Validating rows before inserting into BigQuery from Dataflow - google-bigquery

According to
How do we set maximum_bad_records when loading a Bigquery table from dataflow? there is currently no way to set the maxBadRecords configuration when loading data into BigQuery from Dataflow. The suggestion is to validate the rows in the Dataflow job before inserting them into BigQuery.
If I have the TableSchema and a TableRow, how do I go about making sure that the row can safely be inserted into the table?
There must be an easier way of doing this than iterating over the fields in the schema, looking at their type and looking at the class of the value in the row, right? That seems error-prone, and the method must be fool-proof since the whole pipeline fails if a single row cannot be loaded.
Update:
My use case is an ETL job that at first will run on JSON (one object per line) logs on Cloud Storage and write to BigQuery in batch, but later will read objects from PubSub and write to BigQuery continuously. The objects contain a lot of information that isn't necessary to have in BigQuery and also contains parts that aren't even possible to describe in a schema (basically free form JSON payloads). Things like timestamps also need to be formatted to work with BigQuery. There will be a few variants of this job running on different inputs and writing to different tables.
In theory it's not a very difficult process, it takes an object, extracts a few properties (50-100), formats some of them and outputs the object to BigQuery. I more or less just loop over a list of property names, extract the value from the source object, look at a config to see if the property should be formatted somehow, apply the formatting if necessary (this could be downcasing, dividing a millisecond timestamp by 1000, extracting the hostname from a URL, etc.), and write the value to a TableRow object.
My problem is that data is messy. With a couple of hundred million objects there are some that don't look as expected, it's rare, but with these volumes rare things still happen. Sometimes a property that should contain a string contains an integer, or vice-versa. Sometimes there's an array or an object where there should be a string.
Ideally I would like to take my TableRow and pass it by TableSchema and ask "does this work?".
Since this isn't possible what I do instead is I look at the TableSchema object and try to validate/cast the values myself. If the TableSchema says a property is of type STRING I run value.toString() before adding it to the TableRow. If it's an INTEGER I check that it's a Integer, Long or BigInteger, and so on. The problem with this method is that I'm just guessing what will work in BigQuery. What Java data types will it accept for FLOAT? For TIMESTAMP? I think my validations/casts catch most problems, but there are always exceptions and edge cases.
In my experience, which is very limited, the whole work pipeline (job? workflow? not sure about the correct term) fails if a single row fails BigQuery's validations (just like a regular load does unless maxBadRecords is set to a sufficiently large number). It also fails with superficially helpful messages like 'BigQuery import job "dataflow_job_xxx" failed. Causes: (5db0b2cdab1557e0): BigQuery job "dataflow_job_xxx" in project "xxx" finished with error(s): errorResult: JSON map specified for non-record field, error: JSON map specified for non-record field, error: JSON map specified for non-record field, error: JSON map specified for non-record field, error: JSON map specified for non-record field, error: JSON map specified for non-record field'. Perhaps there is somewhere that can see a more detailed error message that could tell me which property it was and what the value was? Without that information it could just as well have said "bad data".
From what I can tell, at least when running in batch mode Dataflow will write the TableRow objects to the staging area in Cloud Storage and then start a load once everything is there. This means that there is nowhere for me to catch any errors, my code is no longer running when BigQuery is loaded. I haven't run any job in streaming mode yet, but I'm not sure how it would be different there, from my (admittedly limited) understanding the basic principle is the same, it's just the batch size that's smaller.
People use Dataflow and BigQuery, so it can't be impossible to make this work without always having to worry about the whole pipeline stopping because of a single bad input. How do people do it?

I'm assuming you deserialize the JSON from the file as a Map<String, Object>. Then you should be able to recursively type-check it with a TableSchema.
I'd recommend an iterative approach to developing your schema validation, with the following two steps.
Write a PTransform<Map<String, Object>, TableRow> that converts your JSON rows to TableRow objects. The TableSchema should also be a constructor argument to the function. You can start off making this function really strict -- require that JSON parsed input as Integer directly, for instance, when a BigQuery INTEGER schema was found -- and aggressively declare records in error. Basically, ensure that no invalid records are output by being super-strict in your handling.
Our code here does something somewhat similar -- given a file produced by BigQuery and written as JSON to GCS, we recursively walk the schema and do some type conversions. However, we do not need to validate, because BigQuery itself wrote the data.
Note that the TableSchema object is not Serializable. We've worked around by converting the TableSchema in a DoFn or PTransform constructor to a JSON String and back. See the code in BigQueryIO.java that uses the jsonTableSchema variable.
Use the "dead-letter" strategy described in this blog post to handle bad records -- side output the offending Map<String, Object> rows from your PTransform and write them to a file. That way, you can inspect the rows that failed your validation later.
You might start with some small files and use the DirectPipelineRunner rather than the DataflowPipelineRunner. The direct runner runs the pipeline on your computer, rather than on Google Cloud Dataflow service, and it uses the BigQuery streaming writes. I believe when those writes fail you will get better error messages.
(We use the GCS->BigQuery Load Job pattern for Batch jobs because it's much more efficient and cost-effective, but BigQuery streaming writes in Streaming jobs because they are low-latency.)
Finally, in terms of logging information:
Definitely check Cloud Logging (by following the Worker Logs link on the logs panel.
You may get better information about why the load jobs triggered by your Batch Dataflows fail if you run the bq command-line utility: bq show -j PROJECT:dataflow_job_XXXXXXX.

Related

Updating Parquet datasets where the schema changes overtime

I have a single parquet file that I have been incrementally been building every day for several months. The file size is around 1.1GB now and when read into memory it approaches my PCs memory limit. So, I would like to split it up into several files base on the year and month combination (i.e. Data_YYYYMM.parquet.snappy) that will all be in a directory.
My current process reads in the daily csv that I need to append, reads in the historical parquet file with pyarrow and converts to pandas, concats the new and historical data in pandas (pd.concat([df_daily_csv, df_historical_parquet])) and then writes back to a single parquet file. Every few weeks the schema of the data can change (i.e. a new column). With my current method this is not an issue since the concat in pandas can handle the different schemas and I overwriting it each time.
By switching to this new setup I am worried about having inconsistent schemas between months and then being unable to read in data over multiple months. I have tried this already and gotten errors due to non matching schemas. I thought might be able to specify this with the schema parameter in pyarrow.parquet.Dataset. From the doc it looks like it takes a type of pyarrow.parquet.Schema. When I try using this I get AttributeError: module 'pyarrow.parquet' has no attribute 'Schema'. I also tried taking the schema of a pyarrow Table (table.schema) and passing that to the schema parameter but got an error msg (sry I forget error right now and cant connect workstation right now so cant reproduce error - I will update with this info when I can).
I've seen some mention of schema normalization in the context of the broader Arrow/Datasets project but I'm not sure if my use case fits what that covers and also the Datasets feature is experimental so I dont want to use it in production.
I feel like this is a pretty common use case and I wonder if I am missing something or if parquet isn't meant for schema changes over time like I'm experiencing. I've considered investigating the schema of the new file and comparing vs historical and then if there is change deserializing, updating schema, and reserializing every file in the dataset but I'm really hoping to avoid that.
So my questions are:
Will using a pyarrow parquet Dataset (or something else in the pyarrow API) allow me to read in all of the data in multiple parquet files even if the schema is different? To be specific, my expectation is that the new column would be appended and the values prior to when this column were available would be null). If so, how do you do this?
If the answer to 1 is no, is there another method or library for handling this?
Some resources I've been going through.
https://arrow.apache.org/docs/python/dataset.html
https://issues.apache.org/jira/browse/ARROW-2659
https://arrow.apache.org/docs/python/generated/pyarrow.parquet.ParquetDataset.html#pyarrow.parquet.ParquetDataset

Add dataset parameters into column to use them in BigQuery later with DataPrep

I am importing several files from Google Cloud Storage (GCS) through Google DataPrep and store the results in tables of Google BigQuery. The structure on GCS looks something like this:
//source/user/me/datasets/{month}/2017-01-31-file.csv
//source/user/me/datasets/{month}/2017-02-28-file.csv
//source/user/me/datasets/{month}/2017-03-31-file.csv
We can create a dataset with parameters as outlined on this page. This all works fine and I have been able to import it properly.
However, in this BigQuery table (output), I have no means of extracting only rows with for instance a parameter month in it.
How could I therefore add these Dataset Parameters (here: {month}) into my BigQuery table using DataPrep?
While the original answers were true at the time of posting, there was an update rolled out last week that added a number of features not specifically addressed in the release notes—including another solution for this question.
In addition to SOURCEROWNUMBER() (which can now also be expressed as $sourcerownumber), there's now also a source metadata reference called $filepath—which, as you would expect, stores the local path to the file in Cloud Storage.
There are a number of caveats here, such as it not returning a value for BigQuery sources and not being available if you pivot, join, or unnest . . . but in your scenario, you could easily bring it into a column and do any needed matching or dropping using it.
NOTE: If your data source sample was created before this feature, you'll need to create a new sample in order to see it in the interface (instead of just NULL values).
Full notes for these metadata fields are available here:
https://cloud.google.com/dataprep/docs/html/Source-Metadata-References_136155148
There is currently no access to data source location or parameter match values within the flow. Only the data in the dataset is available to you. (except SOURCEROWNUMBER())
Partial Solution
One method I have been using to mimic parameter insertion into the eventual table is to have multiple dataset imports by parameter and then union these before running your transformations into a final table.
For each known parameter search dataset, have a recipe that fills a column with that parameter per dataset and then union the results of each of these.
Obviously, this is only so scalable i.e. it works if you know the set of parameter values that will match. once you get to the granularity of time-stamp in the source file there is no way this is feasible.
In this example just the year value is the filtered parameter.
Longer Solution (An aside)
The alternative to this I eventually skated to was to define dataflow jobs using Dataprep, use these as dataflow templates and then run an orchestration function that ran the dataflow job (not dataprep) and amended the parameters for input AND output via the API. Then there was a transformation BigQuery Job that did the roundup append function.
Worth it if the flow is pretty settled, but not for adhoc; all depends on your scale.

BQ Switching to TIMESTAMP Partitioned Table

I'm attempting to migrate IngestionTime (_PARTITIONTIME) to TIMESTAMP partitioned tables in BQ. In doing so, I also need to add several required columns. However, when I flip the switch and redirect my dataflow to the new TIMESTAMP partitioned table, it breaks. Things to note:
Approximately two million rows (likely one batch) is successfully inserted. The job continues to run but doesn't insert anything after that.
The job runs in batches.
My project is entirely in Java
When I run it as streaming, it appears to work as intended. Unfortunately, it's not practical for my use case and batch is required.
I've been investigating the issue for a couple of days and tried to break down the transition into the smallest steps possible. It appears that the step responsible for the error is introducing REQUIRED variables (it works fine when the same variables are NULLABLE). To avoid any possible parsing errors, I've set default values for all of the REQUIRED variables.
At the moment, I get the following combination of errors and I'm not sure how to address any of them:
The first error, repeats infrequently but usually in groups:
Profiling Agent not found. Profiles will not be available from this
worker
Occurs a lot and in large groups:
Can't verify serialized elements of type BoundedSource have well defined equals method. This may produce incorrect results on some PipelineRunner
Appears to be one very large group of these:
Aborting Operations. java.lang.RuntimeException: Unable to read value from state
Towards the end, this error appears every 5 minutes only surrounded by mild parsing errors described below.
Processing stuck in step BigQueryIO.Write/BatchLoads/SinglePartitionWriteTables/ParMultiDo(WriteTables) for at least 20m00s without outputting or completing in state finish
Due to the sheer volume of data my project parses, there are several parsing errors such as Unexpected character. They're rare but shouldn't break data insertion. If they do, I have a bigger problem as the data I collect changes frequently and I can adjust the parser only after I see the error, and therefore, see the new data format. Additionally, this doesn't cause the ingestiontime table to break (or my other timestamp partition tables to break). That being said, here's an example of a parsing error:
Error: Unexpected character (',' (code 44)): was expecting double-quote to start field name
EDIT:
Some relevant sample code:
public PipelineResult streamData() {
try {
GenericSection generic = new GenericSection(options.getBQProject(), options.getBQDataset(), options.getBQTable());
Pipeline pipeline = Pipeline.create(options);
pipeline.apply("Read PubSub Events", PubsubIO.readMessagesWithAttributes().fromSubscription(options.getInputSubscription()))
.apply(options.getWindowDuration() + " Windowing", generic.getWindowDuration(options.getWindowDuration()))
.apply(generic.getPubsubToString())
.apply(ParDo.of(new CrowdStrikeFunctions.RowBuilder()))
.apply(new BigQueryBuilder().setBQDest(generic.getBQDest())
.setStreaming(options.getStreamingUpload())
.setTriggeringFrequency(options.getTriggeringFrequency())
.build());
return pipeline.run();
}
catch (Exception e) {
LOG.error(e.getMessage(), e);
return null;
}
Writing to BQ. I did try to set the partitoning field here directly, but it didn't seem to affect anything:
BigQueryIO.writeTableRows()
.to(BQDest)
.withMethod(Method.FILE_LOADS)
.withNumFileShards(1000)
.withTriggeringFrequency(this.triggeringFrequency)
.withTimePartitioning(new TimePartitioning().setType("DAY"))
.withWriteDisposition(BigQueryIO.Write.WriteDisposition.WRITE_APPEND)
.withCreateDisposition(BigQueryIO.Write.CreateDisposition.CREATE_NEVER);
}
After a lot of digging, I found the error. I had parsing logic (a try/catch) that returned nothing (essentially a null row) in the event there was a parsing error. This would break BigQuery as my schema had several REQUIRED rows.
Since my job ran in batches, even one null row would cause the entire batch job to fail and not insert anything. This also explains why streaming inserted just fine. I'm surprised that BigQuery didn't throw an error claiming that I was attempting to insert a null into a required field.
In reaching this conclusion, I also realized that setting the partition field in my code was also necessary as opposed to just in the schema. It could be done using
.setField(partitionField)

Python Apache Beam: BigQuery streaming deduplication by row_id

According to BigQuery docs, you can ensure data consistency providing an insertId (https://cloud.google.com/bigquery/streaming-data-into-bigquery#dataconsistency). If it's not provided, BQ will try to ensure consistency based on internals Ids and best-effort.
Using the BQ API you can do that with the row_ids param (https://google-cloud-python.readthedocs.io/en/latest/bigquery/generated/google.cloud.bigquery.client.Client.insert_rows_json.html#google.cloud.bigquery.client.Client.insert_rows_json) but I can't find the same for the Apache Beam Python SDK.
Looking into the SDK I have noticed that a 'unique_row_id' property exist, but I really don't know how to pass my param to WriteToBigQuery()
How can I write into BQ (streaming) providing a row Id for deduplication?
Update:
If you use WriteToBigQuery then it will automatically create and
insert a unique row id called insertId for you, which will be inserted to bigquery. It's handled for you, you don't need to worry about it. :)
WriteToBigQuery is a PTransform, and in it's expand method calls BigQueryWriteFn
BigQueryWriteFn is a DoFn, and in it's process method calls _flush_batch
_flush_batch is a method that then calls the BigQueryWrapper.insert_rows method
BigQueryWrspper.insert_rows creates a list of bigquery.TableDataInsertAllRequest.RowsValueListEntry objects which contain the insertId and the row data as a json object
The insertId is generated by calling the unique_row_id method which returns a value consisting of UUID4 concatenated with _ and with an auto-incremented number.
In the current 2.7.0 code, there is this happy comment; I've also verified it is true :)
https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/gcp/bigquery.py#L1182
# Prepare rows for insertion. Of special note is the row ID that we add to
# each row in order to help BigQuery avoid inserting a row multiple times.
# BigQuery will do a best-effort if unique IDs are provided. This situation
# can happen during retries on failures.
* Don't use BigQuerySink
At least, not in it's current form as it doesn't support streaming. I guess that might change.
Original (non)answer
Great question, I also looked and couldn't find a certain answer.
Apache Beam doesn't appear to use that google.cloud.bigquery client sdk you've linked to, it has some internal generated api client, but it appears to be up-to-date.
I looked at the source:
The insertall method is there https://github.com/apache/beam/blob/18d2168ee71a1b1b04976717f0f955199bb00961/sdks/python/apache_beam/io/gcp/internal/clients/bigquery/bigquery_v2_client.py#L476
I also found the insertid mentioned
https://github.com/apache/beam/blob/master/sdks/python/apache_beam/io/gcp/internal/clients/bigquery/bigquery_v2_messages.py#L1707
So if you can make an InsertAll call it will use a TableDataInsertAllRequest and pass a RowsValueListEntry
class TableDataInsertAllRequest(_messages.Message):
"""A TableDataInsertAllRequest object.
Messages:
RowsValueListEntry: A RowsValueListEntry object.
The RowsValueListEntry message is where the insertid is.
Here's the API docs for insert all
https://cloud.google.com/bigquery/docs/reference/rest/v2/tabledata/insertAll
I will look some more at this because I don't see the WriteToBigQuery() exposing this.
I suspect that the 'bigquery will remember this for at least one minute` is a pretty loose guarantee for de-duping. The docs suggest using datastore if you need transactions. Otherwise you might need to run SQL with window functions to de-dupe at runtime, or run some other de-duping jobs on bigquery.
Perhaps using batch_size parameter of WriteToBigQuery(), and running a combine (or at worst a GroupByKey) step in dataflow is a more stable way to de-dupe prior to writing.

Caching of Map applications in Hadoop MapReduce?

Looking at the combination of MapReduce and HBase from a data-flow perspective, my problem seems to fit. I have a large set of documents which I want to Map, Combine and Reduce. My previous SQL implementation was to split the task into batch operations, cumulatively storing what would be the result of the Map into table and then performing the equivalent of a reduce. This had the benefit that at any point during execution (or between executions), I had the results of the Map at that point in time.
As I understand it, running this job as a MapReduce would require all of the Map functions to run each time.
My Map functions (and indeed any function) always gives the same output for a given input. There is simply no point in re-calculating output if I don't have to. My input (a set of documents) will be continually growing and I will run my MapReduce operation periodically over the data. Between executions I should only really have to calculate the Map functions for newly added documents.
My data will probably be HBase -> MapReduce -> HBase. Given that Hadoop is a whole ecosystem, it may be able to know that a given function has been applied to a row with a given identity. I'm assuming immutable entries in the HBase table. Does / can Hadoop take account of this?
I'm made aware from the documentation (especially the Cloudera videos) that re-calculation (of potentially redundant data) can be quicker than persisting and retrieving for the class of problem that Hadoop is being used for.
Any comments / answers?
If you're looking to avoid running the Map step each time, break it out as its own step (either by using the IdentityReducer or setting the number of reducers for the job to 0) and run later steps using the output of your map step.
Whether this is actually faster than recomputing from the raw data each time depends on the volume and shape of the input data vs. the output data, how complicated your map step is, etc.
Note that running your mapper on new data sets won't append to previous runs - but you can get around this by using a dated output folder. This is to say that you could store the output of mapping your first batch of files in my_mapper_output/20091101, and the next week's batch in my_mapper_output/20091108, etc. If you want to reduce over the whole set, you should be able to pass in my_mapper_output as the input folder, and catch all of the output sets.
Why not apply your SQL workflow in a different environment? Meaning, add a "processed" column to your input table. When time comes to run a summary, run a pipeline that goes something like:
map (map_function) on (input table filtered by !processed); store into map_outputs either in hbase or simply hdfs.
map (reduce function) on (map_outputs); store into hbase.
You can make life a little easier, assuming you are storing your data in Hbase sorted by insertion date, if you record somewhere timestamps of successful summary runs, and open the filter on inputs that are dated later than last successful summary -- you'll save some significant scanning time.
Here's an interesting presentation that shows how one company architected their workflow (although they do not use Hbase):
http://www.scribd.com/doc/20971412/Hadoop-World-Production-Deep-Dive-with-High-Availability