Glue job keeps running and does not write results

Question

I have created a job to migrate my **Postgres data to S3**, I am implementing full load right now, the table consists of a ** lot of records**(count-17496724), that is why I added 10 workers along with **auto scaling** option checked, but i keep getting this error. The job continues to run for long and does not generate any output. 
Have tried other combination of worker numbers too like 5,10, but same error.
Below are the errors from the Logs:

1. ERROR [dispatcher-CoarseGrainedScheduler] scheduler.TaskSchedulerImpl (Logging.scala:logError(73)): Lost executor 1 on 10.0.4.209: Remote RPC client disassociated. Likely due to containers exceeding thresholds, or network issues. Check driver logs for WARN messages.

2. ERROR [dispatcher-CoarseGrainedScheduler] scheduler.TaskSchedulerImpl (Logging.scala:logError(73)): Lost executor 2 on 10.0.4.209: Remote RPC client disassociated. Likely due to containers exceeding thresholds, or network issues. Check driver logs for WARN messages.

**About the job:**
Creating JDBC connection to the Postgres database and writing the data into S3 as parquet files

How do I perform the load, I need to add the incremental logic later.

Answer

Hello,

Above error occurs when exceutor requires more momemory to process the data than configured. For G.1x worker Glue uses 10 GB and for G.2x it uses 20 GB. By default spark read operation from JDBC sources are not parallel and use one connection to read the entire data. To resolve this issue, read the JDBC table in parallel.

You can use hashexpression and hashfield along with hashpartition to read the data in parallel using Glue dynamic frame. Please refer to article for detail explanation:

https://aws.amazon.com/premiumsupport/knowledge-center/glue-lost-nodes-rds-s3-migration/

Regarding the incremental load, You can use Glue bookmark feature to read the data from JDBC sources. Please refer to article for the same as there are some prerequisite to use bookmark with JDBC sources.

https://docs.aws.amazon.com/glue/latest/dg/monitor-continuations.html

If you still face any issue, Please feel free to reach out to AWS Premium Support with script  and jobrunid and we will be happy to help.

Glue job keeps running and does not write results

Relevant questions

Glue Workflow: Add same glue job multiple times, each with different parameters

AWS Glue ETL Job: IllegalArgumentException: Missing collection name.

What are the benefits when I run a Glue job inside VPC?

Glue job s3 file not found exception

How does data retrieval work with a DynamoDB Glue Catalog table?

Data Catalog schema table getting modified when I run my Glue ETL job

I need to read S3 data, transform and put into Data Catalog. Should I be using a Crawler?

I am trying to write an ETL job to the Data Catalog but its writing the Headers as Data

How do I get the output of an AWS Glue DataBrew job to be a single CSV file?