a new table for that. I am afraid of having a lot of table which use different schema for same project.
If I swap the old schema with new schema in current table, I will have only one table for one project but I can't read and get old data anymore because of schema conflict.
What is the best way to use avro schema in case like that?
This is indeed challenging. The best way is to make sure all schema changes you make are compatible with the old data - so only remove columns with defaults, and make sure you give defaults in the columns you are adding. This way you can safely swap out the schemas without a conflict and keep reading old data. Avro is pretty clever about that, it's called "schema evolution" (in case you want to google a bit more) and allows reader and writer schemas to be a bit different.
As an aside, I want to mention that Kafka has a native HDFS connector (i.e. without Flume) that uses Confluent's schema registry to handle these kinds of schema changes automatically - you can use the registry to check if the schemas are compatible, and if they are - simply write data using the new schema and the Hive table will automatically evolve to match.
I added new columns to avro schema like that
{"name":"newColumn1", "type": "string", "default": ""},
{"name":"newColumn2", "type": "string", "default": ""},
{"name":"newColumn3", "type": "string", "default": ""},
When I use default property, if that columns doesn't exist in current data it returns default value, if that columns does exist in current data it returns the data value as expected.
For setting null value as default, you need that
{ "name": "newColumn4", "type": [ "string", "null" ], "default": "null" },
or
{ "name": "newColumn5", "type": [ "null", "string" ]},
The position of null in type property, can be first place or can be second place with default property.
Related
In our database we have over 20 fields which we need to transform from long to timestamp. Why there is no generic solution to transfer all this value ?
I know I can define:
"transforms":"tsFormat",
"transforms.tsFormat.type": "org.apache.kafka.connect.transforms.TimestampConverter$Value",
"transforms.tsFormat.target.type": "string",
"transforms.tsFormat.field": "ts_col1",
"transforms.tsFormat.field": "ts_col2",
but this is not solution for us. When we add new timestamp to db we need to update connector too
is there some generic solution to transform all fields according to avro schema ?
We are using debezium which for all timestamp fields create something like this:
{
"name": "PLATNOST_DO",
"type": {
"type": "long",
"connect.version": 1,
"connect.name": "io.debezium.time.Timestamp"
}
},
so how to find all type with connect.name = 'io.debezium.time.Timestamp' and transform it to timestamp
You'd need to write your own transform to be able to dynamically iterate of the record schema, check the types, and do the conversion.
Thus why they are called simple message transforms.
Alternatively, maybe take a closer look at the Debezium properties to see if there is a missing setting that alters how timestamps get produced.
I'm actually working on a Meltano project where I have to extract data from one postgres database and load to the final warehouse (also a postgres database) using "Key-based Incremental Replication".
After the loading process of meltano, all table with columns of type uuid from the "tap-postgres" are changed to varchar in the target.
Could someone help to solve this issue?
Singer taps rarely capture "rich" string types such as uuid, and they output simple {"type": "string"} types without any additional metadata. Furthermore, Singer usually only supports JSONSchema Draft 4, so {"type": "string", "format": "uuid"} would not even be supported by most taps and targets.
Since Singer is an ELT analytics framework and in that context a uuid column type is not different in any meaningful way from a varchar, this is hardly an issue for most people.
If you need to replicate a database with that level of detail you may be better served by a different solution for database backup/replication.
I'm trying to apply Debezium's New Record State Extraction SMT using the following configuration:
"transforms": "unwrap",
"transforms.unwrap.type": "io.debezium.transforms.ExtractNewRecordState",
"transforms.unwrap.drop.tombstones": true,
"transforms.unwrap.delete.handling.mode": "rewrite",
"transforms.unwrap.add.fields": "db,schema,table,txId,ts_ms"
For INSERT and UPDATE operations I get the messages as expected, but in case of DELETE I get the following as a payload:
"payload": {
"id": 2,
"first_name": "",
"last_name": "",
"__db": "postgres",
"__schema": "schema1",
"__table": "user_details",
"__txId": 5145,
"__ts_ms": 1638760801510,
"__deleted": "true"
}
As you can see above, both first_name and last_name fields have empty values, though the record I deleted has non-empty values for both of those fields. What I expect to see as a value for those 2 fields is their value at the moment of deletion as it is shown in debezium's before payload chunk in case when New Record State Extraction SMT is not applied.
The reason of empty values for all columns except the PK is not related to New Record State Extraction SMT at all. For postgres, there is a REPLICA IDENTITY table-level parameter that can be used to control the information written to WAL to identify tuple data that is being deleted or updated.
This parameter has 4 modes:
DEFAULT
USING INDEX index
FULL
NOTHING
In the case of DEFAULT, old tuple data is only identified with the primary key of the table. Columns that are not part of the primary key do not have their old value written.
In the case of FULL, all the column values of old tuple are properly written to WAL all the time. Hence, executing the following command for the target table will make the old record values to be properly populated in debezium message:
ALTER TABLE some_table REPLICA IDENTITY FULL;
NOTE!! FULL is the most verbose, and as well the most resource-consuming mode. Be careful with it particularly for heavily-updated tables.
I am using a 3rd party CDC tool that replicates data from a source database into Kafka topics. An example row is shown below:
{
"data":{
"USER_ID":{
"string":"1"
},
"USER_CATEGORY":{
"string":"A"
}
},
"beforeData":{
"Data":{
"USER_ID":{
"string":"1"
},
"USER_CATEGORY":{
"string":"B"
}
}
},
"headers":{
"operation":"UPDATE",
"timestamp":"2018-05-03T13:53:43.000"
}
}
What configuration is needed in the sink file in order to extract all the (sub)fields under data and headers and ignore those under beforeData so that the target table in which the data will be transferred by Kafka Sink will contain the following fields:
USER_ID, USER_CATEGORY, operation, timestamp
I went through the transformation list in confluent's docs but I was not able to find how to use them in order to achieve the aforementioned target.
I think you want ExtractField, and unfortunately, it's a Map.get operation, so that means 1) nested fields cannot be gotten in one pass 2) multiple fields need multiple transforms.
That being said, you might to attempt this (untested)
transforms=ExtractData,ExtractHeaders
transforms.ExtractData.type=org.apache.kafka.connect.transforms.ExtractField$Value
transforms.ExtractData.field=data
transforms.ExtractHeaders.type=org.apache.kafka.connect.transforms.ExtractField$Value
transforms.ExtractHeaders.field=headers
If that doesn't work, you might be better off implementing your own Transformations package that can at least drop values from the Struct / Map.
If you're willing to list specific field names, you can solve this by:
Using a Flatten transform to collapse the nesting (which will convert the original structure's paths into dot-delimited names)
Using a Replace transform with rename to make the field names be what you want the sink to emit
Using another Replace transform with whitelist to limit the emitted fields to those you select
For your case it might look like:
"transforms": "t1,t2,t3",
"transforms.t1.type": "org.apache.kafka.connect.transforms.Flatten$Value",
"transforms.t2.type": "org.apache.kafka.connect.transforms.ReplaceField$Value",
"transforms.t2.renames": "data.USER_ID:USER_ID,data.USER_CATEGORY:USER_CATEGORY,headers.operation:operation,headers.timestamp:timestamp",
"transforms.t3.type": "org.apache.kafka.connect.transforms.ReplaceField$Value",
"transforms.t3.whitelist": "USER_ID,USER_CATEGORY,operation,timestamp",
I have an Avro schema with this property:
{"name": "whenDate", "type": ["date", "null"]}
I am using a Python client and the producer
confluent_kafka.avro.AvroProducer
When I load the Avro schema with aforementioned property,
I trigger this error:
not a valid Avro schema: Type property "[u'date', u'null']" not a valid
Avro schema: Union item must be a valid Avro schema: Could not make an Avro Schema object from date.
I presume this error speaks to limitations in the Confluent parser for Avro schema. That is, date is a logical type specified here, but it not as yet supported by the Confluent parser.
Can anyone confirm? More generally, should I assume that the Confluent parser does not handle the logical types for Avro schema? Thanks.
It is a logical type, I think you just need to declare it as a one.
{"name": "date", "type": "int", "logicalType": "date"}